“For Linux to reach the ordinary user, it has to offer more than good office suites and The Gimp and other free software implementations of common applications. Most people won’t make the move just so they can keep doing what they did before. Security and freedom mean a lot to a few of us, but they are not enough incentive for the vast range of Average Schmos. And we need those Average Schmos; the median is the message.” Read the editorial at OReillyNet.
To Push Desktop Linux, Radical Shift may be Required
2003-11-14 Linux 48 Comments
Apple seems to be doing fine business with their 4% of the market, and their community is large enough to support a decent amount of commercial development. I see no reason why Linux can’t just take the desireable 5% of the userbase and leave it at that. Its already well above 1% and growing fast, so I don’t see 5% to be out of the question.
“The Internet was another: Microsoft, AOL, and others had to really scramble to avoid being swept into the dustbin by it.”
Apart from anything else, the fact that AOL stands for “America ON LINE” should show how nonsensical this statement truly is.
re: Microsoft… I suppose that this *might* be an extraordinarily oblique reference to how the company was threatened by the possibilities the Netscape browser offered as a middleware software development platform. However, I think this may be giving the author too much credit, but even if not, that’s still hardly “the Internet”.
I still want to know those people who want Joe Iamdumb to use Linux? Frankly, I really don’t give a damn if the masses or the market can’t adjust to Linux. Their bad, not mine.
/me returns back to his project
Personally, I agree. I’m using Debian Unstable on my desktop with no complaints, so from my perspective at least, “need” is the wrong word.
However, more games for the platform, together with better hardware support from the vendors/manufacturers, would be niceties brought about by more widespread adoption.
Apple certainly doesn’t have a big enough market share for many game developers to provide an OS X port…
Thank you. That’s exactly what I think when I see comments like that. There’s already an Operating System specifically targeted at the average user. Much of what I like about Linux comes from so much of it being targeted at computer geeks. I like being able to tweak the kernel, have multiple window managers, and gui toolkits. If one wants to force Linux into being Windows, wouldn’t it just be easier to use Windows?
According to the anti-Linux zealots, we’re supposed to be cramming the idea that Linux not only will be ready to take over Windows, but already is.
What’s that? We’re not? We actually don’t care if Linux is easy enough for grandma? We’re not blind zealots?
I have to agree with Rayiner. Even if Linux can capture 5% of the desktop market that will be huge. Linux’s desktop market will continue to grow, jus not as fast as Windows. Apple is in the same dilemma.
I do think that Linux will do very well overseas and will be trouble for MS in the corporate desktop.
Why do we talk down to the general populas who have little experince in computers. Come on guys, seriously.
“Apple certainly doesn’t have a big enough market share for many game developers to provide an OS X port…”
Apple does have a smaller marketshare but also has plenty of mainstream games. Not the most but not lacking for all but the most hardcore gamer.
Apple does have a decent collection.Not that big but still good.
“Microsoft is not stuck in the past; they’re pulling as hard as they can to move their users to these upcoming innovations–trying to make them seem indispensable to staying competitive–because otherwise the company will have to stand by and watch the hose that gushes license fees gradually diminish to a trickle over the next couple years. No, Microsoft is pushing ahead. If any developers are stuck in the past, it’s the free software programmers diligently recreating what’s been done before.”
Exactly right. A huge percentage of the Linux crowd use Linux for 1 reason: it’s not Microsoft. Since that isn’t important enough for average everyday people to go out and buy a new computer, or even more difficult for most to go format their partition with an O/S they and their friends know little about, most will just stick with what works with most every available peripheral and software like games – Microsoft.
If we can get 90% of Asia to adopt Linux, then M$ is essentially screwed. Bad news for M$, Asia seems to adopting Linux faster than you Micro…
Linux is still young with the GUI projects even younger. Linux on the desktop would be a nice acheivement but I do not want corners cut just to get there.
Microsoft has to keep pushing features to make money. OSS projects can use the money but can live with being poor. That is the benifit of having access to the source code. Volunteer programmers will keep projects that are needed a live.
With the incremental improvement approach, Linux will come out a head in the long run. A lot can happen in the OSS community in three years. In ten or twenty years, who knows where linux will be.
Its not a matter of talking down. Its a matter of protecting my interests. I have absolutely no problem with various projects that aim to make Linux easier, more powerful, more usable, etc. Heck, I’d like some more usability work in KDE, to reduce clutter and whatnot. I’m all for projects like freedesktop.org and Fedora and whatnot. I think they’re great. I’m not opposed to new users on the platform either.
What I am opposed to is making Linux unusable for the very people that built it and currently form its core base. I’m talking about standardizing and hemogonizing everything. Think about what would happen if Linux was standardized right now. GNOME would become the default DE, Mozilla the default browser, and OpenOffice the default office suite. I don’t *want* to use any of these apps! I (and a million other KDE users) like KDE so much better. Think about what would happen if we mandated 3 years of binary compatibility. g++ bugs would go unfixed and innovation would grind to a halt. Think about what would happen if we got something like InstallShield. We’d have to give up the automagicness of APT for something intensely manual. And where would the freedom go? Would commercial interests take priority over the ideals of Open Source? I (and a lot of other Linux users) don’t see Linux as just an alternative for Windows. We see it as something fundementally different. Many of us would not be happy if Linux had 90% marketshare, but everything was still the exact same, with just the Windows kernel replaced by the Linux one.
So I’m not fundementally opposed to getting new users on the platform. I just don’t want to see it done in a way that destroys what Linux means to its existing users. Mac users can sympathize — would you want Apple to ditch its hardware and run on every $200 Dell?
“Apple seems to be doing fine business with their 4% of the market, and their community is large enough ”
Mac fans PAY to get Apple’s products (get ripped off ???)
zealots rip off everything, even the IE font pack ???
No wonder MS felt scared 😎
They install linux on new boxes to “show” that IPs are respected – then those boxes are stuffed with pirated version of windows, costing less than US $1 per CD vs the US$200 for the legit. version.
Some local software houses lowered their products to US$3 to 4 per copy hoping to fend off the counterfeit versions. Then counterfeiters RAISED their CD price to US$ 2.5 per copy, with exactly the same boxes, instruction manuals, etc. so even a trained experet in the business could hardly tell the difference, let alone an ordinary consumer.
One guy over there told me he once bought a legit copy of a software title, however the thing would not install. Then he got a pirated version, and everything worked fine.
When peached against a $1 version of Windows, the $0 trouble just isn’t worth all the effort on a $250 box.
The article was logical, he had some ideas. The author is going to hit on something because he understands some of the real concepts such as ‘putting the control into the hands of the users’ and he has created a model in order to understand the direction that Microsoft is going in, so that he can think independently of Microsoft, so that he can think out of the box. This guy is one to watch out for.
What I am opposed to is making Linux unusable for the very people that built it and currently form its core base. I’m talking about standardizing and hemogonizing everything. Think about what would happen if Linux was standardized right now. GNOME would become the default DE, Mozilla the default browser, and OpenOffice the default office suite. I don’t *want* to use any of these apps! I (and a million other KDE users) like KDE so much better.
As a Gnome user, my thoughts are the exact opposite. I always considered KDE to be the more popular desktop, and that if homogenization were to occur, I would have to ditch Gnome.
I guess it goes to show why there is sometimes a Gnome vs. KDE war afterall. Perhaps we are a little bit scared that we will no longer have a choice at some point.
I think KDE has a slight numerical edge, but the influential Linux companies (in the USA, anyway) like Sun, Novell, RedHat, etc, seem to be leaning towards GNOME.
This is my take on the Linux strategy.
The key for Linux to be successful is to have tools that make working with the source code easier, AND the focus of open source development must be on creating unspecialized applications. This is the key because than a Linux vendor can take a subset of the open source software and they can differentiate themselves by specializing it.
This is the strategy that I would use.
The people writing these columns are just so unbelievably arrogant.
They seem to think that the developers who created Linux and the massive profusion of other free/open software components somehow owe it to them to make a product that replaces the commercial offering they are stuck with.
It doesn’t work like that, Linux/GNU is about giving people the tools they need to build stuff, and very little else.
It is not, and may never be, a truly ‘consumer-oriented’ product.
People are free to build these offerings on top of Linux, using a plethora of free software components, but these 3rd party offerings should not be confused with what Linux and the other major infrastructure projects undertaken by he free and open source communities are.
Talk about how Ximian isn’t ready and needs to make a change, talk about how Suns Java Desktop isn’t ready and needs to make a change, talk about how Red Hat Advnaced Workstation isn’t working for people and needs refinement.
But just shut your goddamn mouth about Linux.
When you’re ready to to stop viewing your computer as a burden and instead as an opportunity, thats when Linux becomes attractive. Not ‘when the office suite has matured, not ‘when windows are alpha-blended, not ‘when Adobe ports photoshop’, but whenever *you* have something to offer the rest of the world.
As someone else mentioned, I think linux needs better development tools.
This is somewhat related to what the guy mentioned as the cost of writing/porting apps to linux.
Let’s look at the tools for windows developers and focus on the client-side of things. For the most part, windows developers are using Visual Studio. Noticed I said “client-side” programming – java developers have eclipse and IDEA, but most java development is focused on the server side of things.
You’ve got visual basic programmers that usually focus on the corporate RAD front-end app. You’ve got the hardcore c/c++ developers doing games, device drivers, shrink-wrapped software. Now we have the c# who will be coming in from both camps(In the future almost all windows apps will be done using .NET). You’ve got the ASP people too.
The commonality of all these people is that they’re using the same IDE with good documentation for the api, tons of examples, code completion, easy project setup, gui builders built-in, integrated debugging, object browsers, etc…
Now let’s look at linux developers and also on the client-side. First you have people using a myriad of toolkits – gtk+, qt, gnome, kde, along with the the various language bindings for these toolkits in their various levels of maturity and stability. These people are using vim, emacs, whatever. They are using autoconf,automake, makefiles and other build tools that are imo major headaches. I like vim and have used it for c/c++ work, but the problem is that in all it’s glory, vim is still basically a texteditor. Emacs is the same way. The facilities of these editors just don’t expose a proper plugin system for people to build a proper IDE.
Linux needs the killer IDE that has integrated support for some kind of RAD development that developers will embrace. Maybe the new kdevelop will be that(let’s hope so, it’s been a long time in the coming), but when I look at VS.NET and all of the convencies it gives me, I hope that someone will come up with the killer IDE for linux. Too bad that slickedit isn’t opened up.
Actually, AOL did nearly miss the internet. They still to this day cling to their old-school proprietary network protocol [even over DSL] and they “manage” internet pipes that go thru AOL in various ways that normal ISPs just don’t do. In the begining, AOL really did try to ignore the internet. It was a somewhat buggy “extra” that worked “sometimes” and often not as well as it should have. I routinely still have problems with simple connections to “fringe” web sites thru AOL that I don’t have with my work net thru a normal ISP.
AOL would LOVE to control the net…that’s why they made the deal with MS and “knifed the baby” with netscape….to regain the status Quo. We plug in, they tell us what we like. Isn’t that nice!
but he followed it up with CORPERATE USERS! The only use most MS advancements have for normal users is to teather you’re free information to some company running their servers as a service [or requirment for service!] MS has some great corperate tools, they are best at delivery, not innovation. MS takes the pieces out there and packages them to work best…that is a talent. But anymore, they are resting on their laurels with home users. Their plan is to get the companies using MS everything…so you HAVE to have extensions of that system into your home too! Note, they are about to get wiped by Apple and Sony. Those two have been shut out of corperate gravy and learned to deliver adaquate consumer experiences with lots of little devices that “play nice”. That’s the antithesis of the MS strategy…and MS can’t win.
The problem Linux faces is really the same one that MS faces. Computers haven’t become the “great and marvalous instruments of personal edification and exploration of our time” that they were made out to be. Until the web in the mid-ninties and file sharing more recently, computers were glorified terminals or typewriters, nothing more. I’ve said before, nobody has time for hours of home movies, or gigs of picture albums. Most of our drives are littered with photos stashed in email archives, and cmoputers have still not caught up with how we LIVE, so most of us still “live” with paper calanders, bills, checkbooks, etc…re-keying all that into a computer is nice…but who’s got the time for all that…let alone to manipulate PCs for “wonders”. for most users, if the computer was gone tommorrow, it wouldn’t really matter, sure there’s things we like, but nothing we NEED…computers are still TOYS for home users. Until operating and programming is MUCH easier they will stay that way. <p>
Also, MS has succeded in it’s “mission” to put a computer in every house! Pretty much everyone in North america that wants one, has one. Untill something really new is invented, It’s as good as it gets for at least another 5 years or so…So far it’s a big yawn.
I tell you what, give me Gnome and default Gnome applications and nothing else… Then hide the file system for me but show it with a terminal then install that Gnome stuff in /Applications for me, turn around and install just the other Gnome Extras in /System for me like on Mac OS X… Then give me a website as good as BeBits for me to find good software, look at screenshots and see what is the most popular and I’ll give you 100 dollars a year for upgrade just like I do Apple. I never know what is good to install… I get to much on my install, and I can care less about the history of the file system… It’s fine and all for administrative stuff, but I don’t care about it for my personal desktop.
Well it may be early, but lets see if Linux developers heed the warnngs of Longhorn. The mistake of geeks is that they imagine that with a little persuasion everyone can be brought to love their computers the same way they do. (Rather in a similar way to those people who like loud music actually imagine that everyone within a half mile radius enjoys listening to it too). It is an inward looking self delusion. They fail to see that what most people want is ease of use. They imagine that people will really feel better for knowing in great detail the initimate inner workings of their operatng systems… But the problem is quite simply that most people don’t give a single damn about this. It doesn’t matter if you think they should, because the plain fact is they don’t.
Added to this is the utter lack of focus and concentrated effort in the linux community, wich results in a circumstance where you tend to have 50 odd mediocre
text editors, 10 buggy browsers and 30 odd ways to do any other given task. And then only after jumping through a whole bunch of mind boggling puzzles and cryptic instructions can you ever get really anything useful done.
The Linux scene is currently simply ‘too fat’ to be efficient. What is needed is perhaps only 2 or 3 distribustions on which everyone can focus.
The guy above who compains of his loathing for the idea of ‘standardisation’ needs to realise that it was standardisation that made the industrial revolution possible. Prior to this there was no set standard for anything – and the world ambled along in the dak ages for some 500 years… It is standardisation too that has brough MS to its position of global desktop domination.
As I said, there is two years to go before we see Longhorn come to fruition. I can oly hope that Linux developers everywhere sieze this opportunity with both hands – and finally and conclusively provide some real competition to the absolute dominance of Microsoft based operating systems.
You may well loath standardsation, but it is what has made it possible for IT managers all over the world to track and trace faults and bugs – and has enabled them to build up a fairly standardised (that word again) knowledgebase that allows them without much effort to repair problems quickly and easily.
Hardcore Linux users and developers need to learn that they have nothing to fear from standardisation. Instead of the, what is it, 15 or so Window managers we have today, how about one really good effective one?
You shouldn’t worry too much about having applications as ‘default’ as this needen’t mean the only aplication there is. KOffice is already ‘default’ in KDE. It doesn’t mean you would have to use it.
This debate may well go on indefinately, indeed in the two years I have been visiting this site it almost feels like it has. But the Linux community needs t start listening, they need to take into account what their critics are saying, instead of covering their eyes and ears and shoulting about hwo everyone should come to appreciate intellectual the purity of computers.
If they do not, I fear like many other potentially great ideas, they and Linux (and all hope of a world free of the dominance of MS) may all one day just fade away…
Of course you’re exactly right and something that many linux users don’t want to hear.
Choice is good and all(people have different wants and needs), but just think if there was just one major gui toolkit, one or two major browsers(I guess now there really is only a couple), if distros would follow the LSB.
The real problem is that there are limited development resources. Some people think that since everything is “open” you will have millions of developers developing things that will somehow come together into a great thing. that’s not how it works in the real world.
Look at how the linux kernel is developed. Basically, you have a dictator, his lieutenants, regular developers, testers, etc. You don’t have major forks of the kernel. It basically works how you would have it in a commercial shop.
Look how fast Apple got their act together with Mac OS X. Just think if all the userland people had concentrated efforts into standards like that.
Linux isn’t going anywhere, millions of people are happy with it the way it is(the desktop is actually pretty good), but it could be so much better. Just think if there was only one major gui toolkit, one major desktop..
I could go on forever about this, but since i’m tired….
Average people don’t need Linux. The Red Hat guy was right all the time – Windows is much better for the average user. And MacOSX. Even Amiga was better in terms of usability.
Most of usability and integration between UNIXed DE projects like KDE,Gnome,WindowMaker,Blackbox, etc. just works inside it’s own DE. Copy & paste, menus, etc. .
Freedesktop.org has been making some very good efforts, and finally some X forks are being forged – about time!
Remember when people wanted a GPL’ed and POSIX’ed BeOS?
The Linux guys wanted to port the good stuff to Linux, the other guys wanted the critical mass (and apps) from Linux to come into BeOS.
So it’s not about the kernel. Linux best strenghts are called UNIX and GPL. GPL is a good thing for the average guy, POSIX and X are good for porting.
But about hardware installation, services, installing software packages, and desktop integration:
Linux? thanks, but no.
The biggest thing holding Linux back is the attitude of it’s proponents.
Computers used to be mainframes stuck in sanitized and air conditioned rooms that ordinary people couldn’t access. Companies like Microsoft ended that. Bill Gates etal were the ones hacking out a version Basic for the Altair from the apartment headquarters of MS. Stallman hasn’t brought anything to the ordinary people. Linus hasn’t either. IBM and Microsoft have. I bet if IBM had it to do over they wouldn’t have developed the PC either. They never thought it would be the success that it was.
I think some of the poeple here would like to return to the days of old. They could feel that much more special about themselves. Is this what Linux is really about? Some freedom.
Of course if you know you never will have more than 5 percent of the market, this type of attitude appears to be noble, doesn’t it?
>> “Most people won’t make the move just so they can keep doing what they did before.”
Oram’s quite correct. Individual users won’t switch to Linux unless Linux gives them a reason to switch. Go watch someone as they work with Windows. Is there anything they want to do that they can’t do? (Not “do more easily”m but “I need to do this and I can’t on Windows.”)
For most people, Windows defines their expectations of how a computer behaves and what it will do for them. Get used to it, and move on.
Doing the same things Windows does isn’t much of an incentive to switch. Linux needs to offer something compelling that Microsoft doesn’t, and won’t, offer. Whatever that may be, it certainly isn’t more Office clones or more Photoshop clones. Don’t confuse the corporate desktop market with the individual user market. The latter is easier to crack.
I’ll also suggest that Linux needs to become much more visible in the commercial arena, i.e., where people shop. Even if Irene Individual decides to take a chance with Linux, where is she going to buy it? And, no, downloading a free ISO and burning a CD isn’t going to happen for Irene. Odds are, she doesn’t know how to do that, and doesn’t want to learn. Besides, burning a bootable Linux CD in Windows can be problematic. What Irene wants is a $19.95 CD in a box with a book. And she wants to buy it where she shops, which isn’t at CompUSA or Fry’s.
Finally, adolescent ideological posturing and general snobbishness about “do we want those kinds of people?” is an insult to anyone considering using Linux and serves only to color the entire community as simply a cult of intolerant boys. That raod lieds to irrelevancy.
>> “Linux is still young…”
No, it isn’t. Linux is more than 10 years old. Considering that it s clone of an existing OS and has never developed it’s own graphical display capabilities, one could argue that it really hasn’t progressed very far in all those years. Compare the parallel with OS X: Both a “clones” of Unix, yet OS X leapfrogged Linux in terms of usability and desktop polish in a few years. Why? Because the Linux development model lacks the ability to put focused and sustained effort into that kind of project.
>>Microsoft has to keep pushing features to make money…”
Well, of course. Like any business, they exist to make money. That’s an incentive for MS to market products they believe people will want to buy. The Linux community can’t even make up its mind about whether or not developers should pay attention to what people want to do with their computers. Take a look at the number of snobby posts here insulting non-Linux users as “dumb” and “lame”. It’s “OS as ego-boost” for these folks.)
>> In ten or twenty years, who knows where linux will be.
At this rate, still chasing Windows’ tail.
“[Linux] has never developed it’s own graphical display capabilities”
And quite rightly so. This doesn’t belong in the kernel.
“one could argue that it really hasn’t progressed very far in all those years.”
Lol! Why don’t you download one of Linus’s first uploads, and then compare that to the latest 2.6 test kernel. I’m fairly confident that you’ll be able to spot considerable improvement.
“At this rate, still chasing Windows’ tail.”
I’m using it on the desktop quite happily already. It’s a far superior platform to Windows *now*, from my POV. And, since I don’t foresee anything happening that could cause either the user or develop base to diminish, only grow, I’d say the future looks exceedingly bright.
I’m with Syntaxis on this one. The kernel is already big enough as it is. Imagine putting in all the graphical stuff on top of that and it would be humongous! I prefer having the components separate; it makes things more responsive.
“one could argue that it really hasn’t progressed very far in all those years.”
what are you smoking? And, no, I don’t want any. I tried to get into using Linux about three or four years ago during my WinME days (for obvious reasons, and it came preinstalled before anyone asks), but quickly ditched it because of lack of games, hardware support, etc. About 8 months ago, after a fried HD and burner (actually cause by windows, no joke), i switched to RH9 and was extremely impressed by how things had changed. In three more years (i.e. around when Longhorn will probably be released), things will have improved even further. There is simply no question. Why else would so many companies, goverments, etc. be switching over? Sure cost is one factor, but would you use cheap software if it wasn’t useful or stable just to save a few bucks? Probably not.
What there needs to be, however, it not a standardisation to the extreme where by people are told, that xzy MUST be the default desktop. Those sorts of issues can be resolved by the market place through alliances. The most popular desktop at the moment is KDE followed by GNOME. SuSE and TurboLinux are KDE focused distros where as Red Hat is GNOME focused. If an agreed library base can be agreed on, then the compatibility issues should stop occuring.
For example, the LSB should expand out to including what version of XFree86, KDE and GNOME, RPM and compiler. If LSD 1.4 includes those sorts of things then developers can say, “Any distribution that is LSB 1.4 compliant”. If there is a big honkey sticker on the box saying, “LSB 1.4 COMPATIBLE!”, then the user can put two and two together.
If the user can double click on the RPM, it installs and the icons turns up on the menu like one would expect, the end user quite frankly doesn’t give a tinkers cuss about whether it uses QT or GNOME, what the end user does care about is that their DooDacky software works as it should. If the application comforms to LSB 1.4 and the distribution conforms to it, then it should work.
There are too many people worried about the intricacies of the desktop when in actual fact what is holding Linux adoption back is a standard base, hardware support, easy to install routine and a good network of ISV’s. People are now making their own websites so they want to be able to run things like Dreamweaver Lite (the low cost one release a couple of months ago), or Photo tweakers like Photoshop Elements. The user doesn’t care that they can *TRY* and learn how to use GIMP, they want what is out there on the shelf, and they want technical support and all the other palaver that comes with a boxed product.
The day Linux will be ready as a desktop solution is when people can buy a computer installed with Linux, buy the distribution on the shelf and we see the numerous vendors out there like MYOB, Quicken, Adobe, Macromedia, and the numerous games that are available. Double click and install, that is what the consumer wants. Whether you like it or not, the end user doesn’t care or want to know about the intricate details of their computer. The end user simply wants to turn it on, use it to acomplish a task and turn it off.
Many people here don’t get it because they’re not the target audience. The people who read this site don’t represent the vast majority of users, the vast majority of users don’t sit on the edge of their seat wondering about the next IT advancement or what operating system Microsoft will release next. The end user has very little to do with IT, and they’re happy with that.
What WE should be doing is making it EASIER for them. If WE as a community can offer a BETTER way of doing something with LESS effort, then they’ll have a look. Doing something better doesn’t mean doing away with anti-virus’s, because the vast majority see that as simply a fact of life, what people want are applications that don’t have too many features and that the features that are included actually work as documented in the help file rather than having to find a work around because that piece of funcationality is broken. What users also want is the ability to know what is happening if something goes wrong, but in PLAIN ENGLISH, not techno-garble complaining about some port not being intialised by zya memory address and so forth.
If an application crashes, display a message, “[Application Name] has crashed. This could be due to a bug (hyperlink to defintion of bug). Possible suggestions: [list of suggestions to stop it occuring]”. If the operating system crashes with a BSOD or kernel panic, don’t just simply give a memory dump. If if nvidia.sys crashes, then say, “The display driver has crashed…” and continue along the same lines as the application dialogue box.
I think put succinctly, what most people are saying is ‘enough already’. Enough already with the mind boggling variety of applications that all do the same thing relatively poorly; enough with the Linux isn’t for the masses, ‘we are too l33t, why should we develop for the masses’ attitude; enough of the paranoia that makes peeps think that establishing a set of coherant standards everyone can work towards is somehow a bad thing…
The writing is on the wall, not least so because of the potential demise of two of the biggest desktop players, Red Hat and Suse. (I can’t imagine Novel having any real long term plans for Linux on the desktop). If Desktop Linux is to survive it must embrace some fairly fundamental changes.
But as there is no ‘organisation’ as such – and no top down structure – I wonder if this will ever happen? If not I have no doubt Linux will continue to amble along in the minor operating systems league for much of the forseeable future – and will only ever be capable of at best a form of 3rd rate mimicry. Innovtion happens through concerted effort and organistion, which are two concepts that Linux currently lacks…
Put another way, what would you rather listen to? 50 people all in the same room singing a diferent song, often to a different tune? Or a coherant chorus of peple all singing the same song and more importantly, singing from the same song sheet?
Which analogy do you feel fits Linux best at this point in time?
Put another way, what would you rather listen to? 50 people all in the same room singing a diferent song, often to a different tune? Or a coherant chorus of peple all singing the same song and more importantly, singing from the same song sheet?
Why not have African “call and response” style of singing with 1/2 calling and the other half responding? 😉 (yes, I know, now I am spliting hairs)
Who will bring the leadership? the people who bring the leadership are the distributors NOT the coders. The coders are there to produce code, it is up to the distributors to provide direction, create what they need and weld different projects into a coherient product. That is the opensource smorgasbord model in action. If the additions they make are seen as value to the code, the opensource coders will look and expand on that good idea.
When you purchase a distribution, not only are you paying for the support but the time spent integrating the applications into a distribution to form a coherient product.
Too many people here think that the leadership will come from programmers. Programmers are the LAST people one should expect to provide leadership. Sure, there are some rare exceptions, however, what you need is the ability for a core number of people to layout a map and work towards it.
For example, GTK needs to be compared with Windows and MacOS; what features are missing? what are required? when should these features be added? there needs to be a strong development path that needs to be stuck to. If third parties see that there is a development line, then they’ll be assured that the library they’re basing their product on will be continued to be developed in the future along a certain line.
You could have Mongolian throat singing too.. Unfortunately it is still going to remain rather obscure – and I doubt it will ever make it anywhere near the top spot in the hit parade.
Mmmm, well on consideration you made some good points. I’m just being fecicious, sorry. But yes you need that level of organisation, perhaps a group of professional interested parties to band together, to provide the leadershp and direction that the development community is so plainly lacking.
You are right in that for too long it is developers who have been left to guide the direction in which Linux went. The result is, perhaps naturally that they went off in 500 directions all at the same time, with no one around to tell them what was a good idea and what wasn’t.
To my mind the kernel development model is a good model for the rest of Linux’s development – and I feel sure that if this model were universally adopted, with different branches of the model assigned to different tasks and different applications, you could have exactly the kind of organised structure that allows companies to progess – and indeed to become ucessful. Moreover you could have planning, and interdepartmental discusssions and brain storming sessions. What an innovation that would be!
Once more rather that xyz number of development models, perhaps if everything were placed under one big banner in a place where everyone could really start talking to each other, we might actually get somewhere?
But debates like this have traditionally been like pissing in the wind. They go on week after week, month after month – and ultimately no real progress is ever made. What seems like progress can from another perspective very often seem like the whole effort is falling behind, lagging in a quagmire of introspective geek self justification.
Like all movements the open source community needs leaders – and without them we – and the Linux cause in general will undoubtedly become a lost cause.
CooCooCaChoo is correct about the standards that should be used. It is better to have a GUI that follows standards than to have a standard GUI. Freedesktop.org is a very good start. If all window managers and GUIs follow standards then all programs running in those GUIs should behave the same.
LSB and FHS are other good examples of where standards should be defined. Those distributions that follow the LSB and FHS will behave in a similar fashion.
Some people have said that linux should be syncronized singers, but may be linux is already there. The song could be considered the OSS movement and all the different singers could be the different projects within the OSS movement. Each singer is different in size, shape, and color. But put together they sing one song called the OSS movement.
The Register has a great article about the Thai government’s initiative to provide low cost Linux PC’s to first-time users. Support is provided through local Linux users groups: http://www.theregister.co.uk/content/4/33963.html
This is a great model for both community-building and free software advocacy.
linux does not exist to combat microsoft. linux exists because coders want to make good software.
this is all you ever hear from people coming over from windows, linux needs this, linux needs that. the truth of the matter is, linux needs nothing. YOU need this and that. the way linux works is that if you want it, then you make it. linux doesnt have a unified list of goal, people make projects for linux to take it in the directions THEY want.
standardization is great, but not standardization because of arbitrary choices. if something is a category killer (significantly better then everything else in the category) then it quickly becomes the standard. i like gnome as a dm, you like kde, who decides what the “standard” is?
LSB is good to a point, but to think it will result in a unified linux is redicules, and it isnt the point. we need standard directory structure, standard way of doing things, not a standard office suite, or standard window managers.
if you want something in linux, then make it. if you cant make it, learn. if you cant learn someone, hire people. because that is the only way you have a right to say where linux is going, not by giving 60$ to redhat, that just gives you the right to talk about where redhat is going, not linux as a whole.
so just please, please, please stop complaining about linux as a desktop os. unless you are a contributer, you have no right to complain/demand/threaten/insult anything. it is a slap in the face to all the guys who work their asses off for free, which you and everyone else in the world benefits from.
“A huge percentage of the Linux crowd use Linux for 1 reason: it’s not Microsoft.”
So what? It’s the same thing for Apple users. People like to be different because it makes them feel special. But I’d also wager they rather enjoy the other benefits of their systems, be it “free access to the code and internals” (Linux) or “looks pretty and easy to use” (MacOS).
“Hardcore Linux users and developers need to learn that they have nothing to fear from standardisation. Instead of the, what is it, 15 or so Window managers we have today, how about one really good effective one?”
There is no such thing. There are fundemental differences that cannot be reconciled. You either like one way of doing it, or you don’t. What you’re saying is analagous to saying “instead of 15 or so car manufacturers we have today, how about one really good effective one?” Its ridiculous. Even if you have everyone a Porsche, a lot would complain because they like BMW’s ride better. What we need is open standards (ala Freedesktop.org) so it doesn’t matter which window manager you choose — you can still ride on the same roads.
@Roy Batty: There are projects (Anjuta, KDevelop, Eclipse) designed to make a graphical IDE. KDevelop 3.x and Eclipse are particularly far along, and would please most of the Visual Studio using crowd. But stop brining vi and emacs into this. Those aren’t competing with Visual Studio. They have a fundementally different paradigm. Don’t criticize it just because you don’t understand it.
Once again, you didn’t read and understand my post. In fact, what I said was basically what you said. Vim and Emacs are basically text editors and because of the design of these programs aren’t really suitable for building an IDE around them. By the way, I’ve been using vim for years and do understand it completely, probably better than you.
As far as “projects” go, that doesn’t mean squat. Bobby teenager can setup a project on sourceforge. That doesn’t mean it’s any good. Kdevelop has been around for a long time and even the core kdevelop developers would use emacs because of it’s deficiences. Let’s hope that with the new Gideon coming out they’ve made some real progress after all these years. As far as Anjuta, it’s having it’s fair share of problems with the whole Anjuta2 and various forks.
Raynier, it doesn’t really count if these projects are in a continual beta state no matter how much of an open source zealot you are.
I am ready to be able to order pcs with linux on them.
@Matt… I don’t think people have missed the point, I think you have missed the point. indeed it is people like you who advocate a linux dictatorship thar seriously piss me off. Of course users have a valid input; if they didn’t what are we to do? Leave everything to the developers and voluntrary gag ourselves from ever commenting? Even Microsoft goes to great pains to listen to its users. It is utterly foolhardy not to.
I don’t think he analogy with cars and operating systems holds very well. At a root fundamental level even if you open a car (any car) you will find that the building blocks that is consists of ARE made of standardised components – or at the least they are products of standardised components.
That is all anyone is arguing for, which is a more coherant development model that allows the adoption of generalised standards – and perhaps an underlying common common code base from which all other positve developments could emerge. (The emphasis here being on common).
But it is clear you and your betherin do not want this – and so I expect my prediction will hold – which is that this ‘pigs ear’of a development model that currently exists, with no standards and no real means of ensuring, or indeed even of tracking system wide quality control, that one day not too far off, Linux very will fade into obscurity.
What you are advocatig is an anarchic/communistic development model – which, as we have seen in the real world, can only lead to disaster.
But if you are happy in your little ‘see no evil, hear no evil, speak no evil’ totaliterian Linux ideal, then fair enough. Personally I long for a viable alternative to MS. But I do not see Linux getting there in any hurry. Certainly not as things stand.
Novell, Sun and Red Hat and standardize all they want for Corporate customers. Meanwhile for home users that don’t want to be buried in choices, they can just use Lindows/ Lycoris and Xandros. For the geeks there will always be Slackware, Gentoo, Debian and Mandrake. Isn’t that the REAL strength of Linux….it’s flexibility? So you can just give the corporate shmucks there desktop and their StarOffice/OpenOffice/Koffice and not restrict the freedom of hardcore Linux people. I hope to God they are not going to allow the corporate to define what Linux should be…