European governments have long complained about their dependence on Microsoft’s software, but their rhetoric has not turned into a mass migration away from Windows. During the past few years, Europe’s elected officials have made a lot of noise about ambitious projects to switch to open source software, including big migrations of government PCs in France, Germany, Spain and Norway. Yet the actual migrations have been negligible. More than 95 percent of all PCs used by European government workers still run on Windows, according to the market research firm IDC.
Europe Loves Linux, Runs Windows
260 Comments
Archiesteel capitulates and admits OS and browser usage can be measured!!!!
Don’t be an idiot. The paper states that you cannot count individual users or machines. Yes, you can have an idea of usage, but you can’t derive market share numbers from web logs.
From the same sites:
5.1. Data not captured in the logs
* Individuals’ identities: Except for transactions that have required authorization (passwords), no data recorded in server logs reveal an individual user’s name or any other individual identifier, an e-mail address, for example.
* Number of users: A “user”, as reflected in a log, is an IP address–a computer. This does not necessarily correspond in a one-to-one ratio with an individual person.
[…]
5.3 Unsound inferences from data that is logged
Log files can not support the following inferences, although they are tempting, widespread and, to a greater or lesser degree, encouraged by most of the log analyzer software:
* That hits are equal to use: “Hits” are all exchanges between the client and the server. In order to present an HTML page to a user, the server serves the HTML file, plus all image files embedded on that page (unless the user has turned images on their browser off). This makes “hits” a highly inflated figure.
* That “user sessions” can be isolated and counted: “User sessions” are calculated by some log analyzer products by tracking requests received from an IP address until a period of inactivity (say 30 minutes) indicates to the software that the “session” has ended. As this calculation is based on two unsound assumptions–that a host corresponds to an individual, and that the individual would not normally pause (whether to go to another site or another task) within a site visit–user sessions are, at best, gross estimates.
[…]
4. What you can’t know.
1. You can’t tell the identity of your readers. Unless you explicitly require users to provide a password, you don’t know who connected or what their email addresses are.
2. You can’t tell how many visitors you’ve had. You can guess by looking at the number of distinct hosts that have requested things from you. Indeed this is what many programs mean when they report “visitors”. But this is not always a good estimate for three reasons. First, if users get your pages from a local cache server, you will never know about it. Secondly, sometimes many users appear to connect from the same host: either users from the same company or ISP, or users using the same cache server. Finally, sometimes one user appears to connect from many different hosts. AOL now allocates users a different hostname for every request. So if your home page has 10 graphics on, and an AOL user visits it, most programs will count that as 11 different visitors!
3. You can’t tell how many visits you’ve had. Many programs, under pressure from advertisers’ organisations, define a “visit” (or “session”) as a sequence of requests from the same host until there is a half-hour gap. This is an unsound method for several reasons. First, it assumes that each host corresponds to a separate person and vice versa. This is simply not true in the real world, as discussed in the last paragraph…
Only a complete idiot would see these as support your argument.
Is that really the best you can do? An article from 1998 ?!? Fish in a barrel. Look, your objections to collection of Web statistics are fundamentally bogus.
1. Statistics don’t have to be personally identifiable to be useful and meaningful. Nobody (except a few geeks) changes their browser agent string. So, what you receive for the browser and platform is accurate for 99.999999999% of scenarios.
2. Qualitative information is irrelevant. You don’t need to know why someone visited a site in order to collect statistics from them.
3. Large numbers reduce the margin of error. The more samples in a study, the less that a given subset influences the result. Spiders, bots, and other influences are dwarfed by the sheer volume of traffic from their human counterparts.
4. If you want to reduce the margin of error further, you can draw statistics from a wider variety of sources and filter out addresses that create more than a modest percentage of traffic (ie. spiders, bots).
5. If you want to reduce the margin of error further, you can only draw statistics from sources that require verifiable human input (ie. “enter the number that you see in this scrambled picture”).
6. Influences such as caches (ie. proxies) simply don’t matter as long as the sample size is sufficient.
Is that really the best you can do? An article from 1998 ?!?
The fundamentals of how the web functions haven’t changed since then.
Fish in a barrel. Look, your objections to collection of Web statistics are fundamentally bogus.
No, they’re not. The links I gave specifically states that you cannot tell how many individual visitors you get. No individual visitor count, no reliable market share stats. It’s that simple.
Nobody (except a few geeks) changes their browser agent string.
Except Linux users, which would make them *under-represented* – which is precisely my point.
Large numbers reduce the margin of error.
As long as the sample is random. The OneStat clients are not random, they are self-selected. That alone would be enough to invalidate it as a measure of market share – if web stats could give numbers of individual users in the first place, which they can’t!
Spiders, bots, and other influences are dwarfed by the sheer volume of traffic from their human counterparts.
And you came to this conclusion how, exactly?
4. If you want to reduce the margin of error further, you can draw statistics from a wider variety of sources and filter out addresses that create more than a modest percentage of traffic (ie. spiders, bots).
5. If you want to reduce the margin of error further, you can only draw statistics from sources that require verifiable human input (ie. “enter the number that you see in this scrambled picture”).
There’s no indicating that OneStat does that. Even if they did, that would give them a number of individual users. Without that, no market share stats.
6. Influences such as caches (ie. proxies) simply don’t matter as long as the sample size is sufficient.
Wrong. Firefox uses caching by default, as do many Linux setups (using programs such as Squid). This would mean that they are further under-represented.
So, no, my arguments aren’t bogus. Yours are, however.
Wrong. Firefox uses caching by default, as do many Linux setups (using programs such as Squid). This would mean that they are further under-represented.
Firefox “pre-fetches” by default. Firefox users are always counted 5 or 6 times extra because they pre-fetch the links on the page being hit.
In reality, Firefox usage is skewed by a huge factor.
I suspect real market share for Firefox is 1%.
Prefetching only occurs for pages that are specificly marked for prefetching by the author or system administrator. They would only be counted as extra hits if the links they were prefetching were on different sites also handled by OneStats (I imagine that OneStats *at the very least* do not count multiple simultaneous requests as different sources – if they do, then their stats would be *even less* reliable).
If they prefetch links on other sites, which are not covered by OneStats (or whoever is doing the stats), they wouldn’t be counted.
I suspect real market share for Firefox is 1%.
Whatever. Please continue to underestimate the competition. That only means you’ll sink deeper into your delusion.
I myself don’t give a flying fsck what the market share of Firefox is. The point I was making in the FF thread was that you can’t use web stats to get market share and therefore my numbers were just as valid (or just as useless, take your pick) as yours.
I use FF. It’s better than IE. That is what counts for me.
Prefetching only occurs for pages that are specificly marked for prefetching by the author or system administrator.
Like Google the default search engine for Firefox!
“On some searches, Google automatically instructs your browser to start downloading the top search result before you click on it. If you click on top result, the destination page will load faster than before.
Google uses a special prefetching feature in Firefox and Mozilla web browsers to provide this functionality, so results prefetching is not available in Internet Explorer or other web browsers.”
http://www.google.com/help/features.html#prefetch
They would only be counted as extra hits if the links they were prefetching were on different sites
Like Google!!!
WOW. Firefox is overcounted!!!
Thanks for the info!
you can’t use web stats to get market share
Your reference says otherwise.
They would only be counted as extra hits if the links they were prefetching were on different sites
Like Google!!!
…*and* if the pre-fetched sites are also clients of OneStat, *and* if duplicate IP addresses/user ageny combos are counted separately (which would mean that OneStats’ web stats are even *more* useless in determining market share).
That said, I have already stated I don’t give a rat’s butt about Firefox’s market share. If you want to underestimate the competition and believe that it’s 1%, be my guest. I really don’t care.
Web stats are useless to determine market share.
Your reference says otherwise.
No, they don’t. Even OneStats agree with me, which is why they don’t talk about market share, but rather “usage.”
Stop lying.
Lock-in and proprietary formats is the reality with Microsoft and also Apple.
This is a matter of fact that has been recognized bye the EU and European countries. (years ago)
Some are doing more about it and some less.
Some are faster and more determined than others.
It is not easy and it cannot be fast.
But it would be very wrong to belive it has not started.
And it would be equally wrong to think it will stop.
-
2006-12-10 9:00 pmLarsB
Replying to my own comment I would like to add:
That the problem with lock-in and propritary formats is universal.
And like so often before the USA is leading the way out of that unacceptable state.
The one organisation that knows this better than anyone else is Microsoft.
Microsoft is probably spending more money and effort, to day, on lobbying and politics than on improving their products.
They are to day sitting on the wrong side of the fence and their effort to fool us with an open format is not convicing but also shows they are not unavare of the facts.
-
2006-12-10 9:10 pm
“Ironically, the French Gendarmes will probably still have to pay a U.S. company licensing fees for the Linux distribution it uses on its PCs.”
I’m so sorry Mr Gain, but we here in Europe don’t care about where a given company comes from. It’s about business ethics, quality, and the dangers of monopolistic enterprise. That’s it. It has nothing to do with opposition to the present American government, or “anti-Americanism”, whatever that may be.
Now you go enjoy your Freedom Fries, Mr Gain.
-
2006-12-10 8:43 pmgonzo
Now you go enjoy your Freedom Fries, Mr Gain.
And you can enjoy your freedom Mr. EU, you know, the real freedom you didn’t fight for.
Ironically, the French Gendarmes will probably still have to pay a U.S. company licensing fees for the Linux distribution it uses on its PCs.
Explain please, how is this so ?
It’s the same story across Europe. Switching to open source can cause compatibility issues with Microsoft’s file formats, which are proprietary — and still used by the vast majority of other computer users.
How so ? File>Open…. File>Save …. what problem ?
There can be conflicts with MS Exchange servers, commonly used for e-mail and calendaring.
Ahhh, the old FUD. Dont switch to Linux or your email will not work because everyone else uses MSExchange servers.
Can we superglue this guys fingers together so he cannot type any more crap like this ?
-
2006-12-10 9:13 pmtwenex
Ironically, the French Gendarmes will probably still have to pay a U.S. company licensing fees for the Linux distribution it uses on its PCs.
Explain please, how is this so ?
Sounds like they are moving to Novell. I *think* the move was announced before Novell’s MS deal; if so they may well be re-evaluating their choice of distro. If they decided to go with Novell after the announcement, or they aren’t reevaluating, I’m glad I’m not a French citizen. Then again, I haven’t heard HMG talking of moving to FOSS at all.
I was amazed after reading this news because I thought that most of the Europeans use Linux but I was wrong.
They use Windows and ask others to switch to Linux.
Hmmmmm…looks amazing!
I proudly consider to be a strong neoconservative (ohmigosh!), pro American, pro capitalist…
and I am very very much in favor of open source software. The biggest difference between me and other F/OSS supporters is that I am totally opposed to govts trying to break down microsoft with anti-trust rulings. I think open source projects can win on their own merits and not because some politician says so. Microsoft’s monopoly really can be broken down step by step if everyone else could get their act together.
-
2006-12-10 9:32 pmtwenex
I am totally opposed to govts trying to break down microsoft with anti-trust rulings.
Anti-trust laws are there to punish illegal monopolies in general, not to punish Microsoft in particular – they’ve been active long before MS existed and even went after other companies during the lifetime of MS.
The reasons why MS have been involved in such huge and protracted anti-trust cases are:
(a) The computer industry seems to be particularly heavily infested by people who think they have a God-given right to be different from everyone else. Imagine if every coffee maker restricted you to its own coffee machines. If you needed not “a driving licence” but “a Ford driving licence” or a “Toyota driving licence”…
(b) Governments were very lax in getting to Microsoft once it was clear they were using monopolistic tactics.
(c) MS have been very good at hiding their tactics with NDA’s, etc.
(d) They are doing their utmost to wriggle out of the cases and any fallout from being convicted.
Edited 2006-12-10 21:48
-
2006-12-10 9:56 pmarielb
make something better than Microsoft and it will no longer be a monopoly. People will switch like they are doing with Firefox. The problem is that linux and its apps aren’t that much better than Windows. That has to change but there’s point in crying about it or pretending the problems aren’t there.
-
2006-12-10 10:06 pmtwenex
make something better than Microsoft and it will no longer be a monopoly. People will switch like they are doing with Firefox. The problem is that linux and its apps aren’t that much better than Windows. That has to change but there’s point in crying about it or pretending the problems aren’t there.
I’m not “pretending problems aren’t there”, but the problems are very often not with Linux. The developers, for example, can and already do code device drivers for a vast amount of devices – and happily. The ones they can’t are the ones for which the manufacturer refuses to release code and/or specs. It’s the same with manufacturers not releasing apps for Linux.
As far as I’m concerned Linux *is* better than Windows. The gap has narrowed, but the fact that it has *narrowed* rather than widened shows that vendor lock-in is the problem.
EDIT: that wasn’t clear: if vendor lock in was no problem, people would have switched when Linux was miles better,
Edited 2006-12-10 22:25
Let the administration move, french administration, for instance, is slow, extremely slow, and so much paperwork need to be filled to justify its slowness !
It’s on the way, wait another 10 years and you’ll see !
it’s one thing to get XP over windows 98. Win98 had hardware support but had so many problems, crashed a lot and it wasn’t 100% 32 bit. I don’t think most people got XP because of the fisher price effects.
I don’t see anything in Vista that I need. But what I do need is protection from being locked in and made obsolete. I think moving away from that may have a cost that’s worthwhile.
-
2006-12-11 4:24 amgonzo
bold is so frickin irritating, would you please stop? thank you!
I do it actually to make it easier for you, so that you can easily see what is it I am answering to.. How about italics?
-
2006-12-11 5:32 amdeanlinkous
Thank you! Yes italics are great. I tried bold a time or two and it is much easier to read but it is distracting seeing it over and over. I guess I get distracted too easy but man it looked like a checkerboard to me or similar… Thanks again!
Everyone seems to think that going from linux to windows is like a snap. I work for government and It took me about 2-3 years to get everything on linux (software and training) and we are very small part of it (about 100 computers and few servers).
Noting like debating three persons at one and keeping the upper hand…
Still, in 1:30AM and I actually work for a living, so I am now going to bed. I’ll let other brave souls dispel the disinformation of the three anti-Linux Trolls…
-
2006-12-12 6:35 amNotParker
Noting like debating three persons at one and keeping the upper hand…
Still having math problems?
How about more corroboration?
http://www.webhits.de/webhits/browser.htm
-
2006-12-12 3:05 pmarchiesteel
How about more corroboration?
It’s not “corroboration” when all you do is provide more web stats. Let me repeat it again: web stats cannot be used to determine market share. I’ve already given the reasons why, the fact that you have not tried to challenge them shows that you’re out of arguments.
You lose.
-
2006-12-12 5:06 pmNotParker
It’s not “corroboration” when all you do is provide more web stats.
Yes it is.
By the way, do have any usage numbers from 2006 yet?
Let me repeat it again: web stats cannot be used to determine market share.
As your quote says, “small sample size” is problematic.
The sample size on my references (I’m still waiting for yours) is HUGE.
Therefore webstats work just fine. Otherwise people would have to guess at Firefox’s market share (for example). I’d guess 1% tops. Large web analytics firms disagree.
Do you have an non-webstats numbers suggesting Firefox’s market share is above 1%?
-
2006-12-12 5:34 pmarchiesteel
Yes it is.
No it’s not.
As your quote says, “small sample size” is problematic.
No, “sample bias” is problematic. In this case, it’s not a random sample, so it is useless from a statistical point of view.
It’s also only one of several issues with web stats. I’ve enumerated them. You’ve ignored them. Go figure.
-
2006-12-12 6:03 pmNotParker
In this case, it’s not a random sample, so it is useless from a statistical point of view.
75,000 websites. Anyone can go to them. As random a sample as you can get.
-
2006-12-12 7:27 pmarchiesteel
A self-selected sample can never be considered random.
Your ignorance of basic statistical science is obvious – that’s kind of a problem when your whole argumentation is based on misrepresenting statistics!!
-
2006-12-12 9:35 pmNotParker
A self-selected sample can never be considered random.
It is statistically significant if it involves 75,000 collection points day after day after week after week and the numbers all match.
-
2006-12-12 9:48 pmarchiesteel
It is statistically significant if it involves 75,000 collection points day after day after week after week and the numbers all match.
Wrong. We cannot deduct that a sample is biased through consistency.
Let me give you an example. Let’s say you want to have a good idea of the results in the upcoming election and so you conduct a poll. However, instead of using a true random sample method, you decide to go through your business contacts from the State Association of Chambers of Commerce. Let’s say that the Association has a membership of 10,000. You send forms to each member and ask them to fill in the poll questions. You get back the forms and find out that Republicans are getting 67% of the vote. The high number of people responding means your margin of error is very small.
However, you forgot to take into account that businessmen big and small, who make up the bulk of chambers of commerces throughout the country, generally tend to lean more towards the Republicans than the democrats. Therefore, your results are skewed and do not pain a reliable portrait of the overall political picture.
It’s the same here. OneStats clients are *not* random, they are self-selected, and self-selected samples will *not* provide an accurate picture.
That’s besides the point, though, because web stats cannot provide the number of individual users anyway, and so they are useless to provide market share. Of course, OneStat is not going to admit this outright, but they do talk of “usage” instead of market share, which is telling.
-
2006-12-12 10:01 pmNotParker
You send forms to each member
Theres the problem. You don’t understand weblogs. No one is sending anyone a form.
Think of it this way. A person stands on a street corner and counts the type of car that goes by. Day after day after day.
The bias would be geographic location. If the corner was in the US, then more american cars would show up. Etc etc.
None of those people in the car were sent a form. They just drove by.
Now, what happens when someone stands on a 75,000 street corners in 100 countries and record the make and model of each car.
Thats a HUGE sample. No forms. People going about their day to day business.
And then you take a random sample of 20,000 of the recorded make and model, each from 100 countries.
Day after day after day.
Thats what onestat does.
The ONLY bias you can say for sure is that it discriminates against people who don’t own or use a computer.
OS and browser figures on Onestat, corroborated on 10’s of thousands of other “street corners”, show Linux with about .36% of the market.
Live with it.
-
2006-12-12 10:11 pmarchiesteel
Theres the problem. You don’t understand weblogs. No one is sending anyone a form.
Man you are dense. It was an *analogy*.
I do understand weblogs. I’ve administered them. I’ve analyzed them. I’ve used software similar to the one OneStat uses.
Now, what happens when someone stands on a 75,000 street corners in 100 countries and record the make and model of each car.
Yeah, except the 75,000 street corners are not selected at random. Therein lies the rub.
The ONLY bias you can say for sure is that it discriminates against people who don’t own or use a computer.
No, it discriminates against people who don’t visit OneStat’s customers’ web sites.
If it was 75,000 random web sites, then that would mean a better sample (although it *still* wouldn’t be an accurate measure of market share, because web stats cannot give the number of individual users visiting the site). But it’s not 75,000 random sites, it’s 75,000 user-selected sites.
OS and browser figures on Onestat, corroborated on 10’s of thousands of other “street corners”, show Linux with about .36% of the market.
Web stats cannot be used to determine market share, regardless of sample size, because web stats cannot give an individual user count. *Anyone* who’s ever analyzed a web log knows that, and the two research papers I linked to corroborate this information.
Web stats cannot be used to determine market share. Live with it.
-
2006-12-12 11:29 pmNotParker
It was an *analogy*.
A very, very bad analogy.
Web stats cannot be used to determine market share, regardless of sample size,
Wrong. As you yourself said (via your references) sample size is the most important issue.
it’s 75,000 user-selected sites
Kind of like measuring sales of a product by counting how many were “user-selected” and carried out the door.
The users selected the website to go to, but the item being measured is which OS they are using.
.36% and, most likely, much much lower because of Firefox’s prefetching on Google and other websites that skew the data in favor of Firefox users.
-
2006-12-13 2:54 amarchiesteel
Wrong. As you yourself said (via your references) sample size is the most important issue.
I have not said it was the most important issue, and neither have my references. It is *one* issue among many. The main one (and the one you have consciously avoided adressing in all your broken-record replies) is that web stats *cannot* determine the number of individual visitors to a web site.
Kind of like measuring sales of a product by counting how many were “user-selected” and carried out the door.
Sure, but only if you forget the face of the person coming out and count them again when they return.
Also, depending on the web sites, it could be like looking at what kind of lightbulb the person has bought when they’re coming out of a baker’s store.
The users selected the website to go to, but the item being measured is which OS they are using.
Exactly, and since the users selected the website, the results are skewed by the readership of the website. So if you have very few websites that appeal to Linux users, Linux users would be under-represented.
I see you’re finally starting to understand why web stats cannot be used to determine market share.
.36% and, most likely, much much lower because of Firefox’s prefetching on Google and other websites that skew the data in favor of Firefox users.
Uh, no. Caching more than compensates for pre-fetching, because once it is cached the pre-fetching is not activated.
The point is moot anyway because web stats cannot identify the number of individual users. No head count, no market share statistics. The fact that you have refused to even acknowledge this issue makes it clear that you are conceding defeat on this most crucial of points.
Stop grasping at straws, you’re starting to look desperate.
-
2006-12-13 4:56 amNotParker
I have not said it was the most important issue, and neither have my references. It is *one* issue among many.
I read your references. Definitely sample size was THE most import issue. Once thats taken care of, web stats are A-OK according to your references.
Exactly, and since the users selected the website, the results are skewed by the readership of the website
But the HUGE size of the sampe (75,000 sites day after day after day) and the random selection of 20,000 hits from each of the countries takes care of the randomness.
Sample size is KING!!!
So if you have very few websites that appeal to Linux users
Theres the problem You can’t find any. Even Distrowatch is majority Windows users.
No wonder you don’t get it. You are looking for something that doesn’t occur in real life … just your imagination.
The point is moot anyway because web stats cannot identify the number of individual users.
I’m not interested in who they are as individuals. I’m interested in the browser and OS they are using.
And it isn’t linux.
-
2006-12-13 5:52 amarchiesteel
I read your references. Definitely sample size was THE most import issue. Once thats taken care of, web stats are A-OK according to your references.
No, they’re not. They indicate a very definite set of stats that can be derived. Individual visitors isn’t one of them. Sample size is only one of the factors making the stats even less accurate.
But the HUGE size of the sampe (75,000 sites day after day after day) and the random selection of 20,000 hits from each of the countries takes care of the randomness.
It’s not a random sample, it’s self-selected. The size of the sample doesn’t matter. In fact, depending on the sample bias, the sample size can even increase the discrepancy.
Theres the problem You can’t find any.
Oh, I think there are plenty. I think this very site here is one of them. Aren’t you always whining that it’s got too many “FOSS cultists” and that they always gang up on you to mod you down?
Even Distrowatch is majority Windows users.
That’s completely irrelevant to the matter at hand. But just so you know, if there were more sites like Distrowatch among OneStat’s customers, the result for Linux would be much higher. The fact that you can’t understand this and somehow think that this supports your position really shows how little you know about statistics.
I’m not interested in who they are as individuals. I’m interested in the browser and OS they are using.
Are you deliberately trying to misinterpret my words? Because it really sounds like you are in an effort to get out of the corner you painted yourself into.
I don’t care to know who they are as individuals, NotParker. That is not the point at all. It is irrelevant. What *is* relevant is to know the number of individual visitors to the web sites. And you can’t tell that from web logs. If you had any experience with web logs you’d know this.
Each time you post your truth-twisting post, you give me the occasion to prove you wrong in front of whatever readership is left for this thread (which is not much – it’s probably just you and me). I know you’re lying, and you know you’re lying. Give it up.
-
2006-12-13 6:03 amstestagg
I’m still with you. Mainly because I follow all of NotRedmond’s rants for the same reason that I visit Shelly the Republican’s blog, (masochism I guess)
Another factor to include in this is Proxies. With the prevalence of SQUID, many (some?) companies/groups using Linux are behind a caching proxy. Yes NotParker, I realize that windows users might be proxied as well, but probably fewer than us linux geeks.
-
2006-12-13 6:11 amarchiesteel
Yes, I brought up caching, which is much more prevalent in the Linux world than for Windows (mostly because of Squid). As you can imagine, he hasn’t responded to that particular argument…
I think it’s clear that NotParker isn’t interested in rational debate, but rather in repeating his talking points over and over again.
Honestly, I’m hoping that he’s an astroturfer, because if he’s just a fanatic then it’s a bit scary.
-
2006-12-13 8:14 amNotParker
I brought up caching, which is much more prevalent in the Linux world than for Windows (mostly because of Squid).
Since WIndows is on 95% of PC’s, and Linux is on .36% of the worlds PC’s I would guess it was more prevalent in the Windows world.
Do you have any numbers from 2006 to back you up?
Estimates from January 2005 don’t count.
-
2006-12-13 10:24 am
-
2006-12-13 3:04 pmarchiesteel
Since WIndows is on 95% of PC’s, and Linux is on .36% of the worlds PC’s I would guess it was more prevalent in the Windows world.
Prevalent in this context is relative, not absolute, so it doesn’t matter if Windows is on 90% of the worlds’ desktop and Linux on 3 to 5%.
Do you have any numbers from 2006 to back you up?
No, and neither do you.
Estimates from January 2005 don’t count.
Web stats never count. So I guess we’ll have to take IDC’s word on it. That shouldn’t be a problem, since you’ve already accepted them as a reliable source.
-
2006-12-14 12:14 amNotParker
[i]So I guess we’ll have to take IDC’s word on it.[i]
From when and where?
January 2005 estimates don’t count.
.36% until proven otherwise.
-
2006-12-14 12:39 amarchiesteel
You have problems with your italics there, NotParker?
January 2005 estimates don’t count.
Sure they do. They’re from IDC, a company which you respect. And they are actual market share numbers, not web stats.
3 to 5% until proven otherwise, since that’s the only market share numbers we have.
-
2006-12-13 7:55 amNotParker
That’s completely irrelevant to the matter at hand. But just so you know, if there were more sites like Distrowatch among OneStat’s customers, the result for Linux would be much higher.
Tell me about their customers.
You seem be sure they are sites only Windows users would visit.
Tell me how you know that is true for OneStat, NetApplications and TheCounter.
Maybe while you are looking for that, you could find me some 2006 numbers that refute the webstats.
-
2006-12-13 3:11 pmarchiesteel
You seem be sure they are sites only Windows users would visit.
I’m not, but since we can’t know either way we can’t take the data at face value, like you’re doing.
Maybe while you are looking for that, you could find me some 2006 numbers that refute the webstats.
I don’t need to, because web stats are useless when it comes to determining market share.
Let me repeat it again: web stats are useless when it comes to determining market share.
One more time: web stats are useless when it comes to determining market share.
-
2006-12-13 8:03 amNotParker
I read your references. Definitely sample size was THE most import issue. Once thats taken care of, web stats are A-OK according to your references.
But the HUGE size of the sampe (75,000 sites day after day after day) and the random selection of 20,000 hits from each of the countries takes care of the randomness.
Sample size is KING!!!
Thanks for the references by the way. They demolish your case.
-
2006-12-13 3:07 pmarchiesteel
I read your references. Definitely sample size was THE most import issue. Once thats taken care of, web stats are A-OK according to your references.
No, they’re not. Sample size is only one of the many bias. But even when all the bias are in, the stats still can’t be used for market share because you can’t get an individual head count of users.
Sample size is KING!!!
Repeating this false statement like a broken record will not make it true. A large sample that is self-selected is not a reliable source for statistics.
Thanks for the references by the way. They demolish your case.
No, they don’t. They list a *series* of problems with Web Stats, and clearly state that you cannot get the number of different users visiting a web site. Without this *your* entire argument is reduced to rubbly.
Stop trying to argue that 2+2=5, NotParker. No one believes you and it only makes you look foolish.
Has it ever occurred to you that happy Linux users and happy Windows users will not bother debating silly numbers?
If you like typing, I just read that Amnesty International is looking for volunteers to write letters to/on behalf of political prisoners.
Edited 2006-12-12 09:07
The amount of Linux-hating drivel posted by the Windows wankers in the comments for this item since I last posted just shows how desperately they must feel like time is not on their side.
I haven’t seen a single pro-Windows posting in said timeframe whose “argument” was not on the lines of “these are the real facts [which I’ve invented] you smelly parents’-basement-inhabiting communist Linux user”.
It’s really quite pathetic. I can understand why people leave the Mac community if they have to sift through this kind of drivel; if the only really dedicated Windows users are mental midgets^H^H^H^H^H^H^Hmicrobes like this lot it’s hard to believe demand hasn’t forced ISV’s to port more Adobe Photoshop type stuff away from Windows.
Yes I’m biased, but I believe even an unbiased observer would conclude from this discussion that Linux users know something whereof they speak, and Windows users are all insecure little twerps with small d*cks.
Thank God I know different.
Edited 2006-12-12 15:03
..are all organizations funded by governments. It is so easy to engage in those “let’s move to Linux” adventures when you’re spending someone else’s (taxpayers’) money.
How come we don’t see big European companies moving to Linux?
Now why does moving away from windows actomaticly mean a switch to linux?
Because OSX is not an option? It’s just as proprietary as Windows.
BSD could be an option, but why not just use Linux then which has the most momentum of the two?
I never said they should switch to OSX… So i don’t know where you get that from. So insted taking a serious look at whats out there you should just go for the hype?
I never said they should switch to OSX… So i don’t know where you get that from. So insted taking a serious look at whats out there you should just go for the hype?
LOL You’re telling me to be serious?
So, enlighten me, what else is out there, that can be successfully deployed on desktop, as Windows replacement? OK, so it’s not Linux, it’s not MacOS X.. OK, what else? But, please, take your own advice and be serious before you spit it out.
And why aren’t those European governments moving to those other systems? Because they’re not serious?
Nah it was suppposed to be “they”… Switch to linux because the fanboys says so. No need to try it out.
Your sad attempt to talk down doesnt show much else then fanboy behavier.. but thats the way you want people to view you don’t let me stop you.
Whether he’s a fanboy or not, *I’m* interested in hearing an answer since it’s a valid question.
It wasn’t so much a “sad attempt” at talking you down as a valid point, albiet slightly attacking, which you appear to be getting overly defensive about.
I wont get into a pissing contest here about what OS is the best.. Unlike others it seems thats not what i’m getting a kick off. Hell let them learn there own lesson by using linux see if i care. I’m all for using what suits you best.
I’ve been working with linux and i just don’t see it what all the hype is about. The only good thing i can think of about linux must be the installer from ubuntu. Just a shame it locked my computer insted of just work.
Personally I think Linux does a great job in some areas but I’m also all for using a multi-platform environment while steering away from using solutions that are not compatable with competitors (the DOC format for example). That’s just because IMHO vendor lockin can cause problems that could otherwise be avoided and not for some philosophical reason.
Dismissing an OS because you fail to see its strong points is not a good or sensible idea since if you ever have to build a large system you’ll have weak points that you just don’t need to have due to poor research or bias. Now, I like Linux but it’s all about using the best tool for the job which in many areas Linux is not. The same goes for Windows, OSX, *BSD, or whatever other solution you happen like.
Back to the original question, if you’re talking about moving away from Windows but not to one of Macs offerings or Linux which solution were you thinking of. Please understand, I’m not being aggressive I’m actually interested in your view.
“Personally I think Linux does a great job in some areas but I’m also all for using a multi-platform environment while steering away from using solutions that are not compatable with competitors (the DOC format for example). That’s just because IMHO vendor lockin can cause problems that could otherwise be avoided and not for some philosophical reason.”
I’ve never said i wasnt for compatable solutions? Just because i’ve found linux doesnt work for me doesnt mean i don’t support open standards.
“Dismissing an OS because you fail to see its strong points is not a good or sensible idea since if you ever have to build a large system you’ll have weak points that you just don’t need to have due to poor research or bias. Now, I like Linux but it’s all about using the best tool for the job which in many areas Linux is not. The same goes for Windows, OSX, *BSD, or whatever other solution you happen like.”
I have yet to see any strong points in linux that i cant find in other OS’es.
“Back to the original question, if you’re talking about moving away from Windows but not to one of Macs offerings or Linux which solution were you thinking of. Please understand, I’m not being aggressive I’m actually interested in your view.”
You gotta read what i write… I asked why people just asumed a move away from windows automaticly would be a move to linux? It would be easy to diss miss it as fan boy talk from people who think they a the coolest people in the world because they run linux at home on a server.
I never did say they couldn’t move to OSX… Thats just another thing people assume pretty much the same as linux is the solution to all problems.
LOL You’re telling me to be serious?
So, enlighten me, what else is out there, that can be successfully deployed on desktop, as Windows replacement?
I’m sorry. I don’t believe you’ve enlightened us all as to what advantages Windows has in these kinds of government environments, and why Linux is just such a poor fit.
I’m sorry. I don’t believe you’ve enlightened us all as to what advantages Windows has in these kinds of government environments, and why Linux is just such a poor fit.
Migration to Linux is too expensive, takes too long.
Windows is already there, custom applications are already created, people are already trained (both end users and technicians).
Just take a look at Munich migration. It started, what, 3-4 years ago, expenses are now around $35M (initially $30M), and it is not finished yet. Not even close.
Name one company in the world that can afford that kind of migration.
Migration to Linux is too expensive, takes too long.
Why?
Windows is already there, custom applications are already created
It’s a stmubling block, as anything already in place already is, but that’s a woefully poor reason for maintaining the status quo because change is always required in any environment. You’re looking at what you’re going to save over a period of something like five to ten years rather than what you have today.
Just take a look at Munich migration. It started, what, 3-4 years ago, expenses are now around $35M (initially $30M), and it is not finished yet.
And you think rollouts to Windows XP are actually finished in many environments?
Besides, as I pointed out above, organisations are looking at what they can get out of this over a period of something like five to ten years. I can give you a list if you like:
1. Real remote working from anywhere without the woeful administration overhead of Microsoft Office licenses and Terminal Services licensing (had my fill of that personally) and the cost both in licenses an administration involved.
2. Unix and Linux do real centralised management. You mount a directory as if it’s local and apply permissions there. All the applications anyone needs are there and you don’t need to distribute apps through MSI or over Zenworks and with Active Directory. When a knew app goes live you just copy it there. This costs a huge amount of time and money to do in Windows environments.
Microsoft purposefully tries to keep that dividing line between clients and servers, and as long as they do that administration in these kinds of Windows environments is going to continue to be absolutely astronomical.
So, enlighten me, what else is out there, that can be successfully deployed on desktop, as Windows replacement? OK, so it’s not Linux, it’s not MacOS X.. OK, what else? But, please, take your own advice and be serious before you spit it out.
Sun/sunray thin clients. Excellent stuff, great management, tadbit expensive (just as rolling out a big deployment of linux is). It’s tailored for government and big industry, has good support and best of all: it’s not linux 😉
For most government purposes OSX is not viable. Not because it doesn’t work, not because it isn’t pretty, which it is, but mostly because it isn’t a thin client solution. Where I work (in the netherlands), I’ve seen a big movement away from standalone workstations. The organizations where I have been stationed are mostly turning away to either a complete thin client solution (with windowsCE thin clients) or a partial thin client solution (Citrix/Win2K3 terminal services with some programs subscribed in Powerfuse). It’s the management, stupid.
The organizations where I have been stationed are mostly turning away to either a complete thin client solution (with windowsCE thin clients) or a partial thin client solution (Citrix/Win2K3 terminal services with some programs subscribed in Powerfuse). It’s the management, stupid.
So it is Microsoft again.. both on servers and clients?
As far as I have seen, it still remains Microsoft on both clients and servers. Some are retaining a Netware solution, but many companies are phasing it out. I think that it is a shame, the helpdesking gig I did at a (partial) Netware shop was quite enjoyable.
Remote management services were nice, and I happen to like Groupwise for collaboration purposes.
Thin clients are not the magical solution to everything. They are useful in certain situations. You can’t automatically recommend thin clients (or any other solution, for that matter) without knowing the particular circumstances for an organization or company. Try offering thin clients to a design company or an architecture firm and you’ll be laughed out of the building.
Many have tried recommending people to revert to thin clients over and over again. I remember Sun preaching the same ten years ago. It’s just marketing, trying to sell a particular product. The actual demand has relegated thin clients to a niche. It’s good for those who need it, but not for everybody.
These being said, here’s a heads up: Linux can make for excellent thin client setups just as well.
Now why does moving away from windows actomaticly mean a switch to linux?
What else? MacOS X?
So, you’d replace Windows with Mac and therefore replace Microsoft with Apple? How’s that better? Not to mention that hardware-wise you’d be dependent on one company – Apple..
I was kind of thinking the same, it’s so easy to go and buy all those Windows licenses when you’re spending someone elses money.
I was kind of thinking the same, it’s so easy to go and buy all those Windows licenses when you’re spending someone elses money.
No, my company does not spend taxpayers’ money. We spend what we make. And if we spend too much, we’ll all lose our jobs. Big difference.
Of course, but we’re talking about Governments who are spending taxpayers money and as such aren’t concerned about the proven lower TCO of Linux. I’m not sure where you lost me on this one.
No, as I said governments are not good example because they spend someone else’s money – it really doesn’t matter if it’s Windows or Linux, etc. They can make political decision to do it, no matter what the costs are.
That is why I asked how come there are no European companies on that list.
Not sure where you lost me on that one.
Thousands of European companies are currently using Linux for all kinds of workloads and functions. The article did not name them because the article was about use of Linux in the public sector and because the article is written with the intent of planting fear, uncertainty and doubt into those that are in the process of starting their own migrations to Linux.
as I said governments are not good example because they spend someone else’s money
No company has a God-given right to money. They provide service, they get money for it – no different to a government. They are in a far worse position than governments in this sense, because however bad governments are, they do have an obligation to (for example) defend their citizens.
Any company that *does* think it has a God-given right to other people’s money is exactly the kind of company anyone with nous stays away from if at all possible.
It’s because it’s from a political need (ideology and emotionaly driven) rather than a real need (needing to fulfill a specific task).
I’ve seen people ditch AS/400s just because the admin didn’t felt good about them.
Yes, that’s right, everyone who uses Linux in preference to Windows is being “emotional and ideological”.
/Rolls eyes.
-5.
Learn to read.
-50 DKP.
I doubt I need lectures on reading from someone who thinks Linux users (let alone “any members of the human race, besides Microsofties – actually, delete that last bit) – actually want to use Vista:
http://osnews.com/permalink.php?news_id=16553&comment_id=184978
In point of fact, this is more telling:
http://theinquirer.net/default.aspx?article=36259
I bet much of the (lack of) enthusiasm is due to the fact that Vista costs 20% more to build:
http://theinquirer.net/default.aspx?article=36251
Compete with free (as in freedom and price) by slapping a higher price on draconian software. That’s so stupid it’s almost clever.
Edited 2006-12-11 01:22
I don’t think that Linux users needs Vista. I know OSS fundies yearn Windows Vista.
Learn to read.
🙂
I know OSS fundies yearn Windows Vista.
Yeah, I long to waste FIFTEEN GIGABYTES on something that does the same as something I’ve already got, but more expensively and less well.
Then those organisations deserve to fail to listening to those technicaladmins without conducting proper business analysis.
How come we don’t see big European companies moving to Linux?
I am a developer, and don’t know much about office & productivity application. Far as I know Microsoft productivity applications are ahead of competition.
On the other hand, to be honest, majority of government employees are not working that much, and the productivity is not of the essence. That’s why less expensive alternatives are practical.
Another matter is that government institution should not depend too much on particular vendor. I am not talking about conspiracy theories. Imagine that the vendor gets hit by terrorist attack and goes out of business.
DG
…migrating computers takes some time. If you want to use a new technique for building skyscrapers, you test it thoroughly, train your personnel, before you implement it. The same is true with software. You first try all possible scenarios, then train the personnel and then make the switch. Not the other way round.
It will be interesting to see how far the transition progressed in e.g. five years.
First get them to Firefox over IE.
Then get them to switch media formats.
Then openoffice over MS office.
Then and only then will people switch to another OS. It has to be done incrementally.
First get them to Firefox over IE.
Then get them to switch media formats.
Then openoffice over MS office.
Then and only then will people switch to another OS. It has to be done incrementally.
Yeah right.. you know how much time and effort is involved in that? Too much, otherwise we’d all be using Linux by now.
C’mon, if Linux was as good as Windows (I’m talking about desktop usage), who would want to pay for Windows? (And no, we’re not sheep and you’re not smarter than everyone else. The whole “Windows desktop” environment is simply better choice right now.)
But I see something else there: the scenario you described can be used to get better deal from Microsoft. And that is exactly what most of the companies are going to do, for the time being. 10 years from now? I don’t know, I guess we’ll see.
Edited 2006-12-10 19:25
C’mon, if Linux was as good as Windows (I’m talking about desktop usage), who would want to pay for Windows?
Governments that have documents locked in in file formats that only can be read by applicatiosn that only runs on windows. It have nothing to do with windows allegedly being better than Linux.
Governments that need to be independent of companies in foreign companies for doing mission critical things like national defence.
Governments that want to give the citizens in their county equal access to information, regardless of financial capacity of paying expensive software licenses.
Governments that prefer stimulating the local economy rather than sending money to Redmond.
you don’t have to change your entire OS and throw out all your other software to avoid locked in file formats.
Governments that have documents locked in in file formats that only can be read by applicatiosn that only runs on windows.
Not true anymore. How’s using Windows stopping me from using, for example, OpenOffice?
Governments that need to be independent of companies in foreign companies for doing mission critical things like national defence.
Ah yeah.. another conspiracy theory.
Why are they all using USA technology such as jetfighters: F-15, F-16, etc?
Governments that want to give the citizens in their county equal access to information, regardless of financial capacity of paying expensive software licenses.
How’s Windows stopping them from doing that?
Governments that prefer stimulating the local economy rather than sending money to Redmond.
Is that why we all wear shoes and T-shirts made in China (with the exception of Italian guys, sure :-)? Did you ever buy anything from non-European company (Sony, Apple, Toyota..)?
Edited 2006-12-10 20:02
Not true anymore. How’s using Windows stopping me from using, for example, OpenOffice?
You didn’t get it, and I won’t explain it to you, would be waste of 20 seconds, and you wouldn’t get it anyhow.
Ah yeah.. another conspiracy theory.
Why are they all using USA technology such as jetfighters: F-15, F-16, etc?
Because they are stupid? Actually my country is using Migs and Sus. They are cheaper, and better. The only difference might be with parts, not sure how much industry is left in Russia for this.
How’s Windows stopping them from doing that?
By costing too much in the first place.
Is that why we all wear shoes and T-shirts made in China (with the exception of Italian guys, sure :-)? Did you ever buy anything from non-European company (Sony, Apple, Toyota..)?
What was this about? When americans shout “evil chinks ruining our economy” it’s OK, but when some european says that stimulating local economy is good he’s a xenophobe?
Because they are stupid? Actually my country is using Migs and Sus. They are cheaper, and better.
And your country is listed in that article?
The only difference might be with parts, not sure how much industry is left in Russia for this.
Oooh, so what happens when you have no spare parts? You fly like pigs can fly?
Most of the European countries already are or soon will be in NATO. That also means bye bye to MiGs and Sus.
When americans shout “evil chinks ruining our economy” it’s OK, but when some european says that stimulating local economy is good he’s a xenophobe?
There’s a difference between shouting about something and doing it.
“Most of the European countries already are or soon will be in NATO. That also means bye bye to MiGs and Sus.”
For your information: dropping the Russian techniques does not necesseraly mean adopting US ones. For fighters both Hungary and Czechia chose the Swedish Saab’s Gripens from the NATO countries; also Germany, France and the UK runs several concurrent air force projects.
Sorry, I could not decide whether or not it was off-topic, maybe the theories behind are the same.
You didn’t get it, and I won’t explain it to you, would be waste of 20 seconds, and you wouldn’t get it anyhow.
He had you cold: It ISN’T necesary to change OSes to use open file formats. The fact that you don’t want to “explain it” has more to do with the fact that you CAN’T explain it.
Because they are stupid?
No argument here. /sarcasm
By costing too much in the first place.
Again, the cost of Windows has nothing to do with the choice of file formats. You completely dodged that issue.
When americans shout “evil chinks ruining our economy” it’s OK, but when some european says that stimulating local economy is good he’s a xenophobe
There’s a big difference between objecting to an overwhelming trade gap ($30B per year) and trying to stimulate a local economy by purchasing a locally-created product. The former really IS ruining an economy. The latter isn’t.
Edited 2006-12-12 00:04
Why are they all using USA technology such as jetfighters: F-15, F-16, etc?
They wouldn’t be if those fighters or the manufacturing company’s policies required McDonnell Douglas pilots and Texaco fuel. That is the best analogy to the situation we have in the proprietary software industry. Remember, even the UK government, presently headed by the man pretty much everyone describes as “Bush’s lapdog”, roundly told the US that it could either hand over top secret information on the workings of some US technology we were buying or the American govt. could stick the deal where the sun don’t shine.
They wouldn’t be if those fighters or the manufacturing company’s policies required McDonnell Douglas pilots and Texaco fuel. That is the best analogy to the situation we have in the proprietary software industry.
You need Microsoft technician to install and use Windows and you’re forced to use MS Office on Windows? And you can’t use PDF, etc, etc? How’s that?
And any governement can have access to Windows internals, I am sure you know that.
So, what is your point?
You need Microsoft technician to install and use Windows
You need technicians trained in the technology of one single company, which is Microsoft, in a corporate setting, yes.
and you’re forced to use MS Office on Windows? You are if you need 100% compatibility with all Office applications. Until quite recently you were forced to if you needed any compatibility at all.
And you can’t use PDF, etc, etc? How’s that?
Until recently Office couldn’t save to PDF. In fact I may be wrong in thinking that it can now – and PDF creators are more expensive than Office.
You need technicians trained in the technology of one single company, which is Microsoft, in a corporate setting, yes.
So what? You need technicians either way.
Until recently.. Until recently..
Can you please fast forward to “today”?
So what? You need technicians either way.
With Linux, you don’t need technicians who are tied to the oddball products of one company. It seems like you keep trying to avoid this point.
People say they like an end-to-end Microsoft solution because they have “one throat to choke”. The problem is there aren’t that many companies in the world with handspans large enough to choke MS’s throat.
Can you please fast forward to “today”?
Today, MS is still a proprietary company and there is no indication that they are going to change tack in the near future – if they were then they would have worked to improve ODF, not create yet another “open” format which they may have patents on, now or in the future, and which they reserve the right to alter unilaterally.
I beg to differ sir…PDFCreator is free.
http://sourceforge.net/projects/pdfcreator/
Actually to get it loggically correct you have to switch things upside down.
The fuel is synonym of Windows and doc files are synonym of the fighters.
Not true anymore.
And just why is the file format lock-in cycle not true anymore?
If you’re going to say what I think you’re going to say then I would advise you to go away and read some other articles around here. Put simply – you don’t know what you’re talking about, or if you do, then you’re just talking crap for reasons only you know.
How’s Windows stopping them from doing that?
Go away and do some reading before making a fool of yourself here.
that’s my point. it’s a lot easier to switch to firefox than linux. And no I don’t think linux has what it takes. The open source community has to stop with this linux focus and work on apps that are also on Windows. For example, gnucash, koffice, konqueror, etc. don’t work well on windows so Windows users will just use Quicken, MS office and IE and they will never ever switch to any other OS.
I don’t agree that Linux devs need to “switch focus”…I think it’s good that good open-source apps are available for Windows, but work still needs to continue on Linux in parrallel.
Look, guys, this isn’t going to happen overnight. It’s often a chicken/egg situation, except that in this case governments have decided to step in and get things started. Extremadura is a good example that, yes it can be difficult, but it’s worth it, as they saved more than 30 million Euros in licensing fees, and are developing a good local know-how in Free Software (which will help others who want to switch).
The first steps are always the hardest.
“For example, gnucash, koffice, konqueror, etc. don’t work well on windows so Windows users will just use Quicken, MS office and IE and they will never ever switch to any other OS.”
I have been using Quicken on Linux for years and I have colleagues who have run Microsoft Office 2000 and XP on Linux for years as well.
In summary, the three applications you cited for staying in Windows land, Office, Quicken and IE work well enough with Crossover-office, although I’d argue that IE is close to useless and unless you have to use some brain-dead intranet app, stay away from it.
And governments have the power to ask developers to code to w3c standards and to make sure that applications work with all major browsers.
porcel I don’t see crossover as viable for anyone beyond home users and very small offices. Great you can ‘use’ MS Office in linux but who will support you? Microsoft will say “hey don’t run to us if you have a problem! We never intended it to run well on linux.”
Most governments have their own internal support mechanisms. If they chose to use crossover office and Microsoft Office, which most governments wouldn’t do when Openoffice is available, I am sure they would evaluate their usage pattern and workloads beforehand to see if the combination was stable under those circumstances.
And were you to have issues that arose from your use of Crossover Office, you would get support from Codeweavers the makers of Crossover Office.
there’s a limit to what codeweavers can do to help because they don’t control the source to Office. It will always be second class support and when it comes to critical govt documents I’d rather have them use Windows and Office instead of risking it on some frankenMSlinux
The more people get “good deals” from Microsoft, the less money it is going to bring in. That will eventually be a problem.
Given a five-year delay in getting your new system out, a smaller company like Commodore would probably have gone bust.
Oh wait, it did.
Microsoft get by because of their massive cash reserves; it helps to offset these cock-ups like ME being useless or Office98 not having Office95 filters.
Reminds me of the American vs. Soviet approaches to writing in space. The Americans spent millions (billions?) developing pens that would write in space.
The Russians used…
…pencils.
http://www.truthorfiction.com/rumors/s/spacepen.htm
OK, so the space pen vs. pencil meme is a myth. The point still stands that when you have money to throw around, you accept less cost-effective solutions than when you don’t. Necessity, not luxury, is the mother of invention.
Reminds me of the American vs. Soviet approaches to writing in space. The Americans spent millions (billions?) developing pens that would write in space.
The Russians used…
…pencils.
I’m sorry, but that’s just urban legend.
A great story, very appealing, but mythical.
http://www.thespacereview.com/article/613/1
http://www.esa.int/esaCP/SEM9YN7O0MD_index_0.html
http://www.pencilrevolution.com/2005/12/russian-space-pen/
etc.
<edit> Sorry again, I was too late with this.
Uknow, I think I must have spread the story myself, too some day. And I’m 100% sure that similar things actually do happen allthe time. </edit>
Edited 2006-12-10 23:51
Yes, we’ve already discussed the point:
http://osnews.com/permalink.php?news_id=16684&comment_id=190546
http://osnews.com/permalink.php?news_id=16684&comment_id=190551
Edited 2006-12-10 23:47
The more people get “good deals” from Microsoft, the less money it is going to bring in. That will eventually be a problem.
Yeah, but why would I care about Microsoft?
Microsoft get by because of their massive cash reserves
Indeed. THEIR massive cash reserves. Not mine.
Yeah, but why would I care about Microsoft?
Well, considering you apparently love Windows and they are the sole source, I kinda would if I were you.
Indeed. THEIR massive cash reserves. Not mine.
See above. Sooner or later MS are going to go under. Red Hat will too, and Novell will whether they become more profitable because of the MS deal or not. But their technologies (ok, RH’s tech and some of Novell’s) won’t be rendered obsolete by their disappearance.
Well, considering you apparently love Windows and they are the sole source, I kinda would if I were you.
What’s love got to do with it? 🙂 So, considering that is not true, your point is?
As I said, I could not care less if one day Microsoft disappears. I’ll gladly use ANY system that is cheaper, yet good enough.
That is why 95% or so people and companies are still using Windows. And as I said, no, we’re not sheep, and you guys are not smarter than everyone else.
See above. Sooner or later MS are going to go under. Red Hat will too, and Novell will whether they become more profitable because of the MS deal or not. But their technologies (ok, RH’s tech and some of Novell’s) won’t be rendered obsolete by their disappearance.
As I said, why would I care if one day that happens to Microsoft?
But, you said, “sooner or later” and that is not “today”. I use Windows today because it is cost-justified. In 5 or 10 years from now, if Linux solution is still cheaper and does everything I need, I’ll dump Windows. But today is not the day. So.. you know, we’ve all been hearing about Linux on desktop.. for, what, 10 years now? Let’s wait till that happens and then we can all trash Microsoft for good reason.
What’s love got to do with it? 🙂 So, considering that is not true, your point is?
I doubt very much that is not true.
As I said, I could not care less if one day Microsoft disappears. I’ll gladly use ANY system that is cheaper, yet good enough.
It’s good enough for me, and I don’t need my daily fix of config-file editing as some people on this site are suggesting. Even if I did, that’s basically what Windows Control Panel does. And not only is it good enough for me, but for everyone else who uses Linux.
I’m perfectly willing to listen to reasons why “Linux is not good enough”, for one application or all – there are even some people at my local LUG who will say so. Unlike most Windows fanboys on this site, however, their arguments are not limited to “Linux users are smelly communists and moving your screen pointer requires assembly-language hand hacking and a deep knowledge of binary”-type astroturfing.
As I said, why would I care if one day that happens to Microsoft?
You’ll care when you have to migrate all your old data by hand because OO.org 2025 only understands Office files up to 2006, after which MS started using DRM to “protect” Office files from other apps.
But, you said, “sooner or later” and that is not “today”. I use Windows today because it is cost-justified.
I don’t Windows today because, for me, it isn’t.
In 5 or 10 years from now, if Linux solution is still cheaper and does everything I need, I’ll dump Windows.
Not without a lot of pain from DRM-infected files, you won’t.
But today is not the day. So.. you know, we’ve all been hearing about Linux on desktop.. for, what, 10 years now? Let’s wait till that happens and then we can all trash Microsoft for good reason.
I started using Linux on the desktop ten years ago. The last time I tried to install Windows XP I failed (and considering that I’m one of these supposed uber-hackers who uses Gentoo, that’s BAD). I’d be quite happy using openSUSE if it weren’t for the MS FUD, too.
MS said Win98 was a really good operating system. I don’t need an OS that crashes that well. By your own criteria, I have good reasons to trash Microsoft.
twenex, I’m getting too tired to answer every single thing you said in your last post.. Most of them are just plain wrong.
I’ll address one (last) thing though:
I started using Linux on the desktop ten years ago.
Then, you’re such a Linux fanboy, that all discussion is useless, no offense.
Who in their right mind would use Linux on desktop 10 years ago??? No OpenOffice, no decent browser, hardware support was horrible, same with desktop experience, etc, etc..
You need to stop, sit down and rethink your life 🙂
Who in their right mind would use Linux on desktop 10 years ago???
Someone who needed a system that didn’t crash every five minutes. That may be an exaggeration. “Every half hour to every hour” wouldn’t be.
No OpenOffice, no decent browser, hardware support was horrible, same with desktop experience, etc, etc..
WordPerfect, Netscape, it worked with the hardware I had, WindowMaker didn’t have ActiveDesktop, etc. etc..
Then, you’re such a Linux fanboy, that all discussion is useless, no offense.
You’re an arsehole who doesn’t know the first thing about what he’s on about. “No offense”.
Correction, I started using it in 1999; I must have tried Red Hat in about 1997 or so. It didn’t last long at that time.
Edited 2006-12-10 23:07
Then, you’re such a Linux fanboy, that all discussion is useless, no offense.
I don’t know how starting to use something ten years ago makes you a fanboy. If discussion is useless, then why did you bother to comment – or was there some other motivation?
Who in their right mind would use Linux on desktop 10 years ago??? No OpenOffice, no decent browser, hardware support was horrible, same with desktop experience, etc, etc..
Linux and various Unixes were used quite happily in many environments where people were less keen on rebooting their PCs every couple of hours. Remember the memory leak in Windows 95?
Who in their right mind would use Linux on desktop 10 years ago??? No OpenOffice, no decent browser
Well, that’s a Windows fanboy comment if ever I saw one.
I’m still waiting for an actual comment by you with a point, that is related to this article, as to why Linux is such a poor fit in government environments.
The fact is, you’ve given nothing to reply to.
FYI, I know people who’ve been using Linux since 1993 or so. They’re probably more clued-up than me, but whether they are or no, if it weren’t for their programming/advocacy/funding efforts I wouldn’t have been able to use it in 1999. If it weren’t for people like me using it now, the people who will convert in the future wouldn’t be able to either.
//Who in their right mind would use Linux on desktop 10 years ago??? No OpenOffice, no decent browser, hardware support was horrible//
Strangely enough, “No Office, no decent browser, hardware support was horrible” is a valid description of a fresh-out-of-the-box install of Windows even to this very day.
Typical Linux distributions, OTOH, do include a full Office suite, two decent browsers and have vastly more supported hardware than the Windows install disks do.
Well spotted! +5!
Strangely enough, “No Office, no decent browser, hardware support was horrible” is a valid description of a fresh-out-of-the-box install of Windows even to this very day.
Typical Linux distributions, OTOH, do include a full Office suite, two decent browsers and have vastly more supported hardware than the Windows install disks do.
He said that he started using Linux 10 years ago. 10 years ago, I’d easily add MS Office and any other application I needed on my Windows. Not on Linux. Decent desktop apps didn’t even exist on Linux back then. Would be nice to know what desktop Linux usage he (twenex) was talking about
Besides, companies use custom made images to roll out systems and all needed applications. You prepare images the way you want including adding all the drivers (if any are missing), applications, system settings, etc, etc. And then you deploy that image.
Having 5 browsers, 50 text editors and 3 Office suites (out of the box or not) would only mean more trouble for the tech support.
Obviously you don’t know that.. You’re just like most linux fanboys – ignorant. Managing IT infrastructure in any company is not the same as toying with your Linux box in your mother’s basement.
Edited 2006-12-11 02:44
//Decent desktop apps didn’t even exist on Linux back then.//
I did not claim that they did.
You, however, did conflate the following: “I’d easily add MS Office and any other application I needed on my Windows” and shortly after “Decent desktop apps didn’t even exist on Linux back then” … with a hidden implication that decent desktop apps did exist on Windows.
Well, in the original context, which was: “No Office, no decent browser, hardware support was horrible”, certainly you are not correct for Windows circa 1996. There was no decent browser, hardware support for the year-old Windows 95 was terrible, and no-one in their right mind would sensibly clain MS Office for Windows 95 was a decent Office application.
More to the point, the hardware support and Office was not included in Windows 95 (which was the most current version of Windows 10 years ago). Most hardware support came on extra disks which came with the hardware (and a lot of that was still Win 3.1 only), and certainly there was no Office suite application to be found on the Windows disks (and there still isn’t, to this day).
//Besides, companies use custom made images to roll out systems and all needed applications. You prepare images the way you want including adding all the drivers (if any are missing), applications, system settings, etc, etc. And then you deploy that image. //
This only works for Windows & Windows software for large companies, and it only works if all the desktops have the same or very similar hardware.
For Linux, it works for all sizes of organisation, and it works for users who have a disparate mixed bag of hardware. It can even be made to work for organisations which have some non-x86 hardware.
Edited 2006-12-11 03:28
There was no decent browser, hardware support for the year-old Windows 95 was terrible, and no-one in their right mind would sensibly clain MS Office for Windows 95 was a decent Office application.
It was better than anything available on Linux back then. Also, Windows 9x provided ability to run DOS apps, which was a big thing. You could have used WordPerfect, Borland Quattro or Lotus 1-2-3, Clipper, dBASE III+, IIRC. Did those run on Linux?
Most hardware support came on extra disks which came with the hardware (and a lot of that was still Win 3.1 only), and certainly there was no Office suite application to be found on the Windows disks (and there still isn’t, to this day).
Compared to Linux which had exactly what kind of support in terms of hardware and software back then? And, were you able to easily move from DOS to Linux? Yeah, right..
Less than 1% of PC users is running Linux today, yet you’re telling me it was decent choice back in 1996? Maybe it was.. for twenex, you, Linus and Richard Stallman.
For Linux, it works for all sizes of organisation, and it works for users who have a disparate mixed bag of hardware. It can even be made to work for organisations which have some non-x86 hardware.
Yeah, is that why nobody uses it (less than 1%, that is)??
Are you that unrealistic?
C’mon.. Linux is free. Yet, it captured less than 1% of desktop market in almost 15 years. So what’s wrong with it? Obviously, it is not the price.
//C’mon.. Linux is free. Yet, it captured less than 1% of desktop market in almost 15 years. So what’s wrong with it? Obviously, it is not the price.//
There is nothing wrong with Linux other than: (1) it is not file-format compatible with the formats of some applications for Windows where the vendor of that application has deliberately kept the format a secret, and (2) it lacks some drivers for obscure hardware (as does Windows out of the box) where the hardware manufacturer provides a seaparte disk for Windows but not Linux, and keeps the specifications to write a driver for that harware a secret, and (3) it is not installed by default by OEMs in America, because Microsoft will refuse discount on Windows to OEMs who install Linux on some of their machines, and (4) it is not adevrtised.
None of those failings is the fault of Linux, except possibly the last.
However, there are people who are prepared to be patient and tell people about it. This requires that other people listen, and I somehow doubt that you are prepared to do that.
Edited 2006-12-11 03:56
(1) it is not file-format compatible with the formats of some applications for Windows where the vendor of that application has deliberately kept the format a secret,
Welcome to the real world Neo
So why don’t those vendors provide their apps on Linux?
(2) it lacks some drivers for obscure hardware (as does Windows out of the box) where the hardware manufacturer provides a seaparte disk for Windows but not Linux, and keepes the specifications to write a driver for that harware a secret,
Again, you Neo?
So why don’t they provide drivers for Linux?
(3) it is not installed by default by OEMs in America, because Microsoft will refuse discount on Windows to OEMs who install Linux on some of their machines,
Microsoft can’t do that any more, since that trial, I believe. It’s been, what, 5 or 6 years since then?
Anyway, people don’t want to buy Linux, can’t you understand that? When they do, they get pirated Windows on their computers. It’s that simple.
Most Eastern European and some Asian countries are good exaple for that. 99% of the people buy no-name computers over there (because they can’t pay for preinstalled Windows) and then they all install pirated copies at home.
(4) it is not adevrtised.
In order to spend money, you have to make it. It is hard to make money with Linux because everyone expects it is free ($0). Is it Microsoft’s fault, too, that they use that model?
None of those failings is the fault of Linux, except possibly the last.
Rest of the faults are all the Evil Empire’s faults, right, right Always somebody else’s fault..
One last thing — IT departments are WELL aware of Linux, so the fact that it is not preinstalled or heavily advertised and reason for Linux’s desktop failure is just wrong. Name one IT person that doesn’t know about Linux.
//Anyway, people don’t want to buy Linux, can’t you understand that? When they do, they get pirated Windows on their computers. It’s that simple. //
And I quote from the lead-in text provided by Thom: “During the past few years, Europe’s elected officials have made a lot of noise about ambitious projects to switch to open source software, including big migrations of government PCs in France, Germany, Spain and Norway.”
There is a large groundswell of organisations and people wanting to get off the Microsoft proprietary treadmill, but are finding it difficult to do from a practical perspective.
This article (and indeed the whole thread) illustrates the effectiveness of Microsofts lock-in policies, it does NOT support your contention that people don’t want to use Linux. Rather, the very opposite.
Once again, nice try, but no cigar.
Edited 2006-12-11 04:08
You practically didn’t answer any of the questions from my last 2 posts..
So, same to you: nice try, but facts are the facts: Linux is free, yet less than 1% of PC users is actually using it. Heck, even most of the hits on Distrowatch is made by Windows users – that is you, “Linux fanboys”: you guys all have and use Windows, 99% of the time, yet here, you claim that you actually use Linux. I guess it is cool to trash Microsoft in front of your classmates. U dA h4ck3r…
//You practically didn’t answer any of the questions from my last 2 posts.//
What questions?
Or, should I say … what on topic, non-leading questions?
//Linux is free, yet less than 1% of PC users is actually using it.//
The figure is inaccurate, but the thrust of your statement is right … Linux has a quite low usage and adoption rate. That is correct. That is, in fact, what the original article was saying. No-one disputes it. What exactly is your point? Indeed, what is your question?
//Heck, even most of the hits on Distrowatch is made by Windows users – that is you, “Linux fanboys”: you guys all have and use Windows, 99% of the time, yet here, you claim that you actually use Linux. //
I use Windows at work (because that is what the company installs) and Linux at home (because that is what I prefer and it means I own and control my own machines). Do you have a problem with that? If so, what is your problem, and can I help you with anything that is bothering you?
//I guess it is cool to trash Microsoft in front of your classmates. U dA h4ck3r…//
You do seem to be such an angry person. I can fully understand that if you have to use Windows and all. I too have spent many an hour trying to unfrazzle Windows systems and solve questions related to Microsoft software, when if only people would use the open alternatives it would all be so much simpler …
Edited 2006-12-11 04:40
What questions?
Or, should I say … what on topic, non-leading questions?
You mentioned lack of extra drivers and vendor-locked file formats as reasons for Linux’s poor adoption on desktop.
And I asked — well, how come that they don’t provide those drivers, and those applications that understand those file formats?
You see, the problem is not that file-format is “locked”, the question, actually, is — why don’t you have those apps on Linux, that can work with those files.. Same for the drivers, etc, etc.
I use Windows at work (because that is what the company installs)
Why do they install Windows then?
You could go and talk to your boss: tell him/her you have a cunning plan to save them some big $$$ by moving to Linux? Have you tried? )
Let me guess: you don’t work for the government 🙂 LOL
//You mentioned lack of extra drivers and vendor-locked file formats as reasons for Linux’s poor adoption on desktop.
And I asked — well, how come that they don’t provide those drivers, and those applications that understand those file formats? //
Because they are proprietary.
This is not universal. For that reason, I would, for example, recommend HP printers. If I was looking for (for examplee) a wireless card I would look here first:
http://www.linux-drivers.org/
and then here:
http://www.linux-drivers.org/network.html
… and I would simply choose one of the many, many alternatives that did have a native driver for Linux.
I would in preference choose a card where the manufacturer had provided help in getting the driver running as opposed to one where the Linux developers had to work it all out for themselves.
Similarly, if I was looking to buy a notebook, first I would look here:
http://www.linux-drivers.org/notebooks.html
and here:
http://www.tuxmobil.org/laptop_manufacturer.html
… and then I would choose one of the many, many choices that was around the right price and which had the best specifications and compatibility.
//And I asked — well, how come that they don’t provide those drivers, and those applications that understand those file formats? //
And I told you, drivers isn’t a problem. There is better support on Linux than on Windows for drivers. It is far easier to get help if you do have a problem, anyway.
As for file formats … even that is less of a problem than it seems to many. I have no problem opening any file format I need to in my use of Linux at home. At work, the only problem I have is with Microsoft Access files, and even my companies IT department agrees this format is a huge problem, and is looking to replace Access throughout the company.
//You see, the problem is not that file-format is “locked”, the question, actually, is — why don’t you have those apps on Linux, that can work with those files.//
As I say, there is really only one format that is a problem. Everyone is looking to replace that format and the application that runs it on Windows because it is a problem.
Windows fans do not seem to understand this … file formats, compatibility and cross-platform interoperability are requirements of end users, and proprietary vendors that refuse to supply what customers want and need are in jeopardy of being abandonned by their customers.
It is not a case of abandonning Linux because it has trouble with Access files, it is far more on the cards to abandon Access because of the file format lock-in.
//You could go and talk to your boss: tell him/her you have a cunning plan to save them some big $$$ by moving to Linux? Have you tried? ) //
I have no influence with the companies global policy on what it provides on the desktop. However, on my own projects, because of “future proof” and “interoperability” requirements, we do indeed provide Linux solutions to our customers. In this context, I happen to BE the boss that makes decisions like this.
//Let me guess: you don’t work for the government 🙂 LOL//
You are correct. I don’t work for the government. However, government departments of my country and other countries are my customers, so does that count?
In this context, I happen to BE the boss that makes decisions like this.
And since you said you use Windows at work — what are we discussing then? You use Linux at home (for fun; although I can’t see anyone having fun in using Linux) and Windows at work. Because at work you do some real work, right?
So, the article linked is right on spot.
Hell, I can advocate Linux like that, too.
It is not a case of abandonning Linux because it has trouble with Access files, it is far more on the cards to abandon Access because of the file format lock-in.
How’s that?
Almost every database has its own SQL dialect and most of them are far from standard SQL. You know with Oracle, you have PL/SQL, with SQL server it’s T-SQL, MySQL has its own, and so on and on and on.. But there’s a problem with Access??? Also, it is quite trivial to use Access database from, for example, Java.. or to expose it through a web service, etc. Where exactly is the problem?
Should we all get rid of Oracle, MySQL, SQL Server, etc, etc, too?
Edited 2006-12-11 06:00
//And since you said you use Windows at work — what are we discussing then? You use Linux at home (for fun) and Windows at work. Because at work you do some real work, right?//
Why do you guess at things about which you know absolutely nothing?
I said I used Windows at work – on the desktop, because that is what the company’s IT department dictates. I do not use Windows in the products I desgin and sell to my customers, however, because Windows is too closed and locked in and otherwise encumbered (for both their purposes and mine).
//Also, it is quite trivial to use Access database from, for example, Java.. or to expose it through a web service, etc. Where exactly is the problem?//
The problem is that Access runs on only one platform, and that platform has many weaknesses, and that platform requires that I impose purchase & maintenance costs of licenses, and risks associated with noncompliance, and risk of audits, onto my end users. Not to mention the very palpable risk that their data will be compromised or destroyed via a security weakness on the platform where it was being manipulated and stored.
If I want to include a database capability as part of the solution, then I would not suggest using Access as part of the solution for these reasons. I am not talking just about some screens of info delivered via a web interface, I am talking about a whole programmable relational database solution.
My end users will also often require that I ensure that their data can be accessed for 30 years. That is undeniably not met by any Access solution. In fact, the only way I can deliver a solution that meets that criteria is to supply a database along with source code, so that it can at least theoretically be compiled & run on some theoretical platform of the future.
With Access, I have already run into a position where data which is only 10 years old can no longer be supported on platforms available now.
In short, to summarise, my customers require that they own as much as possible of the end solution that I provide. If I provide them with Microsoft products as part of the solution … it turns out that they don’t own the software and neither do I, and neither of us control it. That is unacceptable.
Edited 2006-12-11 06:19
I said I used Windows at work – on the desktop, because that is what the company’s IT department dictates.
So, why don’t you suggest to them to switch to Linux? You can save them a lot of $$$, I’m sure no one will refuse that offer.
You’re simply avoiding to answer the question.
With Access, I have already run into a position where data which is only 10 years old can no longer be supported on platforms available now.
And how is that? Yes, you can access that data on Linux, as I said. You can easily move that data to another database, too.
All you need is appropriate JDBC driver for Java, for example. And those exist. And then you can expose/use that data on any platfrom that runs Java.
//So, why don’t you suggest to them to switch to Linux? You can save them a lot of $$$, I’m sure no one will refuse that offer.//
I told you, I have no place to say what the company uses for its desktops.
//You’re simply avoiding to answer the question. //
How so, since I told you the answer.
//And how is that? Yes, you can access that data on Linux, as I said. You can easily move that data to another database, too. //
My customer does not simply have a requirement to access the data. My customers requirement is to access the data, to run the relational database and to create new forms, queries, reports and macros.
It turns out, BTW, that you can access the data on Linux or on Windows, but you can only run the database on Windows 2000 or before. It turns out that you cannot buy a Windows 2000 platform today, so that once the computers of today wear out, the game is up.
Ergo, Access fails the requirement. Utterly.
Just to add, Microsoft’s OpenXML file format was recently accepted and approved by ECMA. European organization, I believe
Please read..
http://www.eweek.com/article2/0,1895,2070078,00.asp
Get the facts, pal.
Edited 2006-12-11 06:28
Oooh, now wait for the iminant Fatwa to declare a Jihad against Microsoft and all those who ‘co-operate with the infidels’.
Anyone here realise that Microsoft has its marketshare because it actually makes a bloody good product – Office; and as much as the opensource crowd loving screaming and gnashing of teeth, there is yet to be a single opensource equivilant which can match the depth and breadth of features and server/client integration.
People point out wins – good, and then 2 years later they switch back, like what happened in the UK, a huge announcement that StarOffice would be deployed, it was deployed then 2 years later it was replaced with Office.
Anyone here realise that Microsoft has its marketshare because it actually makes a bloody good product – Office
Umm, no. It makes a good product that no-one can copy because it is proprietary and closed-source, not because everywhere outside MS lacks talent and business nous. If MS stayed out of office software, then another proprietary company would have taken the lion’s share of the Office market – unless open source software had already become prominent.
Considering MS’ tight grip on its Office formats, it’s amazing OO.org works as well as it does, and next time MS changes “Open”XML without releasing the details of the changes, the OO.o people will have to futz about figuring out what the hell is going on. Again.
The meme that Microsoft is chock-full of geniuses and experts, and that everyone in the software industry that isn’t Microsoft is incompetent is nothing but an MS fanboy’s wet dream.
//The meme that Microsoft is chock-full of geniuses and experts, and that everyone in the software industry that isn’t Microsoft is incompetent is nothing but an MS fanboy’s wet dream.//
Actually, in a sense it seems more likely that everyone in the software industry outside of Microsoft are the geniuses and experts. After all, it is they who duplicate or emulate Microsoft products (or a bit darker, it is they who find ways past Microsoft security provisions) … and they do this without any documentation of Microsoft’s secret bits.
Very true.
Just to add, Microsoft’s OpenXML file format was readily accepted and approved by the ECMA.
The ECMA is a computer manufacturers’ organization. In Europe. As opposed to ISO, which is a standards organization. Worldwide. And guess who preloads MS products on theirs? Hint: It’s not members of the European Potato Growers’ Association.
Yes, the ECMA has already submitted the standard to ISO via a fast track process. That doesn’t mean the ISO is going to ratify it. (Unless you can say that it definitely will because the ECMA has control over the ISO, in which case neither ECMA nor ISO ratification matters a bean.)
There are several problems with OpenXML, not the least of which is this:
http://www.sutor.com/newsite/blog-open/?p=1145
And if MS were REALLY interested in an open standard, why not just contribute to ODF, on which work had already started, instead of coming up with their “own” “open” standard. That’s a contradiction in terms, by the way.
Get the facts, pal.
Edited 2006-12-11 11:50
You mentioned lack of extra drivers and vendor-locked file formats as reasons for Linux’s poor adoption on desktop.
And I asked — well, how come that they don’t provide those drivers, and those applications that understand those file formats?
Because the manufacturers for devices Linux doesn’t support won’t provide drivers or specs. Also, some file formats are closed and proprietary. That is the definition of lock-in. I suggest that next time you attempt to prove that x does not exist, you first look up the definition of x.
when if only people would use the open alternatives it would all be so much simpler
Until they change the mouse pad and then they have to recompile the kernel, right..
http://en.wikipedia.org/wiki/Flamebait
Was that joke not obvious one? I even finished it with
Dooooh!
It is hard to make money with Linux
Someone should tell Red Hat, then:
http://news.com.com/Subscription+boom+boosts+Red+Hat+profit,+revenu…
Someone should tell Red Hat
What about those that do not work at Red Hat? Wait, I know, EU governments have to fund them.
One company (Red Hat).. compared to whole freaking software industry based on Windows?
LOL
Someone should tell Red Hat
What about those that do not work at Red Hat? Wait, I know, EU governments have to fund them.
Pointless bullshit
One company (Red Hat).. compared to whole freaking software industry based on Windows?
LOL
Sounds like you think I said “Red Hat can make money, but closed source/Windows software companies can’t”.
Problem with that argument is, I didn’t.
Seriously, take your meds.
Pointless bullshit
LOL I don’t think so, and you reply only shows that you know it is right.
Sounds like you think I said “Red Hat can make money, but closed source/Windows software companies can’t”.
No, I said making money selling (software for) Linux is very hard. Much easier for Windows.
You replied saying Red Hat blah blah
Well, you forgot to mention, Red Hat is providing not software only but complete biz solutions, from OS to JBoss, custom biz apps, together with servers and support.
I was talking about software only, but even then.. as I said not too many companies makes good penny developing software for Linux.
For example, that poor kid that is developing one popular distribution, Slackware, was recently sick (I’m glad he’s fine now) but he had no money to buy meds for himself. OMFG! Yep, that’s Linux. Oh yeah, and not to mention Mandrake/Mandriva.. etc.
They can’t sell it to anybody. Guys like just won’t pay for it (I hardly suspect you make any money yourself, probably your parents are funding you. And that is the case with most Linux “users”. Plus, we all know you actually use Windows except when you go to OSNews or Slashdot).
Seriously, take your meds.
No pal, you need some serious help ’cause you’re using Linux since 1996, as you said. That is a clear sign of insanity.
I bet you have something that reads “Proud Penguin User #1911 since 1996” somewhere in your mother’s basement.
LOL I don’t think so, and you reply only shows that you know it is right.
Really? well in that case let’s assume, for sake of argument, that if I’d replied “oh you’re so right” that would mean I knew you were wrong.
In case, “oh you’re so right.”
making money selling (software for) Linux is very hard. Much easier for Windows.
It’s always going to be harder selling software for Linux, provided software for Linux stays mostly FOSS, because FOSS products have to compete on quality, not lock-in.
I was talking about software only, but even then.. as I said not too many companies makes good penny developing software for Linux.
Well for example, the reason why Novell don’t make a buck selling Linux is, as recent events have shown, they don’t have a clue.
For example, that poor kid that is developing one popular distribution, Slackware, was recently sick (I’m glad he’s fine now) but he had no money to buy meds for himself.
If he lives in the States, that’s hardly surprising.
OMFG! Yep, that’s Linux. Oh yeah, and not to mention Mandrake/Mandriva.. etc.
I don’t recall Mandriva having no money to buy meds. I never heard of a software product needing meds. Maybe that’s what MS have been missing.
They can’t sell it to anybody. Guys like just won’t pay for it
I assume you mean “guys like me”. Yes well except that I have paid for Mandrake (twice), Gentoo (once), Debian (once), and was on the point of paying £50 (that’s quid, by the way, as in pounds, as in sterling, as in non-American money) for SLED before Novell took That Decision.
Oh, and Slackware.
(I hardly suspect you make any money yourself, probably your parents are funding you. And that is the case with most Linux “users”. Plus, we all know you actually use Windows except when you go to OSNews or Slashdot).
Ah well, let’s see. Two people I know run their own businesses on Linux, one runs his business on Windows and also runs Linux, one other is employed at a company where they use Linux, one uses it at home and doesn’t use a computer at work. The last time I went to Slashdot I was using Windows, and I’m using it right now for reasons beyond my control. Two other people I know use Linux occasionally but (shock) prefer Windows for reasons that, you know, actually exist (unlike yours). That’s 8 real people, as opposed to I-don’t-know-how-many-people you’ve invented in a pathetic attempt to prove your “point”.
No pal, you need some serious help ’cause you’re using Linux since 1996, as you said. That is a clear sign of insanity.
Actually I said for about ten years, which, in December 2006, could near as dammit mean since 1997; then I corrected myself and said since 1999. You’ve shown yourself incompetent of assessing the state of Linux both now and in the past, however, so probably the only thing that makes me insane is bothering to reply to your trolling tripe.
I bet you have something that reads “Proud Penguin User #1911 since 1996” somewhere in your mother’s basement.
I once used something similar as part of a sig, but I don’t have any object like that – not least because mother doesn’t have a basement.
And I’d rather be a proud user of Linux – scratch that, I’d rather be an ashamed user of Linux – than a proud user of Windows, if the only proud users of Windows are anything like the Windows fanboys who infest this site.
Edited 2006-12-11 15:38
If he lives in the States, that’s hardly surprising.
He for sure is the first person that works in IT that I know of, that can not afford medcare. Especially in the US. Maybe it has something to do with what he does?
Oh, no, no, it is because he lives in the States, right, right..
don’t recall Mandriva having no money to buy meds. I never heard of a software product needing meds. Maybe that’s what MS have been missing.
Maybe you recall how Mandrake went bancrupt?
You’ve shown yourself incompetent of assessing the state of Linux both now and in the past
Assessing the state of (desktop) Linux? ) Pal, it costs $0 yet nobody wants it. That is the reality: people rather pay for Mac OS X and Windows than use Linux for free. There’s nothing to assess basically. It’s crap. No wonder it’s free. Today, I wouldn’t use it even if I was getting paid to.
And you’ve shown yourself very competent by using Linux on desktop 10 years ago? Nice.. Hehehe.. And please, stop denying it now.
People like you.. you install Linux and then you think you’re suddenly competent.. ’cause it is so geek. It is so geek that you must be competent because you’ve managed to install it
Nope.
I once used something similar as part of a sig
Yet another example of your competency.
l33t h4x0r, huh 🙂
If he lives in the States, that’s hardly surprising.
He for sure is the first person that works in IT that I know of, that can not afford medcare. Especially in the US. Maybe it has something to do with what he does?
Oh, no, no, it is because he lives in the States, right, right..
I was alluding to the fact that US healthcare is renowned for being crap unless you can afford to pay for top care…which plenty can’t. As for IT people, you can’t know many IT people like the ones I know. Most of those who I know (and here I mean the ones who work with Windows, or Windows AND Linux) have to treat illness like a planned reboot at best…something to be avoided at all cost if possible. We aren’t talking about Savlon, here, Pat’s illness was serious.
Maybe you recall how Mandrake went bancrupt?
Yes, and like SGI, they are now doing fine. Your point?
Assessing the state of (desktop) Linux? ) Pal, it costs $0 yet nobody wants it. That is the reality: people rather pay for Mac OS X and Windows than use Linux for free. There’s nothing to assess basically. It’s crap. No wonder it’s free. Today, I wouldn’t use it even if I was getting paid to.
As the article proves, plenty people want it. The problem is vendor lock-in on the microsoft side (apologies to anyone with sense, if i’m starting to sound like a broken record. You just can’t get through to these fanboys).
Furthermore, Windows and OS X “feel” free to most users, who only get a new version when they buy a new computer.
As for it being crap, your argument seems to be that people who actually go looking for other systems, use crap. Choosy users choose crap.
What a load of crap.
If you don’t want to use it, fine. That doesn’t mean anyone who does is a fool; it just means you’re a fool to think so.
And you’ve shown yourself very competent by using Linux on desktop 10 years ago? Nice.. Hehehe.. And please, stop denying it now.
People like you.. you install Linux and then you think you’re suddenly competent.. ’cause it is so geek. It is so geek that you must be competent because you’ve managed to install it
No, I’ve shown myself competent because unlike you, I actually know what I’m talking about when I discuss the state of the industry, now or ten years ago. Installing Linux is no longer hard, as you’d know if you knew anything about it.
Yet another example of your competency.
l33t h4x0r, huh 🙂
Honestly, stop trolling, you’re making yourself look like a fool. Can it be any coincidence that you don’t disclose your real name?
Edited 2006-12-11 19:09
No, I’ve shown myself competent because unlike you, I actually know what I’m talking about when I discuss the state of the industry, now or ten years ago.
Only a fool could have started using Linux desktop 10 years ago. It is your problem that you don’t see it. Most (99%) of the people don’t want to use it today and yet you’re telling us it was good enough 10 years ago?
Yeah, you’re competent.. right 🙂 You started using linux on your desktop 10 years ago, so you must be
Only a fool could have started using Linux desktop 10 years ago. It is your problem that you don’t see it. Most (99%) of the people don’t want to use it today and yet you’re telling us it was good enough 10 years ago?
It was good enough FOR ME in 1999. That’s 7 years ago, btw.
If you can’t understand the concept of me being different from you (thank God), then I can’t make you.
See, I’m kinda of the opposite opinion on this. Not that I’d consider everyone that’s been using Windows for the past 10 years fools mind you, but still, I get on Windows these days and man I miss the Linux way of doing things. I mean, cutting and pasting alone is downright painful now on Windows and reason enough for me to prefer unix style desktops. But add to that virtual desktops, a gui that doesn’t all but lock up when something cpu intensive is going on, a window focus that follows my mouse, a taskbar that actually works when you put it at the top of the screen, and the thousand other little things I could come up with if I sat here long enough, and well, at that point it’s no contest. And this is coming from a guy that started out on Windows, so it’s not like I’m set in my ways or something. And then don’t even get me started on how much easier Linux desktops are to admin.
Microsoft can’t do that any more, since that trial, I believe. It’s been, what, 5 or 6 years since then?
Come on, that’s either very naive or equally disingenuous. The reason why they were pulled up in court for doing it is because it was illegal. Short of locking them all up, if they were doing something illegal before the trial, what’s to stop them doing it after?
//Come on, that’s either very naive or equally disingenuous. The reason why they were pulled up in court for doing it is because it was illegal. Short of locking them all up, if they were doing something illegal before the trial, what’s to stop them doing it after?//
Not only that, but everyone knows Windows OEM is a fraction of the cost of Windows retail.
Ergo, Microsoft still discounts Windows OEM.
Nevertheless, all is not doom & gloom, as we still have this to cheer us:
http://lxer.com/module/db/index.php?dbn=14
It was better than anything available on Linux back then. Also, Windows 9x provided ability to run DOS apps, which was a big thing. You could have used WordPerfect, Borland Quattro or Lotus 1-2-3, Clipper, dBASE III+, IIRC. Did those run on Linux?
As I’ve already mentioned, WordPerfect DID exist on Linux in 1999 if not 97. By definition anyone using Linux at that time didn’t need applications that didn’t then exist on Linux (or was dual-booting).
Less than 1% of PC users is running Linux today, yet you’re telling me it was decent choice back in 1996? Maybe it was.. for twenex, you, Linus and Richard Stallman.
I’ve never seen reliable statistics for Linux desktop users, but as you cited “PC users”, not “desktop PC users”, let me point out that sales of Linux servers are above 25%.
Here’s another statistic for you. Less than 0.1% of Windows fanboys on this site know what they’re on about.
C’mon.. Linux is free. Yet, it captured less than 1% of desktop market in almost 15 years. So what’s wrong with it? Obviously, it is not the price.
No, it’s the lock-in from the Windows side, for the umpteenth time.
10 years ago there was a package called Applixware which was even used by brokers on the stock exchange.
http://en.wikipedia.org/wiki/Applixware_Words
(contrary to this article it is not just a word processor but also spreadsheet AFAIK). Wordperfect by Caldera apparently was available for Linux in 1995.
Then there was StarOffice which was later sold to Sun by the developing company.
http://en.wikipedia.org/wiki/Staroffice
//Having 5 browsers, 50 text editors and 3 Office suites (out of the box or not) would only mean more trouble for the tech support.
Obviously you don’t know that.. You’re just like most linux fanboys – ignorant. Managing IT infrastructure in any company is not the same as toying with your Linux box in your mother’s basement. //
First you started with a strawman (arguing against something that I did not say), and now this ad hominem attack.
Linux can be installed in very large deployments across an organisation. Try 100,000 machines, as these people have done (as one example):
http://www.ubuntu.com/support/paid
See where it says:
“Expert advice & large deployments
Ubuntu has been deployed on home computers and in government deployments of over 100,000 machines. “
Having a number of choices gives you a choice, it doesn’t mean you have to install them all.
I think all of that well & truly bounces the “ignorant” tag right back in your lap, not mine.
Nice try, but no cigar.
Ubuntu has been deployed on home computers and in government deployments of over 100,000 machines.
Governments, again? LOL
Home computers bwhahahaha 🙂
Yeah, I also tried it on VMWare.. does that mean I’m Ubuntu user? Yeah, right.
Was that — 100,000 machines, in TOTAL? :-)))
This reminds me of Sun and how they announced (1 or 2 years ago) that they sold, I believe, aroun 10 million Linux copies to Chinese. Everyone made fun of that as it never actually happened.
Freaking Dell sells well over 100,000 Windows PCs every day. Every single day.
Edited 2006-12-11 03:44
//Governments, again? LOL
Home computers bwhahahaha 🙂 //
What exactly is your problem?
“Home coputers” to “government deployments of over 100,000 machines” obviously refers to a range of “scope of deployment”.
//Was that — 100,000 machines, in TOTAL? :-))) //
Are you seriously claiming that you are so ignorant you cannot tell the difference between “government deployements of over 100,000 machines” (ie more than one deployment of that scope) and “100,000 TOTAL”?
You must be even slower than I thought you were.
//Freaking Dell sells well over 100,000 Windows PCs every day. Every single day. //
So?
That fact in no way detracts from the point that it is entirely possible to perform and maintain large deplyments of Linux computers.
You keep trying to argue things that are not the point. You must have got a very poor grade in comprehension at school, I’d wager.
Are you seriously claiming that you are so ignorant you cannot tell the difference between “government deployements of over 100,000 machines” (ie more than one deployment of that scope) and “100,000 TOTAL”?
Well, do you have any evidence to prove that? I thought so.
Sun claimed the same. Never happened.
I’ll address one (last) thing though:
I started using Linux on the desktop ten years ago.
Then, you’re such a Linux fanboy, that all discussion is useless, no offense.
I started using it as such in 1994 or even 93. You’d probably say that makes me Linus Torvalds
Who in their right mind would use Linux on desktop 10 years ago??? No OpenOffice, no decent browser, hardware support was horrible, same with desktop experience, etc, etc..
No OpenOffice? There were Applixware, Wordperfect, although I never needed Office suite.
No decent browser? Same Netscape as on Windows was not decent?
Hardware support? HCL answers all troubles, it was like that, it is like that and it will be like that. No matter the OS some hardware performs too bad to be considered as running.
Desktop experience? Actualy, all my friends envied me on beaty of my desktop. Custom built with custom graphics (first was Afterstep running kdm and all sorts of tweaks, second was Enlightment crossed with Gnome, again custom built). I can say I was more productive in those times than now with vanilla Gnome, desktop was always perfected for my needs in my early linux years.
In those years linux was perfection for “do it your self” kind of people. In that time it was a hacker-only OS, not user-friendly OS. So you might be in trouble only if your skills were not adequate. But if you were good enough, it was a perfection.
You need to stop, sit down and rethink your life 🙂
Shouldn’t you do that? You speak of things you don’t know.
Well said tweenex. The implication of Microsoft putting enforced DRM on Office documents send the shivvers down my spine.
Suddenly things like the Freedom of Information Act etc become irrelevant. You can’t read this government document. Not because its secret but Microsoft has enforced a DRM system thst stops just about anyone but the author from reading it.
This has implications that politicians all over the world have just not woken up to yet and probably will not do so before it is far too late.
IMHO, DRM is a virus that will wriggle its way unseen into our lives
I dare you to say the todays Linux distro Desktops are BETTER than XP or Vista but Linux still lacks in apps and hardware drivers.
Do not forget people use what there used to change is often scary…
Edited 2006-12-10 21:24
That’s the last way to do that. Incrementally you end with one department using Office 2000, a couple of thousand still runnning Word Perfect, another couple of departments on OpenOffice 1.something, another on Open Office 2.something, etc, etc. Different versions of Office cause enough problems, with a myriad of different suites and OS’s.
These guys need to share information, and have hope in hell of it looking the same when chucked over the fence. Oh – and they may have *macros*, charts they want people to be able to read, complex edit tracking / approval scenarios, workflow systems that rely on particular features of Offfice, etc, etc.
Linux just will not run the apps, or hardware, that will run with windows.
You don’t run an OS just to run the OS. The whole point of an OS is to allow your computer to run apps, and to work with hardware. If the OS doesn’t do that, then the OS is not usefull, no matter how good the OS may be otherwise.
Unless Linux can catch up with windows in terms of apps and drivers, then I can not see desktop linux as being much more than a bit player. And I don’t see how Linux can catch up.
none of that matters in a government setting. They’re the government. They can dictate what they will and will not use and support, and everyone else has to cooperate.
Ever work for them?
You might be amazed to learn that linux has support for more hardware out of the box than windows.
For any hardware that isn’t supported there are plenty of other similar hardware from different vendors that are supported.
Applications are still a problem for some industries, graphic designers(Inkscape and Scribe are still playing catch up), musicians(Rose Garden is also playing catch up) and engineers (there is pretty much nothing for CAD) are still going to have to stick with windows for now.
Most office workers that do word processing, web browsing, email, should have no problems running GNULinux
Custom built applications might be a problem, but it might end up cheaper to re-write the application to save on microsoft licencing.
>>
You might be amazed to learn that linux has support for more hardware out of the box than windows.
<<
You mean CPUs. Bottom line, I can not go to my local computer store, and pick up a wireless card, or multi-function printer/scanner/fax, or winmodem, or whatever, and expect it that it will work with linux.
>>
For any hardware that isn’t supported there are plenty of other similar hardware from different vendors that are supported.
<<
I want to use what I want to use. Besides, when it comes to stuff like wireless, linux is just awful.
>>
Most office workers that do word processing, web browsing, email, should have no problems running GNULinux
<<
This is what linux users don’t get, and probably never will. There are tons of apps that only work on windows. Linux users are often high-school, or college, students. They use linux for email and listening to mp3s, and figure that is all anybody needs. The real world is *very* different. One quick example: UPS worldship, it’s used by thousands of small, and large, businesses, and it doesn’t work with linux. Also, 80% of small business use quickbooks.
>>
Custom built applications might be a problem, but it might end up cheaper to re-write the application to save on microsoft licencing.
<<
Not by a long, *long*, shot. BTW: I have worked in IT for 27 years, I use linux at home, and at work. I even have a weenie comptia linux+ certification.
You mean CPUs. Bottom line, I can not go to my local computer store, and pick up a wireless card, or multi-function printer/scanner/fax, or winmodem, or whatever, and expect it that it will work with linux.
When it comes to multi-function printer/scanner/faxes, you have a point. When it comes to wireless, you don’t (see below). When it comes to winmodems, if you want to use those then it’s your funeral.
I want to use what I want to use. Besides, when it comes to stuff like wireless, linux is just awful.
If you pick the cheapest generic crap you can find, then yes, you might have a problem. Being picky about hardware is good whether you use Linux or Windows. And if when it comes to stuff like wireless, linux is just awful, then I need a new adjective to describe the worse state of wireless drivers on Windows.
This is what linux users don’t get, and probably never will. There are tons of apps that only work on windows. Linux users are often high-school, or college, students. They use linux for email and listening to mp3s, and figure that is all anybody needs. The real world is *very* different. One quick example: UPS worldship, it’s used by thousands of small, and large, businesses, and it doesn’t work with linux. Also, 80% of small business use quickbooks.
This is what windows fanboys don’t get, and probably never will. There are tons of apps that only work on linux, or work better than on Windows. Windows users often knnow next to nothing about technology. They use windows for email and listening to mp3s. Others use AutoCAD or Quicken or MS Office, and figure there are no alternative solutions. Two quick examples: When HSBC UK was derided publicly for not letting users do Internet banking from RiscOS, they quickly changed their tune. RiscOS has nowhere near the userbase of Linux even on the desktop (unfortunately), so it can’t be that hard.
Quickbooks doesn’t come with UK tax solutions anymore, so I doubt 80% of UK small business use quickbooks.
When it comes to multi-function printer/scanner/faxes, you have a point. When it comes to wireless, you don’t (see below). When it comes to winmodems, if you want to use those then it’s your funeral…If you pick the cheapest generic crap you can find, then yes, you might have a problem.
**Yep. That’s right, it’s YOUR fault for buying the wrong stuff, even if it’ generic.**
This is what windows fanboys don’t get, and probably never will. There are tons of apps that only work on linux, or work better than on Windows.
**But nobody uses them. But that’s OK, because that’s their fault as well.**
**Sounds a bit like this:
Woman 1: So how did Jonny do on the parade?
Woman 2 (proud mother): He did fine, but everyone spoilt it by marching out of time.**
So what would you say to someone who bought a Lada and then couldn’t understand why it won’t work for more than 5 minutes.
In my experience if a hardware product supports linux it’s plastered all over the box. So finding a product that supports linux, it just isn’t hard.
In comparing Lada’s to Linux, I think that you are underestimating Lada’s.
Whatever.
>>
For any hardware that isn’t supported there are plenty of other similar hardware from different vendors that are supported.
<<
I want to use what I want to use. Besides, when it comes to stuff like wireless, linux is just awful.
When Microsoft releases a new OS, one of the very first thing that comes to mind is to check the HCL that they probably put in place the date that the software hit the shelves so that one can avoid those hardware traps where the scanner/printer/whatever works with one version of the OS but not the latest one.
That happened with Windows NT (which was anything but a consumer OS), happened with Windows 2000, XP and most likely will be the same story all over again with Vista.
I have a scanner which is perfectly capable and functional that doesn´t have a driver for Windows XP therefore I have to resort to the old Win2000 driver (that I´ve found on Driversguide.com as the manufacturer is clueless and couldn´t care less about Windows XP users) and hope that it works.
There is no such a thing as “I want to use what I want to use”, even for Windows users. If the manufacturer does not advertise the product as being compatible with your operating system, then good luck trying to make it work.
You may have a point regarding wireless, but then, this is a smaller issue than you make it appear to be. Wireless Access Points and related technology still is a little bit expensive for third world countries and therefore, it is much less of an issue than, say, application compatibility.
Also, 80% of small business use quickbooks.
Where did you get these numbers from? It has been quite some time since the last time that I´ve seen QuickenBooks here on Brazil. Most small offices seem to be using something that they put together on Excel or Access (which can be converted with ease to a web app, really!) or some custom solution that they bought or hired a company or someone to do for them.
Never heard about the UPS thing, so I don´t think that this is a huge problem either.
What some of you guys need to do is to stop thinking as an American user and try to put things in context and then you will realize that the Linux/BSD solution is closer to being replacement for Windows in certain situations than you think.
I won´t deny that software availability IS a hurdle that must be overcome by Linux in order to turn it into a drop-in replacement for proprietary OSes but this problem has been becoming smaller and tends to keep decreasing for the foreseeable future.
You might be amazed to learn that linux has support for more hardware out of the box than windows.
That’s a useless and irrelevant issue. Fact is, Windows has far wider driver support than Linux could ever hope for. Asking a user to put in an install CD to install the drivers which came with a particular hardware product isn’t a big deal. But it IS a big deal when the Linux drivers DON’T EXIST.
Reality bites, Linux bigots.
//Unless Linux can catch up with windows in terms of apps and drivers//
Linux is already well ahead of even XP in terms of driver support.
It is immeasurably ahead of Vista … Linux will run on a huge array of hardware that Vista will not run on.
There are 20,000+ packages available for Linux in the Debian/Ubuntu repositories, covering almost every imaginable class of application. There would be less than a few percent of users who did not have every application that they could possibly use covered.
The only problem is porting existing data from specific, proprietary applications. As an example: it is not like there is no CAD application for Linux, but rather that AutoCAD is not available for Linux.
See here: http://www.tech-edv.co.at/lunix/CADlinks.html
However, if a sizeable number of “undemanding” users (ie users who did not demand a specific application only, but rather used generic applications such as “web browser” or “wordprocessor” or “spreadsheet”) moved to Linux, then companies like Autodesk would see a demand, and even that problem would be solved.
Edited 2006-12-11 00:01
I think you need to look a little harder and a little deeper; you are correct about drivers, although the installation, configuration obtaining could be improved; given my experience installing ipw3945 on Fedora, its not exactly for the faint hearted.
As for the applications, there is a definlate shortage for those who want to do more than the minimum; for example, look at Photoshop Elements for example, there is nothing that can hold a candle to it on Linux.
Thats at the low end, at the high end, there are such things as Cakewalk, Quark, Creative Suite (the whole suite) and many other titles – no matter how good you make the opensource equivilant, people like using what they’re used to.
Now, if those issues were addressed; then there would be no reason to run Windows – people only stay with Windows because of ease of hardware installation (in regards to setting up when the drivers are provided on cd) and the accessibility of applications, remove those two barriers, and people will jump off the operating system upgrade treadmill and go with Linux or what ever other operating system hits the g-spot that is desktop nirvana.
you are correct about drivers, although the installation, configuration obtaining could be improved; given my experience installing ipw3945 on Fedora, its not exactly for the faint hearted.
Not for drivers to which the Linux developers have the source. The last time I inserted a new NIC all I had to do was find the right module – which wouldn’t even have been necessary if I didn’t compile my own kernel. And no, it’s not necessary to compile your own kernel. It is desirable, however.
You’d have the self-same problem with a Windows device that ignored the “proper procedure” and had its own installation – in fact I did have that problem with some hardware I installed about three months ago. An Epson printer isn’t exactly obscure hardware.
Whether with Linux or Windows, the problem you describe is a device manufacturer problem, not a Linux problem. Reasonable people will see that when the issues are explained to them. Fanboys won’t.
True, I am not arguing with you over that fact; it is a manufacturer problem, but like the issue with Windows, if something does go wrong, the first people/organisation they’ll blame will be the operating system vendor – even if they had no part in the provision of the driver.
On my blog ( http://kaiwai.blogspot.com ) I’ve out lined what I had to do to get the freshrpm ipw3945 package setup along with wpa_supplicant configured.
Like I said, that isn’t for the faint hearted, and I hope that Fedora maintainers do look at that scenario and see that the amount of work I needed to do to get my wireless networking up and running is completely unacceptable in this day and age.
Yes, I have heard before that wpa_supplicant is a bitch.
It’s not Linux which can catch up with windows in terms of apps (or drivers), it’s the developers of the software. They need to support Linux and to do that they need a reason. The biggest reason is demand.
But in the big system space, it is often the case that software is developed to run on specific hardware (not necessarily Windows or Linux), for example specialist hardware such as Tandem (now HP Nonstop) or Mainframe, for example, and the customers wishing to run those systems buy the hardware to match. At that level the apps can be more expensive than the hardware.
Smaller apps are usually cheaper than the hardware, so the customers decide what they will run and buy the apps based on that. In most cases they see that the apps they want run on Windows, so they buy Windows. Occasionally they may choose Linux for web or other server purposes, but when they still think of business apps as MS Office, Outlook, etc. they will buy Windows. If the thinking moves to OOO for example, then they may look more at Linux.
>>
Linux is already well ahead of even XP in terms of driver support.
It is immeasurably ahead of Vista … Linux will run on a huge array of hardware that Vista will not run on.
<<
Non-sense, and I have a wireless card – with a penguin logo on the box – to prove it.
>>
There are 20,000+ packages available for Linux
<<
Doesn’t matter. The apps that are actually used in business don’t run under linux: ms-project, ms-viseo, ups worldship, quickbooks – to name a few. There are thousands of specialized 3rd party apps that only work on windows.
>>
There would be less than a few percent of users who did not have every application that they could possibly use covered.
<<
Way off. Way *way* off.
Linux could catch up, but Linux does not want to catch up. Linux distributions are mainly based on GPL software,
and not every piece of software, particulary a driver is compliant. FSF has got a rigit stance on that, and I can’t see any room for compromise. That’s why the Linux market penetration is limitted and will stay limitted.
Linux community is in perpetual state of ambivalence. They are advocating Linux broader use of Linux and similar software. On the other hand they want software that meets their specifical philosophical and
political agenda, and their own vision of IT, which is not shared by the majority.
People out there are not buying GNU/Freedom ideas today, and the majoity of them never will.
DG
Reminds me when I was in government service once upon a time, long, long ago, unless much has changed new roughly translates to more work without a raise. Not a wonder really, 5% is pretty good for a start, the opportunity to change is the important thing.
everything begins with one step and this sounds like a damn fine one to me…
For a change it is really nice to read this kind of relatively neutral, fact-based studies and even statistics besides/instead of the endless amounts of one-sided propaganda for or against this or that OS/software.
I’m not sure deployments in Extremadura, Vienna, Munich, and a partial deployment in Birmingham count as “negligible”. I’ve probably missed some out, too.
After all, difficulty migrating is the whole point of vendor lock-in. The easiest way to get rid of vendor lock-in is to wait until your vendor goes down the tubes.
In the article, a Gendarm says that OpenOffice is “horrible” and too complicated (even though he also said that MS are a bunch of “thieves”…)
I wonder if they’re using OO.o v1…because I’m using v2.04 right now and I have to say that it’s very, very good!
If they could only add a decent Outline view. To all OO.o enthusiasts out there, please register and vote for this bug:
http://www.openoffice.org/issues/show_bug.cgi?id=3959
I wonder if they’re using OO.o v1…because I’m using v2.04 right now and I have to say that it’s very, very good!
Openoffice Writer is pretty good, I haven’t had any issues with it as a replacement for Word, but Calc is quite poor compared to Excel. The basic features are there, but the whole package is far less intuitive than Excel. I don’t do a lot with it, but for basic stats and graphing Excel is by far the better option. I still wouldn’t spend the money to buy MS Office for myself, so I use Openoffice, but I can’t deny that it is not as good as MS Office in quite a few aspects.
I don’t use Excel enough to judge, though I’ve heard they’re putting lots of effort into this one right now.
I’ve also heard that Impress is very good, though I haven’t really taken the time to play around with it. In any case, the whole project has been progressing at a very healthy pace…now, if they can only get that Outline mode in…
Excel better than OpenOffice? You’ve got to be kidding. Excel is the less intuitive and more buggy IMO. For example, try matrix multiplication in both apps. In Excel you have to select the required amount of cells which really slows you down when you have some big calculations to do. In OO.org you just select one cell.
Well, that is a bit too optimistic – speaking as a user who makes living using word processors. Even leaving format compatibility aside, there’s a lot of usability/efficiency issues with OOo making me way less productive on it than on MS Word 2000. BTW, these are issues with basic things like the lack of “Normal” mode (even Abiword has it, though not completely right!), poor logical model of paragraph formatting, inefficient and inflexible hotkey assignment, poorly documented macro language and incomprehensiveness of the stuff spit by the macro recorder, AutoCorrect working or not working randomly… Surely OOo is good enough for a lot of people, but for a sizeable chunk of workers it won’t be able to replace Word anytime soon.
It depends on what you write, really. I also use Word as my primary worktool, while writing a side project (a novel) OO.o. For that use, it is of course sufficient. I also haven’t had any real issues with AutoCorrect so far…
For general office use it depends if you use a lot of Macros or not. At my job, where we produce design documents for video games, we do not use macros at all, so that wouldn’t be an issue.
As far as normal mode and hotkey assignment go, I personally never use either of them and so that is also not a problem for me.
One place where OO.o actually works better than MSWord is how it handles image placement. We have *lots* of images in our documents, and I regularly curse at Word for getting all funky with images jumping pages and the like. I haven’t seen that weird behavior in my albeit limited time with OO.o.
I’m not sure what your issue is with paragraph formatting, however.
I agree with you, though, that it’s not for everyone…yet. However, I was really impressed with the project’s overall improvement in version 2.04.
‘m not sure what your issue is with paragraph formatting, however.
It’s the fact that I can’t reliable copy just the text of the paragraph or the whole paragraph with the associated formatting. It happens that when I copy even a small fragment of text from the middle of a paragraph, the target paragraph becomes formatted like the source one. This does not happen always, and I suspect that this is related to bullets/numbering. Speaking of which, I don’t know who decided that bullets and numbering are NOT a part of paragraph formatting, but that person surely must be subjected to a cruel or unusual punishment. I often need to clear formatting from a paragraph, but bullets and numbering still remain after that operation.
It’s the fact that I can’t reliable copy just the text of the paragraph or the whole paragraph with the associated formatting. It happens that when I copy even a small fragment of text from the middle of a paragraph, the target paragraph becomes formatted like the source one.
Ah, yes, I see what you mean. Word also does some similar weirdness, but only if you select the entire line (by clicking in the margins or by selecting the normally invisible paragraph mark) instead of selecting just the text.
This is a valid point of criticism. There should be a way to copy only text, and not the formatting (Lotus Notes is also very vexing for that, so much so that I now keep a copy of Notepad open just to use as a copy/past buffer that strips text formatting). They should definitely include a way to turn this off.
Speaking of which, I don’t know who decided that bullets and numbering are NOT a part of paragraph formatting, but that person surely must be subjected to a cruel or unusual punishment.
I agree with that as well. You should bring these points up on the bug database, after all the cool thing about OO.o being an open source project is that you can participate in its development.
Copy
Edit-Paste Special-Plain Text
Yes, you should be able to set the default paste method in any applicaiton!
Temcat: Surely OOo is good enough for a lot of people, but for a sizeable chunk of workers it won’t be able to replace Word anytime soon.
That’s what they said about MS Word in comparison to StarWriter and WordPerfect in the 1990s. I, for example, am a StarWriter user since DOS times, StarOffice 3.1 was my last MS Windows version, StarOffice 5.0 my last GNU/Linux version and my switch to OpenOffice has been painless. OOo2 actually is quite a dog for me because it tries to emulate MS Office so hard, which is alien to me.
Thus, I’d argue that your issue is more about things we are used to and less about how things are actually done.
In the article, a Gendarm says that OpenOffice is “horrible” and too complicated (even though he also said that MS are a bunch of “thieves”…)
Thanks. At least you didn’t call him a moron for not liking FOSS alternatives.
Thanks. At least you didn’t call him a moron for not liking FOSS alternatives.
Was there an actual point for your comment? Because I’m trying very hard to find one, and I can’t.
Why would I call him a moron for not liking something? I think you have me confused with NotParker or some other member of your Trolling club.
For your information, I’ve used MS Office for years, and still use it. I do like it better than OpenOffice, unfortunately it does not run natively on Linux. That said, I’m unbiased enough to recognize that OpenOffice has made great strides over the past couple of months, and that for *many* users, it’s a more than adequate replacement.
You can and should read articles such as this one as part of the damage control and propaganda campaigns of Microsoft. The article provides a few anecdotes based on anonymous sources which cannot be named. You can be certain that this article is bullshit.
If it can deny the incredible momentum behind Linux, it can make people second-guess their own decisions. It is very astute and yet very transparent what they are trying to do.
Mandrake 2007, Ubuntu LTS or Suse’s SLED are better desktop operating systems than anything Microsoft currently offers, particularly if you care about cost, security, rate of innovation and improvements, data transparency, retrieval and archival into the future, and not depending on a single provider of technology who has proven itself unable and unwilling to work cooperatively with the rest of the IT industry.
“Mandrake 2007”
Mandriva 2007.
Seriously, are you high? Where do you get your drugs? Linux has less than 1% of the desktop OS market. You call this “incredible momentum”? Fine, whatever. I think that Linux has done remarkably well in the server room, but not on the desktop. So far, it’s been a non-starter there, and there doesn’t seem to be any momentum which will change that.
Linux has less than 1% of the desktop OS market.
Actually, it has between 3 and 6%.
Actually, it has between 3 and 6%.
There is a 3 and a 6 involved.
.36%
IDC, a company which you yourself have described as reliable, places Linux at 3% as of Jan. 2005, estimating 6% for 2006.
The number you give is based on web stats, which cannot be used to accurately reflect market share.
Stop spreading misinformation in order to further your anti-Linux agenda.
IDC, a company which you yourself have described as reliable, places Linux at 3% as of Jan. 2005, estimating 6% for 2006
Gartner suggested 1.3% at most because of all the people who remove Linux and replace it with a pirated copy of Windows. And that was a 2 year old estimate.
You see, the desktop market has grown condierably in the last two years. Linux hasn’t.
Do you have any numbers from 2006?
Linux on TheCounter has 1/9th the share Win98 does (3%).
http://www.thecounter.com/stats/2006/December/os.php
Hitslink has Linux at .37%.
http://marketshare.hitslink.com/report.aspx?qprid=2
As we all know, Linux market share in servers is dropping. There is no reason not to think Linux’s desktop share has dropped as well from 1.3%
In Q1 of 2006 Linux had 12.2% of server revenue.
In Q2 of 2006 Linux had 12.0% of server revenue.
In Q3 of 2006 Linux had 11.8% of server revenue.
References:
http://www.idc.com/getdoc.jsp?containerId=prUS20180706
“Although Linux servers now represent 12.2% of all server revenue, revenue growth for the quarter was approximately half the growth rate observed in 1Q05”
http://blogs.zdnet.com/ITFacts/?p=11637
“Linux servers now represent 12.0% of all server revenue, up slightly from Q2 2005.”
http://www.itjungle.com/breaking/bn112206-story01.html
“Linux machines accounted for 11.8 percent of overall server sales, and its growth has moderated considerably.”
You just cut and paste your replies. You know what you sound like? A broken record.
IDC had Linux at 3% in Jan. 2005.
Webstats are too inaccurate to determine market share.
Linux server has 28% market share (which by default means per unit).
Windows’ greater market share per revenue is due to it being overpriced.
Linux server growth is moderating towards a healthy 5%, more than Windows, after enjoying double-digit growth because it basically started from near zero.
Okay, I think I’ve pretty much covered all your disinformation here. I’ll do like you do, and just keep copy/pasting this in replies to your own copy/pasting. Deal?
BTW. I wasn’t replying to you originally…are you stalking me by any chance? 🙂
IDC had Linux at 3% in Jan. 2005.
That was an estimate 2 years and 250-300 million PC’s ago (most installed with Windows on them).
Do you have anything current?
Linux server has 28% market share (which by default means per unit).
Not according to numbers from 2006.
It has 11.8% of server revenue share which factors in multiple CPU servers based on revenue.
It really doesn’t make sense to count a 500$ 1 CPU server the same as 5,000,000 128CPU server.
Windows’ greater market share per revenue is due to it being overpriced.
Unix beneifts from revenue numbers because they tend to be multi-CPU servers. Just like Windows.
Linux server growth is moderating towards a healthy 5%
When your growth drops from 30% to 5.4% in one year the last data point on a downward sloping graph is not levelling out.
Linux revenue growth is approaching zero. Once upone a time (3 years ago) it was at 60% per year.
60% *
40% ..*
20% …*
6.1%….*
5.4%……*
are you stalking me by any chance?
Countering your out of date FUD estimates with facts is not stalking.
NotParker, I’m tired of you constantly misrepresenting numbers to make them say what you want.
You sound like a broken record. Since you keep copy/pasting the same disinformation, I’ll just keep correcting you by doing the same.
You just cut and paste your replies. You know what you sound like? A broken record.
IDC had Linux at 3% in Jan. 2005.
Webstats are too inaccurate to determine market share.
Linux server has 28% market share (which by default means per unit).
Windows’ greater market share per revenue is due to it being overpriced.
Linux server growth is moderating towards a healthy 5%, more than Windows, after enjoying double-digit growth because it basically started from near zero. It would have been impossible for Linux to continue growing at such a phenomenal pace, because such a growth would have been exponential.
How much math did you study? Because your gross misunderstanding of statistics is appalling.
Ill correct some more new lies:
Unix beneifts from revenue numbers because they tend to be multi-CPU servers. Just like Windows.
Wrong. Linux tends to be found more multi-CPU servers than Windows. Linux scales better than Windows on multi-CPU servers.
There is no rhyme or reason to your latest fantasy that market share by revenue is more representative than by unit. They’re just two way to look at the same data, but you prefer to present it by revenue because the Linux numbers appear smaller.
Oh, and I love how you throw so many numbers around that you lose yourself in them. On one line you say:
When your growth drops from 30% to 5.4% in one year…
And then you say that it went from 20% to 5.4%.
20% …*
6.1%….*
5.4%……*
Another example of your lack of mathematical knowledge is when you claim that “the last data point on a downward sloping graph is not levelling out.”
As I have demonstrated before, the difference between Year -2 and year -1 is greater than between year -1 and year 0. This indicates that the rate of decrease is itself decreasing. Also, the difference between Q3 and Q4 is much, much less than (year 1 – year 0)/4 which means that the rate of decrease has almost leveled out.
Therefore, this leads to the conclusion that the decrease in Linux growth is moderating to a level near 5%, which as I’ve stated is still more than for Windows last year.
Why do you keep bringing up the same lies when I can easily prove them false everytime? Didn’t you know that truth trumps deceit? Your falsehoods are not suddenly become truer just because you repeat them over and over and over…
IDC had Linux at 3% in Jan. 2005.
And close to 300 million PC’s have been sold since then with Windows on them.
Do you have any numbers from the latter half of 2006?
Wrong. Linux tends to be found more multi-CPU servers than Windows.
Do you have any references?
The Linux server market share by revenue from Q3 2005 to Q3 2006 was 5.4%.
http://www.itjungle.com/breaking/bn112206-story01.html
For Q3 2004 to Q3 2005 it had grown by 34.3%.
http://www.idc.com/getdoc.jsp?containerId=prUS20013705
34.3% growth to 5.4% growth in 1 year.
OUCH!!!!!
COmparing the same quarters 1 year and 2 years is the proper methodology because different quarters have different buying patterns.)
Do you have any numbers from 2006?
Edited 2006-12-12 05:44
Gartner says that Linux growth is “moderating”…you probably do not even know what the term means.
Linux server growth is stabilizing at 5%, which is above Windows server growth.
As of today, Linux has a faster growth rate than Windows. All your juggling around with numbers you don’t understand will not change this very simple fact.
Gartner says that Linux growth is “moderating”…you probably do not even know what the term means.
I would say that dropping from 35% growth per year to 5.4% per year is more than “moderating”!!!!
Its disastrous!!!
It means market is actually shrinking from 12.2% in Q1 to 11.8% in Q3.
Thats negative growth in terms of market share.
Negative growth means it is dropping. Can you grasp that?
Linux share of server revenue dropped from Q1 to Q2 and from Q2 to Q3.
Gartner says..
Gartner also said that Microsoft will delay Vista once again:
“Back in September, the Gartner consulting group predicted that Windows Vista would be delayed until at least May of 2007.”
And then:
(From Gartner Research Vice President Brian Gammage’s Windows Vista blog):
“Following last Friday’s announcement that Vista is on schedule, it appears that Microsoft has made much faster progress than many (including Gartner) had expected in getting its next Windows OS ready for release. “
The thing is, since you claim that OneStat stats are inaccurate.. well, how do you know Gartner’s data is not?
And since you claim that OneStat’s data may be wrong, then you realize that it means that Linux’s desktop market share may be even lower than 0.36%?
Edited 2006-12-12 06:04
Ah, 3 on 1, that’s more like it!
The thing is, since you claim that OneStat stats are inaccurate.. well, how do you know Gartner’s data is not?
I don’t claim that OneStat’s data is inaccurate, just that you cannot use web stats to determine market share. Gartner’s and IDC’s data does *not* come from analyzing web access logs.
Note that NotParker has vouched for Gartner’s and IDC’s reliability before, so if you want to discuss this issue with him, be my guest.
And since you claim that OneStat’s data may be wrong, then you realize that it means that Linux’s desktop market share may be even lower than 0.36%?
Again, I’m not saying the data is wrong, just that it cannot be used to determine market share.
I don’t claim that OneStat’s data is inaccurate, just that you cannot use web stats to determine market share. Gartner’s and IDC’s data does *not* come from analyzing web access logs.
So where does Gartner’s data came from in this case, please? I mean, you have an example of Garter’s accuracy: in September they predicted Microsoft will delay Vista once again. A month later, Vista was ready. Hmmm?
Look, OneStat’s stats are the best that we have right now, so we should use it. Once you provide more reliable source, we can discuss that one.
Note that NotParker has vouched for Gartner’s and IDC’s reliability before, so if you want to discuss this issue with him, be my guest.
Yep, but it was that he vouched for them when their statements were based on real (sales $$$) numbers. That is fine with me because you have real data to rely on. Just like OneStat.
So where does Gartner’s data came from in this case, please?
I suggest you ask them yourself.
I mean, you have an example of Garter’s accuracy: in September they predicted Microsoft will delay Vista once again. A month later, Vista was ready. Hmmm?
That has nothing to do with market share research. It’s like saying the sports guy is lying when he says the Canadiens won 3-0 because the weatherguy said it’d be sunny and it’s raining.
Look, OneStat’s stats are the best that we have right now, so we should use it.
No, we shouldn’t, because web stats cannot be used to determine market share. I’ve already outlined the reasons why, many times.
That has nothing to do with market share research. It’s like saying the sports guy is lying when he says the Canadiens won 3-0 because the weatherguy said it’d be sunny and it’s raining.
Actually, I was giving you an example of Gartner’s faulty predictions when those are not based on real data.
Gartner is fine, if they make conclusions based on real data, like sales numbers (that NotParker is pointing to).
And that, just like Gartner predicting Vista delay, is exactly what you’re trying to do – you claim that OneStat stats are inaccurate, yet you have nothing to prove that claim.
So far, I have no reason not to believe OneStat. Sample is big enough.
Edited 2006-12-12 06:30
you claim that OneStat stats are inaccurate, yet you have nothing to prove that claim.
For the last freakin’ time I am not claiming the stats are inaccurate! Are you deliberately making an effort to misunderstand what I’m saying?
Let me repeat it one last time, after which I’ll be forced to recognize that you have severe reading problems: the stats may be accurate for what they’re measuring, but web stats simply cannot be used to determined how many individual machines/users visit a particular web site, and thus cannot be used to measure market share. Add to this sample bias and user agent spoofing, and you make the measurement even less adequate.
It’s not a matter of bad data, it’s a matter that this data cannot be used in such a way.
Read the links I provided.
So far, I have no reason not to believe OneStat. Sample is big enough.
It has nothing to do with the size of the sample, too. The OneStat survey is done among OneStat customers, not a random sample. If you know anything about statistics (which I’m beginning to doubt), you know that a random sample is essential to get reliable results. In fact, it’s the single most important factor, more than sample size. For example, you can take a national political poll of 20,000 people, but if they all come Red states, it’ll be less accurate than a poll of 1,000 people picked at random throughout the country.
All it takes to skew the results is for MS to be a customer (which they’d want to be, seeing as how they can then astroturf and throw these numbers around).
It has nothing to do with the size of the sample, too. The OneStat survey is done among OneStat customers, not a random sample.
They have 75,000 customers who use their service to measure traffic.
The people being counted for their service are random because anyone can connect to the web site being measured.
I’ve realized you know nothing about web analytics or are just being deliberately obtuse.
The people being counted for their service are random because anyone can connect to the web site being measured.
Anyone can connect, but not anyone does. Let’s say Tucows is one of their clients. They’ll get a disproportionate amount of Windows hits (not visitors) because no Linux users will ever go get software from Tucows.
I’ve realized you know nothing about web analytics or are just being deliberately obtuse.
I’ve realized you know nothing about web analytics or *you* are just being deliberately obtuse. Actually, in your case I believe it’s both.
Anyone can connect, but not anyone does.
But they can. Anyone can connect.
Let’s say Tucows is one of their clients. They’ll get a disproportionate amount of Windows hits (not visitors) because no Linux users will ever go get software from Tucows.
Ditto for Distrowatch. A very specialized site for Linux. And Windows users are still in the majority.
What you are fundamentally objecting to is the results.
As YOUR refence says: “properly compiled and knowledgeably interpreted, Web server log files can still provide some meaningful statistical indicators of Web site usage.”
http://www.osnews.com/permalink.php?news_id=16684&comment_id=191168
But they can. Anyone can connect.
They *can*, but most *won’t*! That’s the whole point!
Ditto for Distrowatch. A very specialized site for Linux. And Windows users are still in the majority.
Sure they are, but by a much smaller proportion that Windows’ market share of around 90%. That is exactly my point: if a majority of sites like distrowatch were OneStats’ clients, the usage share would me much higher for Linux (around 35%).
Thanks for helping me make my point: sample bias will make an inappropriate measure of market share (web stats) even less useful.
What you are fundamentally objecting to is the results.
No. I’m objecting to the use of web stats – ANY web stats – as a indicator of market share, and the facts support me.
As YOUR refence says: “properly compiled and knowledgeably interpreted, Web server log files can still provide some meaningful statistical indicators of Web site usage.”
Of course. But they can’t be used to determine market share at large.
The simple fact that web log analysis can’t even tell you the number of individual visitors to a web site completely invalidates the use of web stats as a measure of market share. No ifs, ands or buts.
Webstats are too inaccurate to determine market share.
Except when web estimates put Firefox higher than 10% of the browser market. Then they are great!
If the same web analytics firm puts Linux at .36% then it is inaccurate.
I personally suspect Firefox is at 1%, but Onestat says otherwise.
This is there list from August:
http://www.onestat.com/html/aboutus_pressbox46-operating-systems-ma…
1. Windows XP 86.80%
2. Windows 2000 6.09%
3. Windows 98 2.68%
4. Macintosh 2.32%
5. Windows ME 1.09%
6. Linux 0.36%
7. Windows NT 0.24%
8. Macintosh Power PC 0.15%
That must hurt to be beaten 3 to 1 by Windows ME.
Do you have any numbers more recent than estimates from January 2005?
Just in case you were wondering who OneStat is:
“OneStat.com is the number one provider of real-time web site analytics in the world. Our superior technology powers more than 75,000 websites in 100 countries. ”
Edited 2006-12-12 04:45
“OneStat.com is the number one provider of real-time web site analytics in the world. Our superior technology powers more than 75,000 websites in 100 countries. “
My point exactly. Nowhere do they claim that their stats will provide accurate market share numbers for OSes.
That’s becase they can’t.
Webstats mean nothing for OS market share. You constantly lying about it won’t change anything.
“OneStat.com is the number one provider of real-time web site analytics in the world. Our superior technology powers more than 75,000 websites in 100 countries. ”
My point exactly. Nowhere do they claim that their stats will provide accurate market share numbers for OSes.
“Methodology: A global usage share of xx percent for OS Y means that xx percent of the visitors of Internet users arrived at sites that are using one of OneStat.com’s services by using the particular number of OS Y. All numbers mentioned in the research are averages and all measurements are normalised to the GMT timezone. Research is based on a sample of 2 million visitors divided into 20,000 visitors of 100 countries each day.”
I think web analytics is the only method capable of determining which operating systems and browsers are surfing the web.
Either there are very few Linux users or Linux users are not using their PC’s to surf the web.
“Caching distorts raw data; audiences vary for each site; methodologies vary for each survey; surveys miss or omit important details; surveys mis-identify browsers or other user agents; small sample sizes exaggerate fluctuations; and stats don’t count those who stay away because their browsers are not supported.”
http://upsdell.com/BrowserNews/stat.htm
“4. What you can’t know.
[…]
ii. You can’t tell how many visitors you’ve had. You can guess by looking at the number of distinct hosts that have requested things from you. Indeed this is what many programs mean when they report “visitors”. But this is not always a good estimate for three reasons. First, if users get your pages from a local cache server, you will never know about it. Secondly, sometimes many users appear to connect from the same host: either users from the same company or ISP, or users using the same cache server. Finally, sometimes one user appears to connect from many different hosts. AOL now allocates users a different hostname for every request. So if your home page has 10 graphics on, and an AOL user visits it, most programs will count that as 11 different visitors!”
http://www.analog.cx/docs/webworks.html
“Statistics are no substitute for judgment” — Henry Clay
“Statistics: the mathematical theory of ignorance” — Morris Klein
“Get your facts first, and then you can distort them as much as you please: facts are stubborn, but statistics are more pliable” — Mark Twain
—
If only because of caching, it’s clear that web stats are not useful to determing OS market share. Note how OneStat calls it “usage share”, and not “market share.”
No one believes you here, NotParker. You were a troll on Digg, and you eventually left that site when you constantly got digged down. Same thing’s going to happen here. (Fortunately, there are no threads about Global Warming, because then you could all show us you are a GW Denier in addition to a Microsoft cheerleader…)
small sample sizes exaggerate fluctuations; and stats don’t count those who stay away because their browsers are not supported.
Please corroborate your claim that OneStat’s users are a “small sample size” and which unsupported browsers do you think are not counted.
You’ve posted that quote 100 times and it still seems downright silly to claim 3 huge web analytics firms month after month are a “small sample size”
1) 20,000 visitors in each of 100 countries is NOT a small sample size.
It is actually a very large sample size.
2) Onestats numbers are corroborated by other large web analytics firms sugch as TheCounter and NetApplications.
And calling someone a troll because they have stats from 2006 and not estimates from January of 2005 is awfully sad.
Edited 2006-12-12 05:57
The quote is not just about sample size, NotParker. Funny how you conveniently avoid addressing such issues as web caching, or web site audience bias, or user agents, or dynamic vs. static IPs…
It doesn’t matter if your stats are from 2006 or not – they are not the right stats! They are not market share stats, they are usage share stats, and even then their accuracy is dubious due to the elements I mentioned above.
The Jan 2005 figure is a *market share* stat, conducted by a firm you have described as reliable, using real market share research methods.
No matter how much you and your buddy tomcat try to spread the lie that web stats can be used to accurately depict market share, it still won’t make it true.
And I call you a troll because you troll, btw. Function defines form.
The quote is not just about sample size, NotParker.
But sample size is the key! Everything else in your quote comes from the asumption that the sample size is small.
But the ones I’m posting are HUGE!!!!
But sample size is the key! Everything else in your quote comes from the asumption that the sample size is small.
No, it doesn’t. It’s only one of many factors.
Web stats are not an accurate measure of market share, no matter how many times you repeat it. That’s why OneStat themselves do not claim that they accurately measure market share but “usage share”.
That’s why OneStat themselves do not claim that they accurately measure market share but “usage share”.
Are you saying you would accept the concept that only .36% of Onestats users use Linux?
Still a disaster for Linux on the desktop.
Are you saying you would accept the concept that only .36% of Onestats users use Linux?
No, I’m saying that web stats cannot accurately measure the number of individual users in the first place. They can count the number of hits, they can count the number of visits, they can count the number of individual IPs used to access the site.
They can’t take into account caching, they can’t take into account Dynamic vs. Static IPs, they can’t take into account User Agent Spoofing…I don’t even think they take into account that AOL surfers are grossly over-represented on web logs (AOL users are allocated a different hostname for each request, that means if there are ten images on a page, then an AOL surfer would appear as 11 different visitors).
They can give stats for these individual hits, but they can’t provide stats on individual users or machines. The Web just isn’t made that way (read up on it).
Also, we would have to see the list of OneStat customers to have an idea of the readership bias. If it’s mostly used by Windows-oriented sites, then the readership bias will be steep. And if Microsoft is among their customers, that would be enough to skew the results dramatically.
These are the reasons why web stats are *not* an accurate measure of market share. These reasons are supported by facts, as explained in the links I provided. I also suggest that you learn a thing or two about web access logs, so that you may understand this issue further and stop spreading disinformation.
You just cut and paste your replies. You know what you sound like? A broken record.
The funny thing about your exchanges with NotParker is that at least he provides data to back up his comments; whereas, you post half-assed conjecture and speculations. NotParker is right: Linux desktop market share is shrinking (not growing) because the overall market is growing and Windows is gaining more seats. Not surprising … but it does require a little mathematical ability to comprehend the dynamic.
Ohe funny thing about your exchanges with NotParker is that at least he provides data to back up his comments; whereas, you post half-assed conjecture and speculations.
You mustn’t have read many of our exchanges, tomcat, because you would have seen that I *have* given data to backup my comments, many times (and unlike NotParker, I do not misrepresent the data and play around word definitions). The problem is that whenever I come up with data, he dismisses the source off-hand, even when he’s used the same source himself.
NotParker is right: Linux desktop market share is shrinking (not growing) because the overall market is growing and Windows is gaining more seats.
Except you have no proof of that at all, just some web stats, which are *proven* to be ineffective in determining market share.
Not surprising … but it does require a little mathematical ability to comprehend the dynamic.
I can safely say that I know more about mathematics than both you and NotParker combined.
It’s sad that you have to tag team in order to debate me, but that I still run circles around you two.
I *have* given data to backup my comments
Estimates. From Janurary 2005.
My data is from this year. Q1, Q2 and Q3.
Linux’s share of server revenue dropped.
Edited 2006-12-12 06:01
Estimates. From Janurary 2005.
Actually, one figure was the market share they had measure, the second was an estimated projection.
BTW, we’re talking about Desktop here.
My data is from this year. Q1, Q2 and Q3.
Yeah, you see, you’re getting all excited and flustered now, you’re confusing servers and desktops.
In any case the OneStat figures for Desktop are irrelevant to market share.
Linux’s share of server revenue dropped.
By a negligible amount (less than 1%). That doesn’t mean that less units are sold, however, as Linux servers on the whole tend to be cheaper than MS’ overpriced products.
“Linux’s share of server revenue dropped.”
By a negligible amount (less than 1%).
Finally he actually reads what I’ve posted!!!!!
Finally!!!
DOWN! DOWN! DOWN!
And if servers is going down, and everyone admits that servers is the only place Linux has some hope, then desktop share had to have dropped like a STONE!!!
And thats why it is at .36%.
Do you have any desktop numbers from 2006?
Finally he actually reads what I’ve posted!!!!!
It’s too bad you don’t understand the numbers you post yourself. Linux revenue share going down (especially by such a minuscule number) doesn’t mean there are less Linux servers out there. In fact, without actual market share (not revenue share) numbers, we can’t claim either way. Since Linux server prices have gone down, and MS server prices have gone up, it’s even possible that market share might have gone *up* while revenue share has very slightly declined.
And if servers is going down, and everyone admits that servers is the only place Linux has some hope, then desktop share had to have dropped like a STONE!!!
Servers are not “going down”, the Linux revenue share has gone down (less than 1%), which is probably due to overpriced MS servers
And thats why it is at .36%.
Except it’s not. It’s a 3%.
Do you have any desktop numbers from 2006?
Do you? No, you don’t, all you have are some useless web stats that even OneStat won’t claim represent market share.
Linux revenue share going down (especially by such a minuscule number)
Actually it dropped from 12.2% to 11.8% which is .4% out of 12.2% which is 3.27% over 3 quarters.
Thats over 1% per quarter.
Not to be taken lightly!
doesn’t mean there are less Linux servers out there.
It means a smaller share of the pie. It means zero growth is approaching.
It means that Gartners predicitions from 2005 based on 35% growth no longer will come true.
It means Windows has an insurmountable lead for decades in servers.
Edited 2006-12-12 06:42
Actually it dropped from 12.2% to 11.8% which is .4% out of 12.2% which is 3.27% over 3 quarters.
Thats over 1% per quarter.
Not to be taken lightly!
One fact you’re missing. Those numbers are based on revenues.
Linux revenues were mainly based on mainframes and there is no real demand for those right now (or in future).
Do you know how much mainframes cost? Most of the latest revenues were based on x86 and x86_64, and by the time you figure out how many of those you can fit in price of one mainframe you’d maybe notice it wasn’t number you think that was going down.
It means a smaller share of the pie. It means zero growth is approaching.
Not really growth was the same if not even bigger.
It means that Gartners predicitions from 2005 based on 35% growth no longer will come true.
They still do, except not in term you’d like people to believe you.
It means Windows has an insurmountable lead for decades in servers.
Really? I always thought there is no best OS. It is only best choice for the job you need to employ. But if you meant numbers, I wouldn’t agree, I only see one thing, I didn’t saw linux servers at smaller companies few years ago, now I do and often more of them than Windows ones. I’d say this new visibility clearly shows that Linux is not dying but spreading. And there is one funny thing, you know which one is leading as most seen? CentOS. Meaning no revenue on Gartners numbers.
Excellent post.
Thanks to the stalker who continually mods me down. Get a life.
You’re a transparent hypocrite. Because on the one hand you call bullshit when Web statistics support products that you like (ie. Firefox) and then say they’re “inaccurate” when the statistics don’t support others that you like (ie. Linux). You only get to choose one delusion. Not both.
Web statistics are inaccurate. I wasn’t “calling bullshit” in the Firefox article – I was stating that the 11% figure thrown around by NutParker was not anymore valid that the 28% percent I found elsewhere. The point I was making, and which apparently went completely over your head, is that neither figure could be said to be more valid than the other, and that if NotParker was going to throw one stat around, I could very well do the same. However, contrary to what you’re tying to make me say, I never claimed that web stats were accurate in this particular case. To me, they were as inaccurate as they are to calculate OS market share.
Ergo, my position is the same both on the Firefox issue and the Linux issue. In both cases, web stats are inaccurate and cannot be used to determine market share. No hypocrisy, no inconsistency, no ambiguity.
Jeez, can anyone here pose a real challenge? I’m tired of these lightweights.
Web statistics are inaccurate.
They can be if you have a “small sample size”. Isn’t that what you posted 100 times?
But if the sample size is huge and ongoing and corroborated by other large sample sizes it can give you a very accurate pictures of what browsers and what operating systems are in use.
Sample size are king.
Onestat, NetApplications and TheCounter have huge sample sizes!.
Onestat .36%
http://www.onestat.com/html/aboutus_pressbox46-operating-systems-ma…
Netapplications .37%
http://marketshare.hitslink.com/report.aspx?qprid=2
TheCounter .3% (20 million hits)
http://www.thecounter.com/stats/2006/December/os.php
WOW. Those are HUGE sample sizes.
Web statistics are inaccurate.
No, they’re not. That’s an overgeneralization. The quality of the statistics depends entirely on the sample size and the methodology for collecting samples. The credible statistics-gatherers use millions of samples and gather them from a wide variety of sources. And, if you examine a bunch of them, a pattern emerges: Linux really does have less than 1% desktop share. I don’t see why you find that surprising. Practically no OEM sells desktop PCs with Linux preinstalled. Which means that the only ones running Linux are those who specifically reinstalled Linux over Windows. That market is absolutely less than 1% of all desktop users; nonetheless, given the size of the desktop market, that’s still a large number of machines. For example, 0.5% of 500M PCs is still 2.5M machines. Not chump change.
No, they’re not.
Yes they are. Read the two links I provided (for someone who complains I don’t give reference, you obviously haven’t checked out mine. One is from Statistics Canada, by the way). They clearly outline the limits of web stats, and make it clear one cannot use them to determine market share. Period.
Provide the links here.
Provide the links here.
Gladly.
http://www.collectionscanada.ca/9/1/p1-256-e.html
http://www.analog.cx/docs/webworks.html
…or, just talk to anyone who actually knows how the Web works.
archisteel provides the links:
Gladly.
http://www.collectionscanada.ca/9/1/p1-256-e.html
1998?
“What Can You Learn From a Log File? ”
“browsers and versions making the requests. ” and, from earlier where they list whats in a logfile: “the browser and operating system used by the requesting computer.”
Cool! You can find out the browser and the OS!!!!
However, properly compiled and knowledgeably interpreted, Web server log files can still provide some meaningful statistical indicators of Web site usage.
Yay!!! archiesteel capitulates.
Thanks.