The Open Source Development Labs has voiced cautious optimism that its latest initiative could finally herald a mass-market for Linux on the desktop. OSDL is pushing ahead with Project Portland, to develop a common set of core technical requirements for Linux and open source software on the desktop, following a meeting of 47 companies and organizations it hosted earlier this month. Portland has identified a core set of areas, spanning the interface, plug-and-play, drivers and the kernel, that OSDL members will flesh out.
Firefox will be to FOSS what iPod has been to Apple?
Good luck to them
Really, I just want Desktop Linux to finally happen…. not to Destroy Windows but help bring back competition to the industry.
http://tieguy.org/blog/index.cgi/523
The only thing he failed to mention is that in order to make either KDE or Gnome better, it’s not just a matter of making KDE or Gnome better, but also creating an *underlying infrastructure* (kernel, daemons, drivers etc) that work better than they do today and work FOR the user. No matter if you make Gnome/KDE work perfectly, if Linus continues breaking binary and source drivers with each minor Linux kernel release, or he doesn’t fix ACPI or unify all Linux distros under the same package/binary compatibility, it’s all going to waste.
Normal users want things that work, not things that don’t. The the “things that work” usually have their root to the underlying system and its compatibility/capabilities and not on just the desktop environment which is simply a carrier for these underlying capabilities.
ACPI works fine. Your laptop/desktop is shit. Don’t blame
Linus that your computer is shit. Buy IBM/lenovo acpi works fine with them.
>ACPI works fine.
ACPI does NOT work fine.
Only SOME of IBM’s laptops work with suspend-to-RAM and that’s because the Intel engineers use them (who are largely responsible for ACPI on Linux).
All the rest of the laptops out there pretty much don’t work. I have 3 laptops. None of them work with suspend-to-RAM on Linux, and no, I don’t want to buy an IBM/Lenovo one.
The x86 market is a vast one, if Linux can’t support the big bulk of it, maybe they should only support IBM machines. Most people don’t use IBM laptops, mostly businesses do. The rest are usually buying HP or Dell.
ACPI works fine. The problem is hardware deviating from specs, making it harder to support it.
You know perfectly well that Linux supports the major bulk of the x86 market and even several other platforms.
No other system has such a wide platform support as Linux. But then again, Eugenia. You like XP, so your thoughts on this hardly counts by any metric.
ACPI works fine.
It does NOT work fine. Eugenia has THREE laptops and on NEITHER of them does Linux’s ACPI work fine. Hell, it’s almost common knowledge that Linux’s ACPI kind of SUCKS ASS.
The problem is hardware deviating from specs, making it harder to support it.
Oh really? Why, I never realized that… To the normal end user, it doesn’t matter $%#Q$^ if the hardware deviates from the spec or not. If it doesn’t work, it does NOT work. As simple as that. If Windows can support all those deviating specs, than why can’t Linux?
You know perfectly well that Linux supports the major bulk of the x86 market…
Yup, I do, but that doesn’t have anything to do with the fact that Linux’s ACPI support for x86 sucks.
…and even several other platforms.
Windows XP is probably just as portable as Linux. In case you didn’t know, Windows NT ran on x86, SPARC, PPC and Alpha. But that has nothing to do with Linux’s ACPI either.
But then again, Eugenia. You like XP, so your thoughts on this hardly counts by any metric.
I do not like XP. Yet, I agree with Eugenia. Your point?
Thom, Thom. Why don’t you read up before you yell at people?
If Windows can support all those deviating specs, than why can’t Linux?
Here is the secret: Windows does NOT support all those deviating specs. It’s the laptops that support windows broken implementation of ACPI.
Say after me. Hardware is designed for Windows ($VERSION), Windows is NOT designed for $HARDWARE. Buy hardware that is supported in Linux if you want working system. Don’t whine because your “Desiged for Windows XP” system doesn’t work with MacOS X or Linux..
Just keeping believing that. Try really, really hard.
Windows does NOT support all those deviating specs. It’s the laptops that support windows broken implementation of ACPI.
Correction. Windows has a crappy implementation of ACPI that is broken even more by many laptops, desktops and motherboards and this is completely masked by the drivers which are shoe horned and panel beaten into shape. That’s also the only reason why ACPI in Linux works on many machines – hacking the individual driver implementations. The only difference is more widespread testing and manufacturer help.
Edited 2005-12-15 22:23
Agreed. I should probably have written “the OEMs” in stead of “the laptops”.
Hell, it’s almost common knowledge that Linux’s ACPI kind of SUCKS ASS.
Everybody’s ACPI support sucks ass. Your point is?
To the normal end user, it doesn’t matter $%#Q$^ if the hardware deviates from the spec or not. If it doesn’t work, it does NOT work. As simple as that.
In terms of getting a solution to an end user it bloody well does matter. ACPI is by no means a standard, and the only reason why Windows works is because hardware Microsoft and hardware manufacturers have hacked it into place. I can give you a list of laptops and even desktops where suspend to RAM and hibernate does not work with Windows, even after all these years and given the fact that Windows is the first thing these people should be testing.
The solution is clear – hardware manufacturers have to really help out and actually test their hardware with Linux, and the OSDL and Linux developers need to encourage them to actually create an actual standard that’s predictable and that works.
“The solution is clear – hardware manufacturers have to really help out and actually test their hardware with Linux, and the OSDL and Linux developers need to encourage them to actually create an actual standard that’s predictable and that works.”
When the desktop Linux standards are there, the hardware manufacturers, as well as software companies, will come (or at least be more cooperative).
The problem now is that there is too much inconsistency across Linux distros, as well as a too rapidly changing Kernel, as well as a too fragmented Linux market, to make it monetarily worth while for hardware manufacturers (and software developers/companies) to target/test Linux.
The standards that will eventually come out of the Portland project will hopefully change all that. The standards will make it so that hardware manufactuers would not have to have multiple code bases (for drivers) or testing platforms, thus making targeting Linux easier. And a consistent file structure and packaging system will make it easier for software devs/companies to target Linux. And with these kind of standards, Linux distro fragmentation becomes completely irrelevent.
“The problem now is that there is too much inconsistency across Linux distros, as well as a too rapidly changing Kernel, as well as a too fragmented Linux market, to make it monetarily worth while for hardware manufacturers (and software developers/companies) to target/test Linux.”
i just wonder:
if linux keeps changing too much, doesn’t accept a closed source binary driver model, has too much fragmentation, why there’re still commercial distros using it? why do they use freebsd instead? i see that many of this claims are resolved in freebsd. It’s only takes to repackage a GNOME or KDE on top and some drivers and make a new OS to conquer the market… (even, they can use some of the linux hype and say that it’s a kind of linux, because in front of normal users differences between linux and freebsd kernels are almost zero).
the only thing lacking is 20 or more years of marketing and commercial experience
My best guess is the license.
It is true that Intel is doing its best to provide the ACPI implementation for Linux, and I’m sure their doing really great at implementing what they consider to be the ACPI standard.
Fact is, MANY laptop manufacturers claim to support the ACPI standard. Now, if memory serves me well, Intel designed the ACPI standard. Please explain to me, how it is possible that anyone can call their implementation ‘ACPI compatible’ if it clearly is not? In the real world, whoever does such a thing gets sued.
For example, you can’t just create a new programming language and call it Java. Or .NET. Two major companies will try to kill you, if you do. Hell, even if you create something new and call it Perl or Python, you won’t live an easy life.
Basically, from a users point of view, it appears that Intel does not give two shits about ACPI. They don’t care if anyone deviates from the standard and still calls it ACPI.
Personally, I hope to see an ndiswrapper for ACPI one day. I seriously don’t care about a native ACPI implementation, I just want the damn thing to work.
Please, give me Windows ACPI for Linux.
From a Linux Journal article:
ACPI, which stands for advanced configuration and power interface, is a power management component eventually set to replace APM. The current state of ACPI as a whole is marred at best. Laptop manufacturers often deviate from following any type of standard when coding certain ACPI-related BIOSes and components. Rather than follow a standard, they rely on proprietary Windows drivers to keep things functioning. Unfortunately, in many cases, this leaves Linux users out in the dark.
http://www.linuxjournal.com/article/6891
This answers your question about why linux can’t support all those deviating specs. If you don’t have the specs, how can you support them? If the manufacturers alter the spec, and then provide a binary driver to one OS provider only, then it’s pretty hard to support that spec isn’t it.
Sometimes you make a complete arse of yourself on your own site.
This answers your question about why linux can’t support all those deviating specs. If you don’t have the specs, how can you support them? If the manufacturers alter the spec, and then provide a binary driver to one OS provider only, then it’s pretty hard to support that spec isn’t it.
Sometimes you make a complete arse of yourself on your own site.
Yeah, all nice, no specs, etc. But what is the end result? Exactly– ACPI doesn’t work as well in Linux as it does in Windows. Period.
I was contesting the fact that “ACPI in Linux works well.” Because, as you just have proven, it does NOT.
I was contesting the fact that “ACPI in Linux works well.”
No you weren’t. You were asking why linux doesn’t support it if windows does. I was just pointing out why, and pointing out that windows doesn’t support the broken implementations either, the manufacturers do through binary drivers.
Unfortunately, you can’t spin your comments when there available for public viewing and scrutiny.
It does NOT work fine. Eugenia has THREE laptops and on NEITHER of them does Linux’s ACPI work fine. Hell, it’s almost common knowledge that Linux’s ACPI kind of SUCKS ASS.
I’ve gotta take the high-road on this one. You can’t say Linux’s ACPI implementation sucks, because it doesn’t. Linux’s implementation is based Intel’s, and Intel wrote the fricking spec. What sucks is Microsoft’s implementation. It just happens that most hardware has broken tables designed to be compatible with Microsoft’s sucky implementation.
You can say “the user doesn’t care who is at fault”, and you’d be write, but at the same time, I’d argue it’s a mistake to think that Linux should break its implemtnation to be compatible with Microsoft’s. Standards are standards, and without standards we are nothing but savages. Compromising standards for compatibility is exactly what made web-browsers as ass-tastic as they are today.
Edited 2005-12-15 22:13
I know nothing of this issue, so could you provide some reference material so I could read up on it myself? And linking a an open-source advocate site or a paul thurrott article doesnt count.
Neither does linking to a pro-Microsoft site, nor does arguing that ACPI works fine on Windows when it doesn’t.
Fact is:
ACPI is broken in itself – especially on laptops – since hardware vendors deviate from the specs and only works with a broken implementation from Microsoft.
When Eugenia says ACPI doesn’t work well on Linux she is wrong. The truth is ACPI doesn’t work well on laptops.
ACPI works fine on all stationary PC’s I’ve used. And not all laptops works fine with Windows.
The problem is the laptops, not the OS.
Um.. hence why I said not to link Paul Thurrott. Sometimes I wonder if you read my full posts :/
Again, I don’t know much about this. But someone before said that Microsoft doesn’t have a broken implementation, but that hardware manf. support a broken implementation with binary drivers. I don’t know if it’s true, but it’s something to at least consider.
Oh sorry, didn’t see the Paul Thurrot thing. I usually avoid his name.
You’re right. It’s worth considering if even MS has a problem or if it can be isolated to hardware manufacturers solely. In case it can be isolated to them, it’s about time we put heavy pressure on them. This would also benefit the Windows platform, since no all laptops works well even with Windows (at least not all modern versions of Windows).
Oh really? Why, I never realized that… To the normal end user, it doesn’t matter $%#Q$^ if the hardware deviates from the spec or not. If it doesn’t work, it does NOT work. As simple as that. If Windows can support all those deviating specs, than why can’t Linux?
Eugenia is no ordinary end user and she knows very well, that ACPI in linux works fine. The problem is that ACPI does NOT work fine on laptops.
She is badmouthing the wrong part in the problem. The problem is Microsoft raping standards once again and certain hardware vendors do the same.
That’s why ACPI doesn’t work well on laptops. Linux is doing fine, it’s doing the right thing. The hardware vendors are doing wrong – and so is Microsoft.
Focus on the real problem, instead of blaming the innocent one.
BTW: Windows XP is not exactly portable. It might be, but we don’t know since it only runs on x86 hardware. I do know it ran on a few platforms in earlier days. But that’s quite some years ago by now.
Of course you agree with Eugenia. I’d expect no less. And I’d think less of you if you didn’t give her backup.
linux’s acpi support does not work fine for a lot of people… in fact, it’s one of the reasons i’m using freebsd on my laptop. it just works. this is one thing linux really needs to work on. end users don’t care if an operating system can run on a game console or pda or somebody’s used gum that they stuck on the bottom of a chair somewhere. end users care that things work, and work well.
Like when you enter on a FreeBSD system : shutdown -h now
And instead of shutting down it write : You can now safely turn off your computer!
for me, that doesn’t work either.
that’s what shutdown -h is supposed to do. if you want it to turn off you’re supposed to use -p.
from the shutdown(8) man page:
-h The system is halted at the specified time
-p The system is halted and the power is turned off (hardware support required) at the specified time
“All the rest of the laptops out there pretty much don’t work.”
I guess it’s time for the hardware guys to stop making useless and undocumented implementations, eh?
Yeah, that pretty much the philosophy behind the Linux fanboy. ‘If it doesn’t work it’s their problem not ours’.
Well, hardly anyone gives a shit about whos problem it is, if it doesn’t work, it doesn’t work. What good can a 100% W3C compliant browser do for me, if it can’t render correctly 90% of sites, because they’re all broken?
Time to point out the old Linux Fault Threshold(LFT).
http://www.adequacy.org/public/stories/2001.10.2.33542.4010.html
Maby I’m just lucky non of the sites I use prakes in FF (Osnews , slashdot, Shandiabankn.no (online banking) and several others. And yes it is the wbsites fault, thera are tools out thehre (sorry don’t remember the name, will post replay with aditional info) that creates w3c html etc. But that code breaks in IE, whos fault is that? Not to bash MS but can we reealy plaim w3c for the bad MS implemantation of the stanndaed?
As for ACPI; my knowlage of tha subject is terribley patchy so I cant’t comment on it.
PS: I know FF 1.5 stil has a few problems with CSS2 and that it fails ACID2 but IIRC that is on the roadmap for FF 2.0 (plz correct me if I’m wrong as I cold not locate the 2.0 roadmap)
“Yeah, that pretty much the philosophy behind the Linux fanboy.”
I dont use Linux and even if I did it’s besides the point. If you cant follow and implement a published standard in a usefull manner you’re a crap engineer and designer. Period. End of story.
It’s not the OS designers problem if a closed and undocumented piece of hardware doesnt work because it’s not POSSIBLE to support it well.
The fact that it works for many companies tells you a lot about the maturity of the IT industry.
One day people will wake up and become responsible consumers of computer technology. Hopefully it will happen before the industry is ruined by substandard, non-conforming products and implementations
unfortunately that is also my experience: not only that my notebook (Toshiba M50) doesn’t really support ACPI (e.g. the notebook just runs 1:30 until the battery is low or no way to suspend the system at all – although all needed kernel modules are loaded – tested with Gentoo, Ubuntu, openSuse), ATi is also too gay to support any of their gfxcards (including that laptop’s x600se). But because I used opensource and linux when the majority didn’t even know that such thing could exist, I won’t get scared by that
Well…actually, two things don’t work: suspend and gfx hardware accelleration. Blame Toshiba for the former and gay Ati for the latter. That’s not linux/unix/beos/skyos/syllable/hurd/bsd/… fault.
Not for me. If I reboot with ACPI on, my soundcard deactivates itself and has to be reactivated in bios on my box. Both FreeBSD (I’ve tried 5.3, 5.4 and 6.0) and windows 2000 worked just fine regarding this… Linux acpi works fine if I “poweroff” and turn my box on again however, thou a quick “reboot” without visiting the BIOS would be nice… But if it works for you, great. Just don’t moan about us having shitty hardware…
>> Not for me. If I reboot with ACPI on, my soundcard deactivates itself and has to be reactivated in bios on my box. Both FreeBSD (I’ve tried 5.3, 5.4 and 6.0) and windows 2000 worked just fine regarding this… Linux acpi works fine if I “poweroff” and turn my box on again however, thou a quick “reboot” without visiting the BIOS would be nice… But if it works for you, great. Just don’t moan about us having shitty hardware…
—
Well, it would be a nice idea to file a bugzilla entry …
if Linus continues breaking binary and source drivers with each minor Linux kernel release
Not even worth discussing. That’s never going to happen, and there are many well documented reasons as to why:
http://www.kroah.com/log/2005/11/03/
Notice some of the ludicrous mailing list entries linked to in there by someone with an OSDL e-mail address! Yay, let’s trust these guys!
The first big one is that hardware manufacturers will go off to a private lab and create binary drivers that are crap (and shitty ACPI implementations!), don’t work well for users, cause problems when testing the kernel and they’ll do it even if it’s downright illegal. If that were to happen we would have absolutely no open source drivers (as the kernel compels you to do), and the whole point of open source drivers would go down the tubes. The OSDL should be educating manufacturers on how to get involved in kernel development, but they’re not.
You and other people are also looking at this totally the wrong way. How many iterative versions of its kernel do you think Windows XP has had over the last four years? None, that’s how many. It’s the same kernel. However, I’m sure that Microsoft have many internal development revisions of it internally and I’m sure they don’t maintain binary compatibility at all. I’m also sure that Vista’s kernal and drivers won’t be binary compatible with XP’s, and they never have been with Windows. That’s the way you have to look at the Linux kernel – a series of development revisions.
The solution for Linux? Grab one version of the kernel and standardise on it for four or five years and get manufacturers writing good quality open source and GPLed drivers. You can backport any major changes that you need, and this is what many Linux vendors like Red Hat have done for years. The only problem is that there is no one official, so to speak, version for everyone to standardise on.
I really don’t know why everyone is whinging because people are actually developing the kernel.
Edited 2005-12-15 22:30
Another thing people are missing is the fact that there’s a pressure at all.
The fact that the industry is negotiating at all means that they’ll eventually cave in.
We just have to wait untill they think the value of supporting Linux is equal to the precieved cost of FLOSS drivers.
Bingo! For instance, there’s a reason all high end ethernet cards work well with Linux. It commands enough market share of the server market that manufacturers have no choice but to make sure that their products work there.
Just one fact segedunum, many windows XP drivers are binary compatible with windows 2003 and even on Vista.
As an example, in Vista, Microsoft has changed the NDIS to support a new packet format. But microsoft invested enough money to write a translation layer so that the older drivers work fine.
Sorry to burst your bubble, but Microsoft puts a lot of effort to make sure that Windows provide backward compatibility.
Linux kernel on the other hand is a pain in the freakin a$$. You got some old hardware, sorry new linux kernel won’t work on it or go hack the driver code yourself. Ahh this is what we call consumer friendly OS.
Linux is generally quite excellent about supporting older hardware. They rarely drop support for something unless it becomes rather marginal.
That said. OS X is considered a consumer-friendly OS. Not only does Apple drop support for older hardware with every single release, but even minor releases generally break a kext or two. Interestingly, Apple is much in the same boat as Linux here. The reason new OS X releases break kexts is because Apple releases so often. I don’t think many Mac users would want Apple to go 5 years between releases like Windows does, just because of the occasional broken driver!
Rayiner either you are another biased windows hater or a chutia (= dumbass_. You decide which one you are.
Apple doesn’t need to bother about hardware because they make all their hardware but think about the number of windows users and weird hardwares windows run on. Let aside that, think of 100s of different drivers for windows. Microsoft is able to develop new driver models without breaking the OLD stuff.
On the other hand, Linux is a mess. SysInternals guys wrote filemon for Linux and Linus broke it in next rev…ultimately they got tired of the goose chase and gave up. I wonder how many people will be willing to update older drivers all the time. How many people will be willing to compile their apps all the time…
Linux is a big freakin mess….As i said in past and i will say it again:
Linux is a monolithic kernel with no unified driver model or fully asynchronous IO wrapped up with ad-hoc GUIs to make 100s of incompatible distributions to confuse the hell out of a user.
Linux = Losers Invented New Unix xeroX (= clone)
“I wonder how many people will be willing to update older drivers all the time. How many people will be willing to compile their apps all the time…”
“Users” dont have to compile anything, they use a distribution and just install the pre-compiled packages.
“Linux is a monolithic kernel…”
It’s not a monolithic kernel, it’s a hybrid kernel just like NT/XP.
“…100s of incompatible distributions to confuse the hell out of a user. ”
It’s not confusing since joe User only uses one distribution.
“Linux = Losers Invented New Unix xeroX (= clone)”
How clever. Nice to see that the art of trolling is still alive and well.
It’s not confusing since joe User only uses one distribution.
Really? And which distro is that? Should they go for Xandros, RedHat, Suse, Mandrake (or whatever it’s called), Ubuntu, Slackware or (god forbid) Gentoo?
“Users” dont have to compile anything, they use a distribution and just install the pre-compiled packages.
Oh yeah? Every time I use to get a kernel update on Ubuntu I had to rebuild ndiswrapper from the console. This sort of thing in the norm, not the exception.
“Really? And which distro is that? Should they go for Xandros, RedHat, Suse, Mandrake (or whatever it’s called), Ubuntu, Slackware or (god forbid) Gentoo? ”
Whatever comes pre-installed or is offered by the seller. If it doesnt come pre-installed you’ll have to do what you do with any other product you dont know much about: research. Yeah, if Joe User wants to use Linux and it didnt come pre-installed obviously he has to do some research and ask around. it’s not different than any other product in this respect.
“Oh yeah? Every time I use to get a kernel update on Ubuntu I had to rebuild ndiswrapper from the console. This sort of thing in the norm, not the exception.”
I dont buy shitty unsupported hardware so I dont have this problem and if it bothers you so much obviously Ubuntu isnt for you and you should use a different distro. Ubuntu == Linux.
Linux is a monolithic kernel with no unified driver model or fully asynchronous IO wrapped up with ad-hoc GUIs to make 100s of incompatible distributions to confuse the hell out of a user.
More like:
Linus is a kernel that is often used for special devices such as cell phones, TiVos, or Google search machines and is sometimes used on the desktop by hobbiests and cheap businesses.
And none of its a bad thing. I can’t see myself going back- its 64 bit support its far better!
Yeah they never win, they have to know stuff, and all they have to show for it is a free commercial quality unix clone that just gets better and better at their beck and whim forever.
Sorry to burst your bubble, but Microsoft puts a lot of effort to make sure that Windows provide backward compatibility.
That’s because it’s virtually the same kernel. Vista is not compatible with XP or 2003, and neither is NT 4.0, 2000 or any of the 9x series.
You got some old hardware, sorry new linux kernel won’t work on it or go hack the driver code yourself.
That’s because it’s a new kernel.
Just one fact segedunum, many windows XP drivers are binary compatible with windows 2003 and even on Vista.
Since XP and 2003 use much the same kernel and inner working, yes. However, you can’t say that about Vista – at all. It will be different, just as XP is not compatible with 2000, NT 4.0 or 9x.
Sorry to burst your bubble, but Microsoft puts a lot of effort to make sure that Windows provide backward compatibility.
Considering the above isn’t true, this isn’t either.
Linux kernel on the other hand is a pain in the freakin a$$. You got some old hardware, sorry new linux kernel won’t work on it or go hack the driver code yourself.
That’s because it’s a different kernel, in the same way as NT 4 is different to 2000, which is different to XP and is different to Vista. The notion that these are all binary compatible is absolute tosh.
“if Linus continues breaking binary and source drivers with each minor Linux kernel release”
And you know damn well that will never change. Linus has always been crystal clear on this point. Linux is a FOSS OS and closed binary only software (especially true for drivers) will always be second class citizens. This is not a problem with Linux, it is a problem for those who want Linux to be something fundamentally different from what it is. If a manufacturer wants their hardware to work very well with Linux they have one choice: release the source to their binaries and maintain them in the main Linux tree. Period!
“he doesn’t fix ACPI”
Others have addressed this issue, but you and Thom have chosen to ignore the real issue here. ACPI support in most hardware is buggy but tuned to work with Windows’ implemetation.
“unify all Linux distros under the same package/binary compatibility, it’s all going to waste”
Again, it will never happen. Not to mention the fact that Linus doesn’t have the ability to do this, even he so desired. No one has this ability.
The underlying problem here is that there is a conyingent who, for some odd reason or another, deperately want Linux to be a general purpose consumer desktop OS. It’s not; it’s actually pretty poorly suited to that purpose. Further, there is almost no finacial incentive to push Linux in that direction. I’m sure you don’t need me to list off all the failed attempts at creating such a beast. There’s very little money to be made in that market and an awful lot of headaches. That’s why you don’t see Linix distro companies vigorously pursuing it, not because they can’t get Dell to push pre-loads on the front page.
That said, Linux is doing just fine in those markets it is well suited to: servers, embedded appliances, and, slowly but surely, centrally managed desktops. Linux is, technically speaking, actually well suited to these markets and, more importantly, there exists a financial motivation for doing so.
In brief, find another bandwagon to hitch a ride on.
The truth hurts, Amiga users had the same unhealthy habit of denying facts as well.
While I understand Linus reasons for not stabilizing the platform, on the long run is very bad, Microsoft is keen on improving their “server” abilities more and more, displacing GNU/Linux because all the commercial software available for windows plus nearly all of the FOSS ones, (that is the biggest ever advantage an OS can have).
More and more I find people who replace their UNIX boxes with windows ones because of improvements in Windows (Win2003 is good enough for 95% of the population out there) and because of the software advantage, and not to mention that anybody (with or without a clue) can do a minimal administration on a Windows box, and not to mention basic troubleshooting.
Heck, all the GNU/Linux – UNIX power users I know run windows or MacOS as their main desktop. I know the purpose of GNU/Linux is not to replace MS, but if the GNU/Linux crowd doesn’t eat their own dog food…
GNU/Linux, UNIX looks and feels very abstract to most people, this continuous changes it suffers only accentuate this.
There is a big chunk of computers users who classify as “average”, not geeks or hackers, but not newbie’s either, those people are the ones who drive adoption, those are the people Linux scares out the most.
ACPI? Linux’ implementation is textbook perfect. The problem lies with broken ACPI BIOSes. In Windows, you rarely noticed the BIOS problem because the OEM Windows you use has a patch to work around the errors. Linux provides a way of applying the work-arounds too, but the problem is that someone needs to generate the work around first and you need to make sure it is applied.
Decent vendors (like Lenovo) have correct ACPI implementations. Crappy vendors, like Dell, have broken ones that need fixes.
You can’t blame Linux for not supplying post-release vendor patches to their own broken BIOSes.
I think the only thing that will bring Linux to the average desktop is being able to install a program and not have to worry about what distro they’re using. And sadly even a much larger covering by lsb isn’t going to make that happen.
Linux had its chance for heavy adoption, it didn’t catch on. If changes are being done to make desktop linux a better experience for the sake of making it a better experience, that’s fantastic. But continuing to chase after the average user years after they’ve said ‘no’ is a little sad at this point.
Well, the same problem exist on the windows platform when you look at it closely.
Not all modern windows programs works on all modern windows platforms, and some only works with limited functionality on some of these platforms.
I still have faith in autopackage
There are so many things wrong with your comment I don’t know when to start.
1) You assume Linux isn’t getting adopted. This is patently false. It’s made large inroads into the server market, it’s making large inroads into the workstation market, and outside the US, it’s making some high-profile inroads into the desktop market.
2) You assume, falsely, that desktop Linux has “had it’s chance”. This is due false ideas:
a) Desktop Linux has been around for years. This may be what Slashdot has told you, but that’s inaccurate. Desktop Linux is probably a few years old at this point. Do have make inroads into the desktop market, you need to have a product and you need to have people pushing that product. Things don’t just sell themselves! The oldest Linux distributers, namely RedHat and SuSE, were (and still are) focused on the server market. Novell and Sun are the first big companies to focus on the desktop market, and their Linux desktop efforts (NLD and JDS) are only a couple of years old at this point.
b) That becoming a player in a mature market with a monopoly competitor is something you measure the success or failure of on a timescale of a couple of years. It took Apple 6 years to get its market share above 10%, and the Microsoft wasn’t even a monopoly back then! If a competitor can take out the Microsoft OS monopoly*, it will be a progression measured over a decade, not over a couple of years.
* I actually don’t think this will happen. If Microsoft can be taken out, it won’t be in OS space, but Internet space, by someone like Google.
3) You say it’s “sad” to keep persuing users after they’ve presumably said “no”. By your logic, Apple is a dismal failure, with only single-digit market share after decades of effort. Is it “sad” for them to keep improving OS X?
Yeah… After reading that ars technica article about the histoy of personal computing, i realized how boring the pc market is today….
Yeah… After reading that ars technica article about the histoy of personal computing, i realized how boring the pc market is today….
That’s why you should support the alternatives! Buy Zeta! Buy a PegasosPPC! Buy boxed Linux! Buy RiscOS machines! Buy [insert name]!
Y’all can complain, but without doing something about it, it won’t change!
Get an Amstrad 664 emulator or even find a real one – write an OS in assembler..
Or smack your forehead or something… or what
Article states: “The last three years has been ‘the year of the Linux desktop.”
But I believe the “Year Of The Linux Desktop” tradition actually dates back to the late ninteen hundreds (1999) with 2000 being the first predicted year.
I once tried to look up some of the people that were SURE 2000 was the year and tried to interview them but I was not able to track them down for comment.
I think interviewing some of those people would make for an interesting article though.
The problem is the redundancy of work, without which we wouldn’t even have an option.
But seriously, we can’t expect a lot from so many different commanding entities, each with its own interest — it’s like as if every person that owns a share of Microsoft could decide in something, but in a greater scale.
But don’t worry too much. If the app is interesting, it will be available for your distribution in an easy to install package — we should adopt this tested and tried mechanism once and for all.
For example, there is the Debian world and there is the “other” world. At least in a Debian world you will have many packages in an easy to install way. Every distro would be advised to adopt the Debian packages once and for all. Or don’t complain about lack of support for your platform.
But don’t worry too much. If the app is interesting, it will be available for your distribution in an easy to install package — we should adopt this tested and tried mechanism once and for all.
root@nand:/home/blah# apt-get install skype
E: Couldn’t find package skype
Worked for me:
hac@ws:~$ dpkg -l skype
Desired=Unknown/Install/Remove/Purge/Hold
| Status=Not/Installed/Config-files/Unpacked/Failed-config/Half-installe d
|/ Err?=(none)/Hold/Reinst-required/X=both-problems (Status,Err: uppercase=bad)
||/ Name Version Description
+++-=================================-================================ =-==================================================================== ==============
ii skype 1.2.0.18-1 Free Internet Telephony – The whole world can talk for free
I have Skype installed as well:
dewd@marieta:~$ dpkg -l | grep skype
ii skype 1.2.0.18-1ubuntu0 Free Internet Telephony – The whole world can talk for free
Maybe if you tell what your distribution is (sarge, sid, etc) someone could tell you where you can find the package. I have Ubuntu Breezy.
No problems for me.
emerge skype
No problems for me.
So what you are saying is that if it works on one distro but not another then it isn’t a problem?
Basically. What do you want? Some sort of “approved by Linus” stamp on distros? Linux is a market with competing implementations. I know this is a hard concept for those accustomed to a monopoly market, but you have to choose your product (your distro) with regard to whether it does what you need it to do.
Which is great for a niche market. But the general public doesn’t like that kind of fragmentation. It’s too confusing. No, not because they are stupid. But because there are people out there whose lives do not revolve around computers and are too busy with the rest of their life to learn all about differences.
To rephrase your point:
No, not because they are stupid. But because there are people out there whose lives do not revolve around [cars/cellphones/TVs/shoes/watches/TV dinners] and are too busy with the rest of their life to learn all about differences.
A computer is a product just like any other. People have merely gotten lazy with regards to computers, is all.
You are right. Competition is bad for buissiness. Forcing people to choose between Windows or MacOS X is hurting the Desktop market. They should just merge so that people doesn’t have to choose…
Don’t put words into my mouth. In fact, read my other posts in here where I state that having Linux for competition is great for everyone.
My point is that having the choice between an ungodly number of distributions for Linux confuses people. I never even said it was bad.
Sorry, but your sarcastic implications won’t work on me.
Sorry. I guess I’ve should have gone back and read the thread.
I’m just so tired of this constant miss conception that Linux is _a_ system and that there exists some gruoup of people developing this system to compete with MS Windows.
It’s this constant whining from new users about how hard it is to pick a product that gets to me.
If the distros have bad marketing then that might be a problem for customers. I’m just saying that Linux is not _one_ project with _one_ goal so _it_ can’t possibly be hurt by fragmentation becase there is no _it_ to fragment.
That was part of my point, that there isnt a group of people developing it as a system. And yes, there is an IT to fragment, and that would be the GNU/Linux platform.
I didn’t know that the GNU/Linux platform was fragmented. How many sources of different GNU/Linux released do you know of?
There’s linux, linux-mm and a few others. All of them exist for a good purpose. And everyone using them as a platform is well aware of their differences.
When it comes to GNU I’m only aware of Emacs and XEmacs (I’m not saying that’s the only one though)
I guess you meant that distributions are implementations of _the_ platformt. I don’t agree. I think they are implementations of different systems using GNU/Linux as a platform.
To avoid some confusion here’s some definitions I use:
1. Linux = The project lead by Linus Torvalds to develop a source code distribution usable as a base to create operating system kernels.
2. GNU = The project lead by FSF to develp a source code distribution usable as a base to create UNIX like operating systems. Mostly as a way to convince other projects to use the GPL license and thus spreading software freedom.
3. linux or Linux = Some ideal non-existing operating system loosly defined as what would be the result if all existing systems and distributions based on GNU/Linux suddenly and magically merged inte one and together decided to exclusively target Microsofts customers for no good reason. Sadly this is also what most people outside the FLOSS community belive is the real state of things.
When I speak of the platform I mean the union of 1 and 2. When I earlier spoke of a non existing _it_ i was referring to 3.
I guess you meant that distributions are implementations of _the_ platformt. I don’t agree. I think they are implementations of different systems using GNU/Linux as a platform.
You just contradicted yourself.
GNU/Linux is what most people would call any distribution using the linux kernel and gnu tools.
I don’t know why you’re having such a problem admitting that there is a “linux platform”, where there are many many distributions, and that is fragmentation.
Because there is not reason to belivd that the different distributions should be compatible with eachother more than they should be compatible with any other system.
I don’t expect to be able run Windows programs on MacOS in the same way that I don’t expect to run Ubuntu programs on RedHat.
I never said they should be compatible…
I guess this is where threaded comments are useful.
Someone said “If the app is interesting, it will be available for your distribution in an easy to install package”
So I responded by pointing out an interesting app (Skype) that is not available for my distro.
In response to my response, someone chimed in by pointing out it IS available for their distro, but what good does that do me? It angers me when people constantly reply with “Well that works in X distro so I see no problem”
Desktop Linux is not some kind of magical combination of all the strengths of all distributions combined.
Aparrently Linux support is not easy enough to do for commercial software (closed source) companies, partly becasue “Linux is a market with competing implementations”
One company may build packages for version n of distro Y, while another company may build packages for a different competing distro, or only a specific version of that distro. Switching distros only shifts the problem.
What is your prediction rayiner? When if ever do you see Linux taking > 70% of the global desktop market?
Aparrently Linux support is not easy enough to do for commercial software (closed source) companies, partly becasue “Linux is a market with competing implementations”
Bingo. That’s one of the problems. And that’s why I said if you spooge out some source code, scream freedom, and hope for the best you’re going nowhere. These people are competing among themselves instead of competing against Windows or Apple. So it’s 2006 in a couple weeks and just now this Portland project is starting. This stuff should have been done 8 years ago and things would have been different.
Aparrently Linux support is not easy enough to do for commercial software (closed source) companies, partly becasue “Linux is a market with competing implementations”
That’s a reality of life, and the sign of a healthy market. I’m not going to argue about this with you because you’re simply wrong. From an economics point of view, the computer OS market is messed up, and you’ll get nowhere trying to pretend that this is somehow a good thing. Yes, some things are easier when there is no competition. But hell if I’d rather live with a monopoly just to avoid choosing my long distance carrier!
What is your prediction rayiner? When if ever do you see Linux taking > 70% of the global desktop market?
Windows has messed up people’s minds. Your question is like asking “when if ever do you see Ford taking > 70% of the global car market”? That said: I think Linux will settle at about 10-20%, over the next decade, with heavy emphasis on corporate and professional desktops. I optimistic that Apple might make it to 10% in the same time-span, with Microsoft keeping the remaining 70-80%. This is assuming, of course, nothing interferes with Microsoft’s monopoly.
EDIT: Removed foul language.
Edited 2005-12-16 05:32
Nope. I said it wasn’t a problem for me.
It’s just like wanting to use MSN Messenger 7.5 with Windows 2000. Not possible. Solution: Get another Windows distribution (WinXP or newer).
If something won’t work on your linux distribution: Get another one.
Or add some more repositories. Besides that. You can get a static build from skype, a version which runs on all Linux distributions. Just download it from Skype like you would do on Windows. Problem solved.
It’s not impossible to due on Linux what is down on Windows. It’s just not a solution we like (or it would be the most common solution, which it is not).
>Every distro would be advised to adopt the Debian >packages once and for all.
I couldn’t agree more. I’m not saying all should adopt apt or whatever front-end. All distros should adopt a common package format. I think deb is the best solution since it is the chosen format for so many distributions.
But don’t worry too much. If the app is interesting, it will be available for your distribution in an easy to install package — we should adopt this tested and tried mechanism once and for all.
Is Open Office 2.0 final popular on linux? I can’t seem to get the final current version for my distro and have been told to wait another six months before I can have an updated distro with the final Open Office 2.0 pre installed!
The same was said for Firefox 1.5 – I guess I should just install windows, as I can have all those programs up to date and current without waiting for another release of windows!
For example, there is the Debian world and there is the “other” world. At least in a Debian world you will have many packages in an easy to install way. Every distro would be advised to adopt the Debian packages once and for all. Or don’t complain about lack of support for your platform.
Uhuh! Then I should uninstall Ubuntu(Doing it the Debian way) and use Debian? Would this give me updated programs without waiting 6 months?
Noone listens to music on their pc anymore, nobody plays PC games, nobody writes email on PCs, noone uses PCs anymore to do anything useful – PCs are only for developers.
sorry that boat has sailed.
Noone listens to music on their pc anymore, nobody plays PC games, nobody writes email on PCs, noone uses PCs anymore to do anything useful – PCs are only for developers. sorry that boat has sailed.
Where can I get some of what you’re smoking?
troll!
Anything bigger than an ipod is wasted.
(with apologies to Frank Zappa)
What Eugenia (standard binary APIs for drivers), and Emerson (install software no matter what distro) are talking about are the type of standards that the Portland project is working on.
Being that the Portland project is led by the OSDL, and has participants spanning many open source projects and corporations, it has a fantastic chance of succeeding in it’s goals (desktop standards).
Once these desktop standards are well established, one of the major hurdles blocking mass desktop Linux adoption will be eliminated. Once this occurs, the ecosystem built around desktop Linux will really grow, and will finally reach the kind of critical mass where reluctant OEMs like Dell, as well as consumers reluctant to try new things (for fear of compatibility issus), will really start taking the desktop Linux plunge.
Now, I’m sure this thread will be full of pro Windows posters trolling that “Linux is too hard”, or “Linux can’t do X”, or “Linux users are all zealots” or “Linux is socialism”, or that “desktop Linux will never make it”. Well, we all know they are either ignorant or are lying.
They won’t want to admit that Portland project standards are all that is needed to push desktop Linux over the hump.
Anyway, the Portland project is huge, and I wish them the best of success.
“Adobe Systems” is mentioned in the article as being one of the attendees. Is adobe wanting to get on board with linux to develop their previously-macromedia server products on linux? Or are they doing this because they want to port the likes of Photoshop to linux to be used in conjunction with major studio quality 3d apps like maya and softimage which already run on linux?
I hope it’s the latter, then I’d have no reason to use windows anymore.
I think a lot of people are severely misunderstanding what binary drivers mean to Linux. Linux, as a system, has its own particular dynamics, and binary drivers just don’t fit with those dynamics.
Consider Xgl. One of the holdups with XGL has been driver issues. Currently, there is only one set of chips that can run the preliminary Xegl implementation — the Radeon R200, because it’s the most modern card that has open-source DRI drivers. If the ATI or NVIDIA drivers were open source, it would be massively easier to get something like Xgl running.
As I said, it’s a matter of dynamics. One of the aspects of Linux as a system is that the Linux developers don’t have access to the source code of binary drivers like Apple and Microsoft’s developers do. Both Apple and Microsoft can futz around with their OpenGL implementations, because their developers have the source code to ATI’s and NVIDIA’s drivers. You’d better believe that this makes things like a GPU accelerated GUI much easier to develop!
Another aspect of Linux that’s commonly overlooked is that it thrives on rapid releases. Linux kernels are released every couple of months, and Linux distros release new versions as often as twice a year. Compare this to Microsoft’s 4-5 years between major releases. It’s a simple fact of life that if Linux kernels were released every year or two, the development model simply wouldn’t work. There is no way to get a worldwide group of developers to coordinate on that sort of time scale. This sort of release schedule is unfavorable to binary drivers, however. Extremely dedicated companies like NVIDIA, with large driver development teams, can afford to make new releases every month or two, but smaller companies (interestingly, almost always those companies for whose hardware Linux drivers aren’t available), cannot do that for Windows, much less a secondary platform like Linux.
From the user point of view, it’s not so much a matter of convenience as a matter of reasonable expectations. People cannot expect Linux to be like Windows with regards to hardware. It’s status as a secondary platform means that people should expect it to be like Apple. Most drivers that Apple users use are first-party (supplied by Apple itself). Users have to do a modicum of research to figure out if their new hardware is going to be support on their Mac before buying it. Linux is really no different in this regard, and expecting it to be so is unreasonable.
I agree with the “Buy” part, most in the oss community never buy an oss os, my suggestion to get the ball rolling is this, try all the distro’s or os’s you wish, when you find one you want to stick with, either a.)buy it, b.)donate to it , or c.)contribute to the project with time, coding, usability, bug reports, or graphics work.
Almost everybody can do one of the above things, and if they can’t, that’s cool to, but if everybody that was able did the things I pointed out above, the alternative os’s would be Windows and OSX.
I’d say standardize for 1 year. 4-5 is long even by Microsoft’s standards.
Obviously this should have been done years ago. All the infrastructure and kernel-related stuff equals someone taking the bull by the horns and producing a real desktop OS from the kernel on up and not just slapping Gnome or KDE on top of it.
Firefox is a killer app, but like Gnome and KDE, its not Linux specific. It’s sad that these people are delusional enough to think that KDE or Gnome is going to propel desktop linux into the mainstream.
Until Google or someone else with money comes along and makes a big commitment then nothing is going to change. By the time all this infrastructure is in place the OS after Vista will be around the corner and OSX will be a few generations ahead.
Sorry, but this is the type of stuff that should have been going around ’98 or so. Linux on the desktop could have been something. But Gnome and KDE aren’t going to cut it.
Luis Villa and others are starting to realize the truth of the situation http://tieguy.org/blog/index.cgi/523
It’s interesting that he brought up a hypothetical Knome and that it wouldn’t change anything. He’s right, but 7 years ago it would have made a difference. You just can’t spooge out a bunch of code, scream freedom, and hope to take over the world. The nature of open source is its own worst enemy when it comes to the desktop
Open source “just spooged” out a bunch of code, and now has a desktop market share approaching Apple’s. Companies have employed much more focused and well-funded efforts on competing on the desktop, with less success. What’s the market share of OS/2 these days? How about BeOS? Linux has a couple of percent, in a market completely hostile to non-Microsoft competitors, and people act like its a failure!
Open source “just spooged” out a bunch of code, and now has a desktop market share approaching Apple’s
Linux has been around for 15 years. If it takes this long to approach Apple’s market share with all the supposed benefits of open source then it’s already lost the game. You people will never get it, but that’s not surprising.
A million distros, 2 less than stellar desktops, no binary drivers, wacky cult-like ideologies = Linux has already failed on the desktop. You and the other fanatics around here just don’t realize it yet.
Linux has been around for 15 years. If it takes this long to approach Apple’s market share with all the supposed benefits of open source then it’s already lost the game. You people will never get it, but that’s not surprising.
A million distros, 2 less than stellar desktops, no binary drivers, wacky cult-like ideologies = Linux has already failed on the desktop. You and the other fanatics around here just don’t realize it yet.
Harsh! Well, pretty accurate, but still harsh! 🙂
The Linus Thorvald blowup on printer dialogs earlier this week highlighted one of the biggest problems. Linus (incorrectly) chastised GNOME’s developers for trying to make a simpler, more consistent UI. So, a godfather of Linux wants GNOME to be more complex. After 10 years in IT, I can tell you that end users like simpler, easier, more consistent interfaces. That’s one of the reasons Windows has succeeded on the desktop for so long – you can sit down in front of any version of Windows from 95 to Vista, and start easily using it to get work done.
Which brings me to my other point – having apps to get work done on. The killer app for business desktops are groupware email, word processing, spreadsheets, and browser – in that order. I can go to Sourceforge right now and find probably 300 text editors for Linux. All are cool and interesting in their own way, but no one cares. Take all the wasted talent churning out utterly useless repetetive utility apps, and make a damned email system that competes with Exchange. Make a directory services system that competes with Active Directory (forget about Edirectory – it’s dead, and never offered out of the box management tools for desktops that worth a wet diaper). Make a word processing app that does not use 10x as much RAM as winword and has fonts that don’t look retarded.
To paraphrase, “it’s the apps, stupid!”
“Linux has been around for 15 years. If it takes this long to approach Apple’s market share with all the supposed benefits of open source then it’s already lost the game. You people will never get it, but that’s not surprising.
A million distros, 2 less than stellar desktops, no binary drivers, wacky cult-like ideologies = Linux has already failed on the desktop. You and the other fanatics around here just don’t realize it yet.”
Wow, the Windows zealots/trollers are out in full force on this thread.
Some points:
1. Apple’s market share – According to IDC numbers, Linux passed Apple’s market share, ever so slightly. And IDC numbers come strictly from unit’s shipped, which means box sets sold, or PCs with the OS pre-installed. Considering that the majority of Linux use comes from free downloads, and considering the media’s anti-Linux bias, and considering all of the MS FUD, and considering all the paid MS trollers, and considering MS lobies politicians heavily, Linux on the desktop has already done remarkably well.
2. “A million distros, 2 less than stellar desktops”. First, yes there are lot’s of distros. However, there are only 10-15 “major” distros, and among these they are divided up into big business/server distros (Red Hat, SuSE), geek distros (Slackware, Gentoo, pure Debian), hobbiest distros (Ubuntu, Fedora) and newbie oriented desktop distros (Mandriva, Linspire, Xandros, Mepis, PCLinuxOS). And distrowatch.com has a section that makes distro selection very easy. Second, if the comment was refering to Gnome and KDE – well, eye of the beholder. To me, and many, many others, both Gnome and KDE are vastly superior to Windows – more configurable, easier, more stable, nicer looking. And the newbie oriented distros are every bit as easy as both Windows and MacOSX. I’ve mentioned it before, but my completely non-technical, computer neophyte, 71 year old Mother is currently using Linspire Linux with the greatest of ease, pleasure, and lack of problems (no viruses, pop ups, worms, crashes, or spyware).
3. “no binary drivers” – wrong. Nvidia drivers to name but one. Linux does support binary drivers. It’s just that completely “free” distros, like Fedora or Debian, do not load proprietary drivers by default. But “for pay” packaged distros, do include them out of the box. And Linux supports more hardware out of the box than Windows. Window requires drivers installed separately. That is a fact, that most people don’t know because most people don’t have to install Windows.
4. “wacky cult-like ideologies” – The usual Windows zealot trolling comment. So because you like competition and freedom of choice, and want to buy/obtain qaulity software from vendors/distributors who obey the law and care about the quality of their product, you have “wacky cult-like idealogies”. Nice.
5. “You people will never get it, but that’s not surprising.” – another typical Windows zealot trolling comment. Oh do tell us, oh enlightened one, how us poor ignorant Linux users will never get it. Actually, go stuff your ignorant holier than thou attitude up your @ss.
Wow, the Windows zealots/trollers are out in full force on this thread.
Oh please, is that the best you have? Some of us have been using Linux for years and years…on the server, but have yet to be impressed with the desktop efforts.
1. Apple’s market share – According to IDC numbers, Linux passed Apple’s market share, ever so slightly. And IDC numbers come strictly from unit’s shipped, which means box sets sold, or PCs with the OS pre-installed.
You might be right. If anything it’s even. My point, actually, is that it should have surpassed Apple a long time ago.
Considering that the majority of Linux use comes from free downloads, and considering the media’s anti-Linux bias, and considering all of the MS FUD, and considering all the paid MS trollers, and considering MS lobies politicians heavily, Linux on the desktop has already done remarkably well.
Oh, if you can’t compete then resort to whining that your “superior” operating system is making more headway because of evil forces. That statement is related to wacky cult-like ideologies…see below.
First, yes there are lot’s of distros. However, there are only 10-15 “major” distros, and among these they are divided up into big business/server distros (Red Hat, SuSE), geek distros (Slackware, Gentoo, pure Debian), hobbiest distros (Ubuntu, Fedora) and newbie oriented desktop distros (Mandriva, Linspire, Xandros, Mepis, PCLinuxOS
Yeah, and that’s exactly the problem and why the Portland project exists. It’s hard to distribute linux software if you’re not in their repository ecosystem…ISVs hate it.
3. “no binary drivers” – wrong. Nvidia drivers to name but one. Linux does support binary drivers. It’s just that completely “free” distros, like Fedora or Debian, do not load proprietary drivers by default. But “for pay” packaged distros, do include them out of the box. And Linux supports more hardware out of the box than Windows. Window requires drivers installed separately. That is a fact, that most people don’t know because most people don’t have to install Windows.
Binary drivers are a bitch on Linux. Just take the wireless 8180 chipset from Realtek. They had a binary-part (ala Nvidia) for kernels 2.4.22 or below and then didn’t even produce a driver for the 2.6.x kernel. So people are forced to use ndiswrapper and we know that’s a pain if you have to get windows drivers for it to work. Realtek finally did get a pure open source driver, but it took a long time. The problem is that there is no stable ABI. You want to make it easy, not hard. And someone like Broadcom are definitely not in the mood to produce a binary driver if there is no stable ABI. So it is an issue.
4. “wacky cult-like ideologies” – The usual Windows zealot trolling comment. So because you like competition and freedom of choice, and want to buy/obtain qaulity software from vendors/distributors who obey the law and care about the quality of their product, you have “wacky cult-like idealogies”. Nice.
I’m not talking about “open source” – not people like OSDL. I’m talking the FSF extreme wing. I guess some people would rather not acknowledge that the majority of people will always look at code and computers as tools and/or technical curiousity and will never buy into the whole “freedom” thing. I’ve been making money off of Linux and open source for a long time, so I don’t have any problem with it. But I use it as a competitive advantage and not as some philosophical entity. And if extremists hurt linux desktop adoption because the source code is holier than thou, then that’s their problem.
5. “You people will never get it, but that’s not surprising.” – another typical Windows zealot trolling comment. Oh do tell us, oh enlightened one, how us poor ignorant Linux users will never get it. Actually, go stuff your ignorant holier than thou attitude up your @ss.
Yeah, the truth hurts I see, but until you recognize the marketing, timing, organizational, technical, and other aspects of promoting software, you are doomed to live in fanboy fantasyland and not able to realize the concerns that even OSDL and Luis Villa have expressed.
“Yeah, the truth hurts I see, but until you recognize the marketing, timing, organizational, technical, and other aspects of promoting software, you are doomed to live in fanboy fantasyland and not able to realize the concerns that even OSDL and Luis Villa have expressed.”
I work for a small software company that produces a server/enterprise level sofware that runs exclusively on Windows (but connects to other systems using open standards). I’m accutely aware of marketing, timing, organizational, technical, and other aspects of promoting software – it’s how I make my living.
And guess what? I agree with you. The Linux community, or Linux distros, need to do a better job of those business aspects.
But that’s why the OSDL is doing the Portland project (along with promoting standards). Also, companies like Mandriva, Linspire, and Xandros are actually doing very fine job of promoting/marketing their desktop products. They are gradually chipping away at the MS monopoly, getting in with governments, big competitors like Wal Mart, etc.
What brand of idiot are you?
1) Linux is not 15 years old. At best, it’s 14.
2) For the first third of its life, it was a hobby OS. For the next third, it was present in the commercial sector only has a hobby OS. Only in the last few years have companies like Novell and Sun been selling it on the desktop. You can’t gain market share in a market in which you don’t compete, and Linux is only beginning to compete in the desktop market.
3) By your method of counting, you could consider NT a “failure” on the desktop, since it was started in 1988, but didn’t garner significant desktop market share until 2000 (12 years later).
4) The comparison to Apple, is silly. Apple was a billion dollar company when Linus was barely out of middle school! Apple had hundreds of developers when Linus was learning algebra!
5) The comment about “the power of open source” is ridiculous. The fact that we’re even having a discussion about an OS some guy created for fun competing with multi-billion dollar companies is a testament to the power of open source! The fact that Linux on the desktop is even viable, in a world where OS/2 is dead and BeOS is dead is a testament to the power of open source!
Jesus. IBM couldn’t do it, Be couldn’t do it, Apple has peaked and languished in attempting to do it, and you’re criticizing Linux for not doing it fast enough?
What brand of idiot are you?
The brand of idiot that has to tell you idiots the reality of the situation.
1) Linux is not 15 years old. At best, it’s 14.
So it’s now an excuse because it’s been 14 years 3 months since the release of 0.01 and not 15 years? In 9 months are things going to be dramatically different. In 9 months will it finally be the year of the linux desktop?
2) For the first third of its life, it was a hobby OS. For the next third, it was present in the commercial sector only has a hobby OS. Only in the last few years have companies like Novell and Sun been selling it on the desktop. You can’t gain market share in a market in which you don’t compete, and Linux is only beginning to compete in the desktop market.
And that’s an excuse because somehow hobbyists didn’t care about guis in 1991? Where’s all this magic of open source development? Is Linux somehow absolved because it didn’t recognize the importance of the gui sooner?
3) By your method of counting, you could consider NT a “failure” on the desktop, since it was started in 1988, but didn’t garner significant desktop market share until 2000 (12 years later).
Microsoft had other operating systems that were geared towards the desktop – that were by all acounts inferior to Linux at the time. I’m talking the 1996-2000 era. Linux should have exploited that weakness at that time, but didn’t. If Linux can’t compete then on the desktop then what makes you think it can now when Vista will be coming out next year? We know that Vista probably won’t be crap. Microsoft does tend to get a bit better as time goes along. We know that they will have developer tools and SDKs that many developers like to use. If Microsoft is such the doofuses that some of the crowd around here make them out to be why can’t they exploit it?
4) The comparison to Apple, is silly. Apple was a billion dollar company when Linus was barely out of middle school! Apple had hundreds of developers when Linus was learning algebra!
Apple was tanking for years with failed OS project after project. I don’t consider Apple some pinnacle of operating excellence until OSX came out. There’s another opportunity that Linux didn’t exploit and with Apple going intel now we’ll probably see bigger sales and another missed opportunity during the OSX on PowerPC phase.
5) The comment about “the power of open source” is ridiculous. The fact that we’re even having a discussion about an OS some guy created for fun competing with multi-billion dollar companies is a testament to the power of open source! The fact that Linux on the desktop is even viable, in a world where OS/2 is dead and BeOS is dead is a testament to the power of open source!
So I presume from that comment we should just have lower expectations. I seem to have higher expectations of the open source talent then you it seems. When I started using Linux in ’97 (back in the redhat 4.x and fvwm95 or whatever days) I thought things would be further along and things would have come together in cooperation.
Jesus. IBM couldn’t do it, Be couldn’t do it, Apple has peaked and languished in attempting to do it, and you’re criticizing Linux for not doing it fast enough?
Coincidently, we moved to Linux because OS/2 was dead in ’97. If anything, that’s just more competition out of the way for linux.
“A million distros,”
Even though there are quiute a few they arent even close to a million.
“2 less than stellar desktops”
Says you, many think otherwise. An opinion is not a fact.
“no binary drivers”
I guess the nVidia closed drivers arent, uh, binary?
“wacky cult-like ideologies”
Yeah, caring about stuff like ethics and small companies rather than greed and multinational conglomerates is some wacky, far-out hippyshit.
It’s must better to just bend over and take it like a man from big business, innit.
I think one thing people fail to understand is that the nature of the desktop market. Posters here seem to just want a better version of Windows. Nobody is going to get any marketshare being a better version of Windows. The incentive to do that is just not there. Apple is a much better OS in the same marketing model as Windows, and it still only has a 3% market share after decades of working on things.
People fail to remember the reason why we’re talking about Linux on the desktop at all: freedom. For some people, its a philosophical issue. For others, it’s a matter of practicality: freedom allows flexibility. Flexibility is what Windows lacks, and flexibility is why anybody is using Linux in the first place. Eliminating flexibility might make Linux a better OS for a reviewer, but will make it one that nobody wants to switch to, because it offers them nothing Windows doesn’t.
You don’t need to eliminate flexibility. Also, Windows is a lot more flexible than most people realize, considering it is a closed source Operating System.
But about freedom. Freedom is definitely good. I for one am glad “Linux” exists and am grateful for what they are doing. It gives us another choice, and that is great. Some of the more fanatical Linux advocates seem to think that freedom means using *nix. Freedom means using what you want, which includes Windows.
You don’t need to eliminate flexibility. Also, Windows is a lot more flexible than most people realize, considering it is a closed source Operating System.
It’s not just flexibility of configuration (though that’s a big part). It’s flexibility of application, flexibility of specialization, and flexibility of deployment. Linux is simply easier to apply to certain tasks (eg: embedded ones) than Windows. It can be easily specialized to certain tasks, while Windows, by virtue of its closed-source nature, cannot. For example, I recently learned that FermiLab has customized versions of Linux, and its well-known that the NSA has their own version as well. Lastly, Linux is a lot more flexible in deployment. Not only are low to no cost licenses available, but commercial Linux vendors generally have less arduous licensing terms than does Microsoft, and if you don’t like those terms, there are a range of providers to choose from. A home user might not care about these things, but remember that a huge percentage of the desktop market is business users, and they certainly do care about these things.
You’re definitely right about that. Linux has the advantage there and probably always will, or at least in the forseeable future.
But then again, you’re talking about niche areas. So while you are 100% correct in what you said, it has little relevance on the desktop, which is what this article about.
Why does everyone assume that the only desktop machines are used at home? Corporate desktops are part of the desktop market.
Yeah but even in a lot of corporate environments, it’s just basic desktops that don’t need mcuh specialization, if any, and are not embedded. Again, it’s a niche area that Linux has cornered and Windows has little chance. Granted, it’s more prevelant in the corportate market than home.
I have a strong feeling that the only thing that will enable Desktop Linux for the masses is pre-instsallation on the machines. If OSDL or someone else starts providing desktops in the majority of shops right alongside the pre-installed MS Windows machines then they will seriously be considered and probably starting selling maybe 1 or 2 years after they are first presented everywhere.
Of all shops even PC World doesn’t even provide this, so I expect little chance of pre-installed Desktops being popular.
Ever tried to change out the window manager in Windows XP? Is it even possible? Most of the desktop environments I’ve seen for Windows XP end up using XP’s window manager… Sometimes they skin it for you, but you can still see the complete lack of features.
Ever tried to do it in Gnome or KDE? Really easy, two commands
. Of course, making the change apply at startup in an efficient manner is more involved.
It’s not the FOSS nature of FOSS that makes the software tend to be more modular. It’s the modularity
. Someone could write a completely monolithic program and release it under a free license and find almost no one hacking on it or making things changeable. It was done, we called it xfree86, and eventually it got bad enough for a fork to happen.
Windows, and by Windows I mean Windows and not NT, is monolithic by nature. Microsoft has a lot of work to make it more and more modular; but they’re working from a bad initial design and I’m sure they’ve figured this out by now.
Unix has always been modular by nature. It’s always been about small programs which have terse outputs and accept liberal inputs so that they can be programmed to work together easily. Windows has been about API level hookups that let c++/c developers change things around in prescribed ways.
Unix is so modular you can run Mac on it
.
But I think you’re right. Standards will not eliminate flexibility. They, if done right, can actually increase it. It was unwritten, and written, standards that have made Linux of today flexible (and likely some technological limits which’ve influenced things). But, hyper-intrusive standards will just chase developers away.
I’m not sure I’ll care which way OK/Cancel is until I can, easily, write programs to create useful inputs on X11 apps. Which, btw, is being developed and has been for some time but never seems to get done and used. I wish it’d get easier; it’d be nice for testing! If anyone knows of a good package, I’d love to hear about it
.
There’s an oft not thought of reason why flipping buttons around on users can actually be good: It teaches them to read before they click. The first time you delete 2 hours of work cause ok/cancel were flipped you’ll read what that dialog actually said
.
Maybe we should just look at alternatives to dialogs; they seem to be an awfully big point of contention between users and developers: On every platform.
I think the key is the business desktop. Many people will want what they have at work, and if a lot (10% is a lot here) use it at work you’ll see places shipping it pre-installed.
That, in itself, wouldn’t make a huge adoption of Linux. But, I think that’s the way to get Linux pre-installed: Get it in front of people at work. There they have someone to help them with silly quick questions and do that hard parts for them. If they like it they’ll be more willing to go to the work of having it at home.
There are already some shops, here and there, where you can buy pre-loaded Linux computers. The trouble with this though is that small shops have been burning people on PC sales for years; and people, rightfully, don’t trust anyone in the PC business, especially not small businesses in the PC business.
I am running linux and windowsxp on same computer,
I had tighten every single setting on winxp for security reason, now ACPI part doesn’t work with winxp, but it works fine with linux..
This might sound strange, but that combination would have a chance to push the open source desktop over the hump.
KDE needs to get past C++ programming. It’s 2006 and it’s time to move on to a managed environment of some kind for most app development. We’re in the final year wait for Vista and Microsoft hater or not, it’s going to have some impressive tech as well as sdks…Balmer’s right, it’s all about developers:)
Java is another possibility, but I see C++ programmers really appreciating C# (and even the class libraries) more than Java. But if petty, political bickering stops Mono then Linux on the desktop will just continue to be in low single digits.
Remember, OSX going Intel is going to change things..(at least I think). Linux desktop has to fight that front too.
There are a lot of idiots in here. One camp saying that linux on the desktop is here. One camp saying it will never come no matter what.
Both camps are equally retarded, biased, or both.
I initially disagreed with killing off various packaging management systems but now that I think about it, that is a good start for linux as a whole. Of course reaching a standard would be hard. As a package, it would have to be something that can be transferable, easy to maintain and locate. But the source code could potentially produce the package format very easily. The package names would have to be in a standard format so it would be possible to put various packages in a tree that checks for dependencies (ala portage).
What i thought would be a good idea is if Linus released a straight up distro and called it Linux (the linux distro), so the linux community could have something that partially resembles something official. After those gnome comments however, I guess we can chalk that idea.
…until desktop environments (KDE, GNOME, XFCE) start implementing proper window management.
Read the manifesto: http://modeemi.cs.tut.fi/~tuomov/ion/intro.html#manifesto
Learn about dynamic window management: http://wmii.de/index.php/WMII/DynamicWM
Instead of settling to produce lame copies of the clumsy and awkward “desktop experience” that MS Windows and Mac OS X now offer, Linux desktops should concentrate on creating something better and more innovative than the proprietary OS’s.
the funny thing is that ion and wmii (i belive that was its name) is basicly a trowback to systems like windows 1.0 and the original desktop prototypes.
another funny is that wmii screenshots are exclusively of cli windows stacked side by side. now why is that?
maybe its because office apps, browsers and similar works best when they are more or less full screen apps? 99% of the time i browse with firefox in fullscreen because pages are made to that size. same deal with spreadsheets and word prosessing.
about the only time i use side by side windows is when i need to transfer files from one place to the next. and with say kde’s konqueror i can do that inside one window by splitting it.
then there are apps like im that most of the time sits idle in a corner. those are most of the time made up of multiple seperate windows and a main contact list. a layout like that basicly makes a mess of ion last time i tried it.
You’ve pretty much missed the point I was trying to make. The idea is that there are several window management modes to choose from and the ideal window manager should make it easy for you to switch between these modes. You should be able to switch without effort between the fullscreen mode and the mode where one window fills half the screen (and to decide if the splitting is done vertically or horizontally) while other windows show in smaller tiles, or you can switch to the traditional mode where windows are scattered all over the desktop, overlapping each other.
I also prefer using most apps in fullscreen mode (with tabs showing all the currently open windows) but for some tasks it’s useful to switch to another window management mode. I’m not saying that Ion or Wmii are perfect — just that applying these ideas (and maybe improving upon them) in Linux desktop environments would certainly make the “desktop experience” more enjoyable. The ability to switch easily between different window management modes could even become the “killer feature” that everyone wants to try out, something that hasn’t yet been implemented in other operating systems.
err, your original post more or less had no point. it had two links. one to the ion manifesto and one to the wmii site. outside of that there was some vague request to the gnome&kde camps for a “better” window manager…
i seems to recall a wm that behaved like was you are talking about. the main window was more or less full screen. the rest where basicly scaled down images of the windows along one edge of the screen.
thing is that something like ion will have to work with the existing apps available. ones you introduce something that dont work in either the one-window-pr-task/file or window/tab-in-window styles the ion way of doing things break down.
i have been thinking about a system that would try to marry the best of commandline and gui. this by floating gui dialogs on top of the cli (or basicly making the cli the desktop surface/wallpaper). i again ran into problem with this when thinking about a im system and similar that just sit in the background most of the time and then spit out a window whenever there is something happening.
the im client is a child of the desktop metaphor, and it have allready become a very important part of how we work.
hmm, one option is to rework the im system to behave like the text messages of a mobile phone.
upon a new message you get a icon somewhere. click it and a new window appear with the im message and two buttons: reply and dismiss. hit reply and the window change into a new window with the im message in the top half as a refrence and a text input area in the bottom half.
problem is that this would require people to change the way they use the im system, or rather go back to how say icq at one time behaved. currently people use im very much like they use irc. long trains of tought, multiple single lines. atleast thats my experience.
one would have to go back to thinking about is as email with a status addon. maybe a good thing in a way…
another problem is that any kind of gui breaks down the moment the nunmber of windows on ones desktop passes about 12 or so, maybe outside of exposelike systems that is. there is basicly to limited an amount of display space. hell, even expose can be overloaded, you just need so many windows that the details gets reduced to a small group of pixels
ok so one can have a single window zoom out when dragging the mouse over it but one still have to go looking for the window rather then instantly spoting it based on some shape or colors.
im guessing we have to trow away the desktop mentality alltogether. problem is that i dont know what to replace it with…
until desktop environments (KDE, GNOME, XFCE) start implementing proper window management.
If you find theat wmii better than the DE’s default window manager, just use it instead. The DEs do not require their default window manager to work, any sufficiently modern WM will do
1. create a lib that makes the app independent of the gtk/gnome or the qt/kde groups. basicly the app should talk to the lib, the lib should talk to the desktop in use at the moment. a bit like SDL for desktops
this would eliminate the need to have both sets of desktop base libs installed even tho your using only one. and it would allso allow the app to change apperance depending on what desktop it was used on.
face it, what the app need is access to a clipboard, the ability to load and save data (and therefor a load/save dialog) and maybe a “tray” module for apps like im’s and others that will just sit in the background most of the time.
sure, there may be a bit of a performance hit, but come on, we are not talking about the latest 3d engine here. most desktop apps waste their cpu time waiting for user input.
2. a similar layer for kernel drivers. a stable end towards the drivers, a recompileable one towards the kernel.
ok so it will not benefit kernel development. but just as the GPL style hooks that allow a dev to trow away the crash log on basis of a unfixable flaw in a binary driver, this atleast can put a stable front towards the desktop devices while not comprimising the kernel development to much.
i can see a problem with security, but then if security is a issue, dont use this module. its a third party item, much like ndiswrapper, that will never be taken into the kernel proper. its for ease of use, not security or performance.
The biggest problem Linux faces is the ever growing number of distro. It’s great to have choice, but too much choice can also be a bad thing. Do you go with RedHat, Suse, Ubuntu/Kbuntu, Slackware, Debian, Gentoo etc. Do you apps come in rpm, apt-get or source form? Then you’ve got the desktop managers. Do you go with KDE, Gnome or something more lightweight?
You’re average Joe isn’t really that interested interested in any of this. People just want to run applications. Most non-IT people I know use a computer for surfing, email, managing MP3s, storing digital photos and basic word processing. Now, all these things can be done on Linux, but there’s too many choice for each area and, most importantly, most apps fail to live up to their Windows counterpart. The FOSS world is full of “Yet Another MP3/photo/email” app there’s hasn’t made it to version 1.0 yet.
Even Apple realized that OSX was always going to be a niche player and ported iTunes to Windows. Sure it’s all about iPod sells and yes it will push some people towards buying a Mac, but given Microsofts huge market share you’re always going to see some drift. Prediction: When IE7 comes out there will be some drift from Firefox back to IE.
FOSSs bigest advantage is also one of its major problems. Its open nature means that theres no overall body to steer it in any direction. Groups do their own thing without a lot of thought for what someone else might be doing. Commercial outfits like Microsoft and Apple focus on one area and (if we’re honest) do a good job. Neither company push out 4 different mp3 players or 5 email clients. They produce one product and tend to do a good job of it. If the FOSS world really wants to make inroads into the desktop world then it’s going to have to learn to do the same.
I suspect that in the long term Linux will move in and take over the role that Sun use to enjoy. It will become a major player in the server market and will occupy a major position in the workstation market, but it’s never going to become a major player in the desktop market.
Really? And which distro is that? Should they go for Xandros, RedHat, Suse, Mandrake (or whatever it’s called), Ubuntu, Slackware or (god forbid) Gentoo?
Oh my god! Which car am I supposed to buy? Ford, Dodge, Chevy, …? It’s too hard for me to decide! Someone help or I’ll freak out!
Oh my god! Which car am I supposed to buy? Ford, Dodge, Chevy, …? It’s too hard for me to decide! Someone help or I’ll freak out!
Erm, yes, I must make sure the maker of my open source car gets more market share so that it can crush the really successful car company. I know, I’ll buy the car that isn’t compatible with the gas hose at 95% of the gas stations and has a completely different set of controls to all the other cars.
Each new release of the car is different to the old one, but hey, who cares about compatibility….
non-compatible with the gas hose? that only if you define software as gas. sorry but i dont see it as that. data may well be gas, but software is more like aftermarket mods to said car.
compleatly diffrent set of controls? in a car you have the wheel, some switches, buttons and levers for lights and so on, a gear shit or similar, and a set of pedals.
on both windows and the two major linux desktop you have all the makings of the wimp (windows, icons, menus, pointer) setup.
between diffrent cars, some of the functions change places. like say the way you twist the handle on something to change the lights and similar. what your trying to do is the same for all cars, but the tools to do so may be located in a diffrent place and use a switch rather then a button etc…
the real diff between buying a car and buying a pc is that when you buy a car it have a small set of tasks and requirements. a pc can be molded into just about anything.
i would say that you can compare the install of any hardware or software into a computer as doing after market mods to a car. only entusiasts do so for a car, but why is every user expected to do so for a computer?
i dont expect grandma to be able to check the oil or exchange the entertainment system.
im more and more thinking that a joe user computer should be more like a gaming console. a black box that you feed storage media filled with programs into.
want to write a letter? pop the media that carrys the office software into the machine and choose the word prosessor form the menu that appears.
files should be stored on removable storage units. be them flash based or similar. alltho avoid moving mechanical parts, atleast stuff like reading heads and so on, as part of the storage unit itself. having them as part of the console, sure, but not inside the storage unit itself.
thing is that the console should have enough ram to be able to handle a nice number of apps running at the same time.
im/mail and browsing apps should be part of the main rom of the box. as should video and audio playback. both im and audio/video playback should however be able to handle loadable modules so that you can have the same im support multiple systems. and the audio/video should be able to find the codecs needed alongside the files its supposed to play.
all this should happen automaticaly, so if i insert a movie and the software cant find the codec it needs internaly, it should look around on the media inserted.
the funny thing is that the im bit would allow me to bring my storage with me to a friend. insert it, fire up the im client and expect it to be able to log onto my system of choice, and show my list of contacts (what im system do not allow you to store that on their servers these days? still, one should be allowed to make a offline backup).
Edited 2005-12-16 15:37
For the third time on this thread, I’ll mention my 71 year old, non-technical, computer neophyte, mother happily, easily, peacefully using her Linspire Linux computer purchased from Fry’s.
Also, my brother (who’s never used Linux and mistakenly thinks it’s only for techies like me) and I made the decision at one point to install Windows on that machine. Huge mistake. within about 3 days my Mom was calling my brother and I constantly because of all the problems she immediately got with viruses, spyware, and pop-ups. I should have had her install an anti-virus. But she did not want to spend anymore money (she like the low price of that computer at Fry’s), and I don’t blame her. So, against my brother’s wishes, I promptly re-installed Linspire on that computer. And Joila’ all problems went away immediately. And my Mom is doing everything she wants to do with a computer (email, web browsing, games, documents, etc) very happily and easily.
My Mom’s experience with both Windows and Linspire is irrefutable proof that desktop Linux is definetely very viable.
Also, at least 95% of my personal computer usage is on Mandriva (my current distro of choice) Linux.
Desktop Linux is excellent, it just needs better, more consistent standards, and better marketing. It will continue to grow, and chip away at the MS monopoly. It won’t be easy, and it won’t happen over night. But happen it will.
For the third time on this thread, I’ll mention my 71 year old, non-technical, computer neophyte, mother happily, easily, peacefully using her Linspire Linux computer purchased from Fry’s.
Also, my brother…
Also, my mom…
We’ve been through this before many times. Personal anecdotes are almost meaningless to the big picture. I can claim that 50 people that I know are happy with Zeta, but that would be pretty much meaningless. It’s not that people can’t do things on Gnome or KDE. It’s a whole host of other issues that have already been discussed on this thread and many other threads countless times before.
Desktop Linux is excellent, it just needs better, more consistent standards, and better marketing. It will continue to grow, and chip away at the MS monopoly. It won’t be easy, and it won’t happen over night. But happen it will.
Your (and others) problem is that you see Linux development in a dynamic environment and Windows and Apple in a static environment. In other words, Microsoft and Apple don’t just sit around to wait for Linux to catch up. I would hope that some kind of free desktop on some other operating system other than windows or OSX, (Linux presumably because of its plethora of drivers), would be able to put a dent in Microsoft’s desktop somewhat monopoly. OSDL, Freedestkop, and the Portland Project are not living under the delusion that just because its free and has source code that somehow Linux will ever be significant on the desktop – hence “cautious” optimism.
My main point is that don’t automatically assume that the current KDE or Gnome organizations, philosophy, personnel, and development strategies are the ones that will propel linux into the mainstream.
“We’ve been through this before many times. Personal anecdotes are almost meaningless to the big picture.”
Personal anecdotes are far better than assumptions, factual errors, or just plain old FUD. When someone says technology X can’t do A, then a bunch of people come on and say they did A with technology X, then it completely blows the first person’s claims right out of the water.
Yes, I know that just because my Mom’s experience with Linux has been great (as has mine), it doesn’t mean everyone’s has. But the same goes with Windows, Mac, or anything else.
I’ve personally had tons of very aggrivating, time wasting, ridiculous, inexcusible (like you can’t believe this product has this problem) problems with Windows. This even though so many posters on OSNews say Windows is a great desktop and that Linux is a terrible desktop.
In short, anecdotal evidence is actually very very important. It reflects the real world experience of real people.
” In other words, Microsoft and Apple don’t just sit around to wait for Linux to catch up.”
No, but they aren’t really making big advancments, now are they? It’s been 5 years since a new Windows came out, and the changes to Vista are all rather ho hum (many already implemented in the open source world). After MacOS X came out (which was a big improvement) MacOS X has only become more refined, but nothing breathtaking.
And in my experience, KDE and Gnome on Linux have been better desktops for me than either Windows (which I’ve used extensively) or Mac (which admitadely I haven’t used much). KDE has features that are way beyond both Windows and Mac. Gnome is easier to use than all of them.
But in any event, the impovements to Linux, and the gradual chipping away I was refering to had to do with improving standards/consistency (making Linux an easier platform for both hardware manufacturers and software developers to target), and gaining marketshare. Windows can’t improve in this category, since they already dominate it. Mac won’t improve in this category since they target specific hardware. Linux has a lot of opportunity here.
Desktop Linux is excellent, it just needs better, more consistent standards, and better marketing. It will continue to grow, and chip away at the MS monopoly. It won’t be easy, and it won’t happen over night. But happen it will….
I’ve personally had tons of very aggrivating, time wasting, ridiculous, inexcusible (like you can’t believe this product has this problem) problems with Windows. This even though so many posters on OSNews say Windows is a great desktop and that Linux is a terrible desktop.
In short, anecdotal evidence is actually very very important. It reflects the real world experience of real people.
No, in short, it’s not important because you just don’t know enough people for it to be statistically significant when talking about markets. Some people that get on forums would like to believe that what they say is so important that some trend will follow, but 99.9999999999999% of the time it’s not.
No, but they aren’t really making big advancments, now are they? It’s been 5 years since a new Windows came out, and the changes to Vista are all rather ho hum (many already implemented in the open source world). After MacOS X came out (which was a big improvement) MacOS X has only become more refined, but nothing breathtaking.
You put your foot in your mouth again and fell into the same old “Microsoft and Apple are in static environments and Linux and Gnome and KDE are in dynamic environments” trap. I can say that the changes to KDE and Gnome in the past 5 years have been ho-hum and you would probably throw out words like FUD and Windows troll and all sorts of nonsense. But the fact remains that Microsoft has not been sitting on their hands for the past 5 years. They’ve got a whole new graphics system for Vista, .NET will be somewhat integrated, among other things. So you can claim that Vista will be much ado about nothing, but in reality we know that’s not true.
And in my experience, KDE and Gnome on Linux have been better desktops for me than either Windows (which I’ve used extensively) or Mac (which admitadely I haven’t used much). KDE has features that are way beyond both Windows and Mac. Gnome is easier to use than all of them.
And once again, your experience is meaningless to the discussion about what it takes for the Gnome or KDE or something else on top of Linux to break through. And some of the problems of desktop linux aren’t even related to desktop code.
But in any event, the impovements to Linux, and the gradual chipping away I was refering to had to do with improving standards/consistency (making Linux an easier platform for both hardware manufacturers and software developers to target), and gaining marketshare. Windows can’t improve in this category, since they already dominate it. Mac won’t improve in this category since they target specific hardware. Linux has a lot of opportunity here.
Microsoft already dominates, but that doesn’t mean that the status quo will not be maintained. You can wish for a lot of things, but they don’t necessarily come true. Any predictions of Linux gaining significantly in the desktop market is just pure speculation. I’m not saying it can’t happen, but I’m saying there’s a lot of crap that has gone on that has hurt Linux’s chances.
Stop being such a lamer.
If I knew 0 other linux users 10 years ago, and I know a whole bunch of them now, that clearly indicates that the linux user base is growing. The fact that you meet unrelated people who also know of Linux users, when you previously never did also indicates that there is growth. You can’t deny that going from 0 to 10 or more without actually doing some active searching for them doesn’t indicate anything. The only thing your reasoning indicates with accuracy is that you have stuck your head firmly in the sand. People are sick and tired of Windows and the problems related to it. As alternatives get more viable they jump ship. That is fact. It begins with the more technically skilled people, and proceeds from there as the alternatives get viable for less knowledgable users.
Also, as JeffS points out, personal anecdotes beat FUD, assumptions and factual errors any day, since those anecdotes is based on reality, as opposed to the common myth that you have to use the commandline to use Linux.
The whole point is that the linux desktop only gets better, continously. And users that find Windows adequate for their needs today aren’t giving a hoot in hell about Vista and the fluff Microsoft puts in there. They want something that is cheap, reliable and does the job. You forget about the magical phrase “Good enough”. If you’ve got two products where one is filled with fluff you don’t really need and is significantly more expensive than the cheaper but adequate alternative, it will lose on a free market, always. The evidence of this is plentyful.
If I knew 0 other linux users 10 years ago, and I know a whole bunch of them now, that clearly indicates that the linux user base is growing.
Maybe it means you met new people who use linux? There is no direct correlation based on such a small sample size, sorry.
Note the word indicate, as oposed to proves. Anyhow, it certainly doesn’t prove that the amount of users is notgrowing..
It certainly doesn’t prove anything either way. I don’t think indicate works either, because that implies some proof, even if not concrete.
But obviously, this is all kind of irrelevant, since we all already knew that that the number of desktop linux users has grown.
Actually it is kind of relevant, since the guy I replied to was trying to dismiss the fact that the last 1-1.5 years I and a lot of other people suddenly noticed a lot more Linux users around, as an “statistic anomaly”. Now, if it were only I who had this experience it might be some flux, like me moving to Linux-city or something. But when you suddenly get a lot of people who all tell the same story, either they are all lying (not plausible), or there is something going on. And that something is hardly a “statistic anomaly”.
Windows is already having, officially, 95% or so of the desktop market and frankly doesn’t have a lot more room to grow in, so it’s pretty stagnant. We all know that apple isn’t going anywhere, and that Mac users are probably less prone to drop their platform than a Windows user. So you tell me where all these new Linux users come from? “Statistical insignificance”?
What I meant is, us arguing about whether or not one person knowing more linux users means anything, is irrelevant. We KNOW there are more linux users, but that does not automatically mean that this is the cause of that one person (you in this case) knowing more linux users. Likely? Maybe. But you can’t assume a correlation.
In my opinion, the number one reason why Linux will never become mainstream on the desktop is development, testing and support costs.
We currently live in a world dominated by MS Windows operating systems. This is a nice place to be for software businesses like ours because it limits the number of different configurations, libraries, development tools, compilers and design philosophies. However, even within that nice and clean environment, there exists already too many different platforms and installation configurations. Having in the past worked on Irix, Solaris and Linux OSes, I can say that supporting theses OSes would require extensive testing, support and developement costs. In a competitive market where profit margins are low, having one more OS and it’s subsequent revisions/patches to support would rapidly become overwhelming to any small businesses.
Currently, we already support at least a dozen of different MS windows configurations and service packs. Putting Linux in the pack would at least add four new distributions plus at least two or three main versions of the kernel. What is already a testing burden would rapidly become a money pit nightmare.
“We currently live in a world dominated by MS Windows operating systems. This is a nice place to be for software businesses like ours because it limits the number of different configurations, libraries, development tools, compilers and design philosophies. However, even within that nice and clean environment, there exists already too many different platforms and installation configurations. Having in the past worked on Irix, Solaris and Linux OSes, I can say that supporting theses OSes would require extensive testing, support and developement costs. In a competitive market where profit margins are low, having one more OS and it’s subsequent revisions/patches to support would rapidly become overwhelming to any small businesses. “
Try using cross platform tools like Java/Swing/SWT or C++/QT. Development is easy, and you write to just one API. With the Java solution, you compile once, then run everywhere (with some testing, of course). With C++/QT solution, you write once and compile everywhere.
Then you can deploy to all platforms (Windows – muliple version, Linux, Mac, BSD, Solaris, etc), and deal with only one code base and one API.
So the solution is not to go with the Windows platform only. That has proven, as you mentioned in your post, to be a moving target. Even .Net has constantly changed.
No, it’s much, much simpler to go with one consitent cross platform API and language/toolset, like Java (with either Swing or SWT, the latter uses native GUI APIs, and the former has improved immensly), or QT (which runs/compiles on all platforms, looks great, performs great, and is very easy to use).
Interesting comment..
Since our application is CPU bound, Java won’t do the math for us. However, I’ll take a look into C++/QT.. It’s always interesting to have different views on such common problems.
Have a nice day.
to ordinary users – 80% of the population it does not matter whats underneath – if it doesn’t work – it doesn’t work. Have any of you ever created a product to sell? It is a widely known fact that first impressions always count, if the users can’t get something to work, they will return it and say its broken. Users have lives – they are for a vast majority not hardcore techies, they have things to do, money to be made. An OS simply just has to work out of the box and be intuitive to the user
anon
peace
“to ordinary users – 80% of the population it does not matter whats underneath – if it doesn’t work – it doesn’t work. Have any of you ever created a product to sell? It is a widely known fact that first impressions always count, if the users can’t get something to work, they will return it and say its broken. Users have lives – they are for a vast majority not hardcore techies, they have things to do, money to be made. An OS simply just has to work out of the box and be intuitive to the user “
Very good point. I might add the question – what if ordinary users had to install windows, rather than always having it pre-installed on the PC they bought?
The answer is they would encounter stuff that they would not know what to do with, and a lot of stuff would be broken, and they would have to manually install drivers (something that is beyond what ordinary users are willing to do and/or capable of doing). Thus, the same type of thing would happen with Windows that currently happens with Linux.
Then I’ll add the question: What if many PCs had Linux pre-installed?
The answer is – everything would work right out of the box. This is already the case with the Linspire PC you can get at Frys (as my previous post pointed out with my Mom’s experience). This is already the case with the Linux PCs you can get at WalMart.com, same with many other online sellers.
Yes, the vast majority of users are non-technical and just want stuff to work. But with Linux, since the major OEMs are too spineless to package Linux with their PCs (for fear of punishment from MS), Linux quite often has to be bought and or downloaded separately, and installed by the user. To even somewhat technical oriented people, this is trivial. To non-tech people, it’s a no-go. The same would be true of Windows, if it did not have the luxury of always being pre-installed.
This is perhaps the biggest obstacle desktop Linux faces. But with the OSDL’s Portland project trying to bring on more consistent standards, it will be easier and more cost effective for the Dells and HPs of the world to package Linux as an alternative (and perhaps gain a competitive edge, and perhaps tell MS to cater to their needs, instead of the other way around).
It’s going to happen, slowly but surely. Again, it won’t be easy, but it will happen.
Firefox provides a graphical installer and also:
http://www.wildgardenseed.com/Taj/autopackage/
There’s an autopackage for firefox.
If you want OOo 2.0, they have rpm’s available from their website. If you want to use these in a debian dist:
alien http://packages.debian.org/cgi-bin/search_packages.pl?searchon=name…
If you use an rpm distro, rpm. If you use slackware: rpm2tgz. rpm2tgz should actually work with any distribution, but uninstallation will be quite difficult if you don’t do it right:
Make a new directory, say /OOo
tar -xzf oo.tgz -C /OOo #ought to untar it to /OOo
Then, add these lines to your shell startup:
export PATH=”$PATH:/OOo/usr/bin” # or wherever the binary gets thrown for OOo (I don’t use OOo)
export LD_LIBRARY_PATH=”$LD_LIBRARY_PATH:/OOo/usr/lib”
Etc etc etc.
I think the OOo install package has changed since it went stable, maybe they’ve reincluded the GUI installer? It was always nice.
http://ubuntuforums.org/showthread.php?t=83308
There are OOo 2 install instructions for Ubuntu.
Final step: Quit whining; OOo 2 isn’t even very good
.
I do believe Linus has promised not to do this
. And for good reasons:
1.) I doubt he has time, making distributions is more work than most people realize.
2.) It’s probably not something he enjoys doing.
3.) It’s just cause schism’s in the communities.
4.) It probably wouldn’t be as good anyway
.
5.) Linux (proper) would suffer.
What if every time a peripheral or a piece of hardware didn’t have support from its manufacturer, one would e-mail the manufacturer, asking for a linux driver, and if none exists, ask if they could cooperate with the linux group to develop one.
If instead of quietly grumbling each time something like this happens, we make a point of following up on our opinions and take them directly to the manufacturer, perhaps we’ll start to see some changes.
My brother did this for his NGS Robbie2.0 webcam. He e-mailed the manufacturer, asking if they had a linux driver, and if not, if they could either provide the specifications of their hardware, or the name of the chipset used. They gave out the chipset name. After searching that chipset, we eventually found in it’s fabricator’s home page a 30-page pdf file which, to my layman eyes, contain the necessary information for a driver to be written for it.
What can we do from this point forward? What’s your opinion? Maybe tell this story in one of the kernel mailing list?