Linux (Still) Sucks Video

Linux Fest Northwest 2010 wrapped up less than 24 hours ago, and it was (as usual) a great show.  The full Jupiter Broadcasting crew was there and had an absolute blast.

On the first day of the show I gave an updated version of last year’s “Why Linux Sucks.  (And What We Can Do To Fix It)” session that I know many of you wanted to see (the room got pretty crowded, and I know many couldn’t make it inside).  So here it is, in all its glory…

Why Linux (Still) Sucks.  (And What We Can Do To Fix It).

And here are the slides from the presentation (in Open Office format, of course!).

For those interested, we also talked about Linux Fest NW in the latest episode of the Linux Action Show.

Ogg and h.264 versions of the video can be downloaded from the Jupiter Broadcasting page as well.

Share Button

82 Responses to “Linux (Still) Sucks Video”

  1. Mike-Linux-NL says:

    And still we have not come further than last year. Bryan is right. we need to standardize and organize.
    People also need to willing to support (pay for) opensource software. Only then Linux will become a major competitor and alternative to windows and OSX.

    If you as linux enthusiast want quality for your desktop, then help the developers to fund their linux projects. Developers need to eat too, and they cannot code for free, alltough every linux user expects them to do. If the user is willing to pay up, even though a small amount, linux will have much better software than what we have now.

    On the packaging side of things, we surely need to standardize to 1 or 2 formats. i suggest .Deb and .RPM should be the major ones. Software in those package formats should be released simultaneously.

    On the drivers side, we need to have better support, even from the major hardware vendors, like intel, nvidia and Ati. Also notebook and netbook support have to get a lot better, and should be tested well before releasing.

    As for games… lets hope STEAM comes out for linux, and with games that are newer than the games we have on linux for the past 10 years. nexuiz, open arena, WOP, etc are old…we neeed new quality games for linux, and we need to be wiiling to pay for it.

    The Linux OS you run on your desktop, mostly doesnt cost anything. compared to windows, or OSX we save about 70$. lets donate half of that (35$) to linux projects, in order to make it better.

  2. me262 says:

    One of the more interesting problems I’ve seen with pirating on Linux is in the games area. Specifically copy protection. This is one of the big problems facing it is that everyone under the sun is pirating games, not because they have the know how, it’s because there is real method of CD authenticity verification, and I don’t really know offhand of any that exist.

    A case in point, LGF released a copy of a game on torrent sites that had a specific flaw in it. They got more hits on this flaw than actual sales!

    A way to verify a CD’s authenticity does need to be researched and/or implemented, which will prevent mom and pop from copying it, and leaving it the hacker community and people who know how to break it.

  3. anon says:

    wheres the link?

  4. Bryan says:

    anon: The link to what?

  5. mugginz says:

    Mike-Linux-NL says
    “The Linux OS you run on your desktop, mostly doesnt cost anything. compared to windows, or OSX we save about 70$. lets donate half of that (35$) to linux projects, in order to make it better.”

    If only 50% of Linux users set aside $20 each to donate across various projects then you’d probably get a bit more developer interest. It’s what I’m going to do.

  6. Michael says:

    I know that Bryan was trying to avoid the whole marketing issue, but seems to me that much of this revolves around marketing. Communicating to the community at large that the developers need to be funded. Communicating to the community what software is available. Bryan used the example of his comic book reader that he got great sales on until he exhausted the market the LAS affords him. Lack of marketing. We as a community need to get the word out that linux is still growing and a viable operating system choice, plus we need to explain to the community that they need to throw a few bucks in somebodies direction to continue development. maybe each distro could have one nag screen on first start up? or during install? but we need to get the word out on these issues and that is marketing.

  7. .net jerkface says:

    Hey Lunduke you no longer look like a Mac hipster douchebag. The short haircut really works for you. That is the only improvement I see from last year.

    Since you are a programmer who has lived by your own hands and mind how can you support all these open source ideologists who want to reduce a programmer’s salary to tips in a fucking jar? At least a waiter gets paid an hourly salary so he still gets something if the customer skimps on the tip.

    Why do you even bother with these people? They’re against closed source and yet collectively they can’t duplicate a tenth of the work of closed source developers. Fuck them. Let them wallow in broken video drivers, they deserve it.

  8. factotum218 says:

    Sorry, pointless question, I know. Did you shave your head for the occasion?

  9. Junior says:

    I propose we create a website where we as a community could go and vote for projects to get funding and then everyone that pledges money will have their money split up to the top five voted projects. For example, a donor will go to supportlinux.com (or something) and vote for their top five projects which are:

    1. Suspend/Resume Fix
    2. Video Editor 1
    2. Linux Sound Editor 1
    3. MediaPlayer X
    5. TotallyAwesome Program Y

    This person then pledges to donate $10 a month (or $120 at once) and that money will be split up between the projects. This would work better if we all get together, as a team, as a cohesive group and pledge our monies together to support some badly needed projects (like the sound system) so that these projects could hire the developers they need to do the work that is needed. If we as a community set a date (say every six months) to do this and market it across the social web, we could come together and sponsor (really sponsor) development. As stated in the talk, dropping $20 here and there on obscure projects doesn’t make a huge difference, but if we do it all at once, we could make linux a much better platform than it already is.

  10. SlickMcRunFast says:

    Bryan, I agree with your thoughts on games.

    Why don’t distros focus one month a year on making their disto fun. Make the user happy and give the devs a break.

  11. Justin B. Burris says:

    Bryan, have you seen some of the rumors revolving around Steam and Linux? According to some sites, http://www.devicemag.com/2010/04/23/valves-steam-coming-to-linux-officially-confirmed/ people discovered linux download tools on the steam website. They are no longer there, afaik, but as of the articles debut, it was apparently in the works. This could all be one depressing rumor, but it’s been reported across multiple gaming sites, so there could be some credibility.

    Just something to watch.

  12. WeasleX says:

    Well, I’ve said it on Jupiter, I’ll say it here. Linux doesn’t need games, they need game designers. If anyone who’s anyone has ever done this, you’ll know there are different levels of difficulty. If you just make a simple eventer Sprite Pixel based system, it’s pretty easy. If you try and build a futuristic all in one 3D game designer, that’s going to be pretty hard. I think ASCI/Enterbrain made a few that might work for Linux, but I’m not sure. Look up RM2K3 or XP or VX, and if they aren’t, try and get them for Linux. They are great, easy to use, cheap ($30 each I think) and they run great.

    They aren’t difficult to learn how to use, in fact I picked up on the basic’s within a month, and more advanced eventing/scripting in 3 months, so, it was really easy. Like I said, try and get them in Linux if they aren’t yet. Then see if you can’t sell your works, and sell them well.

  13. Dan Douglas says:

    If all you care about is the quality of a few specific niche programs then I see no technical reason that Linux in itself is superior to any other OS. If I wanted to use proprietary software developed under a commercial model then I would use Mac OS or similar. I use Linux because I do not want to go to download.com to install some crap freeware that the user will probably never update unless they manually check for updates from the author. Linux works in large part because of the work of packagers which integrate the efforts of upstream developers in a way that doesn’t happen on commercial platforms where developers work in a bubble – dumping their code on to the platform without regard for interoperability because the goal of selling copies of software to end users is in many ways counter to the development model that makes free software so successful. You don’t see teams of organized developers working in GIT or SVN along with packagers at the distro level enforcing certain standards such as preventing unnecessary bundled libs etc in the proprietary world. Download.com etc is just a bunch of independent developers who spew tons of useless unmaintained garbage freeware/adware utilities that nobody on Linux would touch with a 10 foot pole. I want to search my package db, install powerful serious tools, and be free to use, study and tweak them with no proprietary bullshit getting in the way.

    It’s funny how in practically the same breath someone can talk about the lack of quality free video drivers resulting from vendor lock in and trade secrets, and at the same time promote the spread of the very software that hinders progress in those areas. That’s just insanity. For instance you might complain about Flash sucking on Linux. The solution isn’t to pay Adobe to improve flash. The problem is caused by the proprietary Flash being so ubiquitous. Free/open standards like html5/ogg need to replace those technologies.

    If what you’re really trying to say is that it would be nice to have the option of using Linux ports of proprietary software such as Adobe Photoshop on the occasion that no satisfactory free alternative yet exists, that I can understand and is a sentiment that many share. I don’t really see that as a particularly catastrophic or long-term problem though. The best way to develop that kind of software IMHO is as you mentioned to create commercial interest in it’s development and support. Selling software to end-users just sucks on every possible level and causes nothing but problems – there are so many other and better ways to make money in software as has been discovered by many of the companies we admire such as Novell.

    It’s also interesting that you lump Gentoo’s ebuilds in with RPMs. Ebuilds are certainly not an ideal “package” specification but a large part of the problem in that area I think has more to do with bad build systems (cough autotools cough) than the lack of a standardized package format. Who really cares about binary package formats? Packages would be really easy to create across formats if there were a better more standardized way to change the way software is built for specific platforms.

    In summary the current way that software is developed and used in the free software community is the primary reason that I use this platform. If you took that away then I would see little reason in continuing to use it.

  14. factotum218 says:

    Okay now that I’ve actually watched the video…

    The idea of porting more commercial software to the Linux based platform is starting to get thrown around more and more, especially in the last year or so. Myself included.

    I’ve worked in graphic design for the last decade and I would be more than happy to stand in line for a copy of CS7 to run on my x64 desktop for the same amount of money as the other platforms. Same with games. Who knows, native commercial apps could maybe (huge maybe) make people more aware of needed funding for other projects in the open world. I know just the discussion of commercial commerce has gotten at least a few more people to entertain the idea of supporting their platform financially.

    I don’t use Linux at home out of politics. I could care less about sticking anything to anyone. I went through my punker anti-establishment phase in my late teens and early 20’s listening to Profane Existance releases and all that. But it was just a farce in reality. In a nutshell I don’t want to take my work home with me. I don’t want to sit down in front of a Mac when I get home. I’ve spent 40-60 hours a week in front of a Mac since OS 7 and I loathe it.

    I like Linux. It’s a big ball of quirky fun with the perk of efficientcy and stability if I so wish. If I could sit down with it and be just as productive and plug in to the workflow at the pixel plant I would. But I can’t, so Adobe doesn’t get my money anymore. Not since CS2 anyways (Hi WINE!)

    Sorry, i know this jumped all over the place and there are a lot of spelling and grammar errors, but it’s 3am and it’s all Bryan’s fault.

  15. Riku says:

    I have to comment on the package management issue. Apt-get and .deb are quite nice, but they have their problems. RPM’s used to be the Linux equivalent of DLL hell, but that might be fixed since I last stumbled into RPM’s.

    However, there are several practical scenarios where deb & rpm-based solutions don’t work. Unprivilidged user mode install (install to home directory), co-existance of multiple versions of the same software (esp. libraries) and non-standard install destinations are examples of features that should be easier (or possible) to handle.

    Some of these problems are solved by newer (and less used) package management software, such as Nix, the package manager used in the NixOS distro.

    I wish there were a coordinated and well-funded effort to design and implement a package management system that could handle most of the practical issues involved in distributing complex software with dependencies.

  16. Nero says:

    Brian, can you upload the video in Theora? Thanks!

  17. smspillaz says:

    I really hate to yell this but

    GSTREAMER IS NOT AN AUDIO FRAMEWORK.

    gstreamer is a (you guessed it) media streaming interface for linux. When I say streaming, I mean it in the sense of the word that you stream data from one pipe-end to another, for example, a source file to a decoder to a demuxer etc.

    People also need to stop bashing PulseAudio for being fundamentally broken in some way – that “linux audio” diagram might be famous, but it also shows a shit-ton of deprecated audio frameworks that hardly anyone ever uses. The fact is that about 90%+ of all desktop linux applications use the Pulse->ALSA->Hardware framework. The biggest application that uses JACK is Ardour, and that is because JACK was designed with the low-latency requirements of Ardour in mind. And when RealtimeKit actually hits the distros and PulseAudio starts to seriously adopt it then you will see that any apparent latency (which makes audio production really annoying) is going to skyrocket down, since you’ll be using the CPU in real-time to get as much sound to the sound card as possible.

    PulseAudio is actually one of the most advanced audio frameworks that exists in the world of computing. Imagine it as an X.org for audio – it abstracts all the quirks in audio frameworks which are involved in talking to all kinds of devices (think bluetooth audio) into a single mixer and allowing things like per-client volume control (which gets very interesting indeed when you consider things like Ear Candy).

    On the issue of X.org, you’ll usually find that if you run from X.org git master like me, the quality of the drivers is actually quite good. Developers never ever EVER check in code with severe regressions and/or broken code into master (unless they have a really really really good reason too). Usually apparantly broken drivers are a result of including things like the wrong libdri + driver or the wrong xserver + driver combination in your distro, which results in it taking slow rendering paths for things that are easily fast otherwise.

  18. SJWoodChipper says:

    Considering that I was finally able to hear this on my Windows XP box, I find it ironic that the first thing the guy mentions is that Linux sucks because come 2010, they still haven’t sorted out audio.

    It’s simply unreasonable to expect desktop end-users to sort out all this audio playback stuff when we (speaking for all desktop users) simply want it to WORK.

    You could basically say that about 100 of other things (many of which listed in this video that you can hear on Windows without having to learn everything about computer audio just to listen to something) which prevent me from ever recommending Linux to anyone who would otherwise buy a Windows box.

  19. Silvio says:

    Nice work man! I like your concepts and think you’re damn right about how to make Linux desktop successful. I feel we need leadership to standardize the platform. But who could bring this leadership? Now that Ubuntu is becoming increasingly successful I feel like Canonical should be able to bring this leadership and impose standards and lead the development of the Linux Desktop. Ubuntu has taken some long fast strides. Do you think they should focus some more on imposing standards?

  20. The Leopard Print Couch Company | Piestar says:

    […] One year on, same presentation, absolutely no significant change what-so-ever. His hearts in the right place but I really don’t think that even if the community did exactly what he said it would really change a thing. […]

  21. Keith Eberle says:

    once flattr (http://flattr.com/) goes live, it’ll be another sweet method to donate money to projects that you care about.

  22. Eric says:

    Excellent presentation. I apologize, but I didn’t have the hour to spend to listen, although I would have liked to.

    I’m a computer science major with a concentration in professional computing. And it embarrasses me to say that the only real experience I had with Linux outside of using it to code in started this February once my brother wanted me to refurbish my old laptop for him.

    I’ve been using it for about 3 months messing around with it, and the biggest thing Linux has against it is a lack of standards. There are so many different standardization companies out there (ISO, IEEE) but there is no Linux standards. It makes everything such a pain in the butt.

    In contrast, it took me less than a month to get used to OS X (I’m also quite the graphic artist, if I say so myself, and used Mac’s for my design classes). But between hardware issues, software issues, random issues (like random shutdowns while running DeVeDe) I cannot recommend Linux to anyone who wants anything more than a Word Processor and Firefox.

    Not only that, but support is very limited to none (at least for Ubuntu). Got a problem, post it on the forums or IRC. No one will get back to you, or you’re a “#%%&ing n00b”. Like I said, I’m a Comp Sci student and I don’t know the intricacies of Linux. Therefore I’m a totally loser in the minds of the Ubuntu forums. Because, you know, there is nothing more to Computer Science than coding in a million different languages and knowing every platform imaginable by heart (I’m way more into software engineering, UX design, and automatia).

    There aren’t even any good websites that I’ve found that helps even a moderately computer savvy person to use Linux. It’s always with these crazy terminal prompts and things that normal people can’t understand. Get off your high-horse Linux users!

    That said, I hope the documentation for 10.04 is at the very least usable. It’s sorely needed. I’m sorry, but several forums, websites and Wikis written in techno-babble does not make Linux (Ubuntu in particular) user-friendly.

    I’ll still place Linux on top of OS X in terms of preference (because I’m a geek who loves to %$$@ around with the inner workings of the Kernel and learn new things and OS X has been simplified to the point where even basic functions are ridiculously complicated…try opening Safari in full-screen with the ‘+’ key, it won’t go full screen…also, pressing delete on a file prompts for a file rename? WTF?), but it has a far, far way to go before it can dig itself out of this go-nowhere cycle.

    My favorite slide was that Linux feels like you’re trapped in the late 90s. It’s very true. And what I liked more is that you have answers, which is more than I do.

    Thanks for the good read.

  23. Johan says:

    I felt compelled by your shouting about paying for software. So I payed for the good ole game “Uplink” from Introversion. I just love feeling like a cyber-badass.

  24. Joe says:

    There should be some kind of standard Linux package repository that all the major distros can pull from. I think all the different package and repository formats retarded things in Linux right now.

  25. Michael says:

    When Canonical starts offering pay for software in their Software Store, they should also be offering the ability to Donate to a open source projects they offer in their store. Basically they should take the role of paypal that most Free Software developers use for collecting donations, allow the developers to set up the payment mechanism on the projects launchpad account and then integrate the ability to set up one off or standing donations directly into the “Software Store” application. I would be really dissappointed if Canonical gives the ability for proprietary software to make money from their store and deny the free software projects in their store from any revenue by donation.

  26. eigentum says:

    I think its really a strange Idea to say that closed source Software will save the open source world. I mean it’s fine to write what ever programs for Linux. But the thing we love about Linux is the openness. The Way we can tinker with the System. If your want closed source, commercially available applications there are Windows, OSX and iPhoneOS. Why do we also need linux for that job? What’s the advantage?
    His proposal is nothing new. It’s the same Bill Gates wrote in his famous letter claiming the only way for Programmers to make a living is through commercially closed source software. Personally I don’t like Linux to go down that same rode again some 30 Years later.
    Yes there are Problems, but I think his ideas are really boring and don’t help to fix them

  27. Nathan Hartwell says:

    What if we had something close to a contest to fix certain problems in apps. So say “Hey, you will get $1,000 if you fix the audio problems here!”. And so you pick the winner, add on to that, and fix things, and you also have 100 other ideas from 100 other people….. might work?

  28. Michael says:

    @Nathan Hartwell – the problem with what you suggest, which are known as bounties and are already available are outlined here, basically the bounties offered are nearly always far too low for the amount of work involved http://ardour.org/node/2674

  29. .net jerkface says:

    @Dan Douglas

    “Linux works in large part because of the work of packagers which integrate the efforts of upstream developers in a way that doesn’t happen on commercial platforms where developers work in a bubble – dumping their code on to the platform without regard for interoperability ”

    Oh you have to be kidding me. Linux application developers care about interoperability? Is that why most of them use GTK+ and only build for a single distro? In Linux you don’t even have interoperability between distros. That’s why package maintainers exist, to collect source and then re-build it for their respective distros. If there was interoperability then Linux developers could compile once and then have their application be deployable on any distro.

    Most Linux developers aren’t even making use of open source toolkits that encourage interoperability. GTK+ is lousy for porting to Windows and OS X compared to Qt. Most Linux developers use GTK+ and don’t care if works on other platforms.

    “Selling software to end-users just sucks on every possible level and causes nothing but problems – there are so many other and better ways to make money in software as has been discovered by many of the companies we admire such as Novell.”

    That model doesn’t work for consumer software. How many people would buy a $100 support contract for Photoshop? Especially when there is so much free help online? Who would buy a support contract for a game that they will beat in a month?

    Only on planet loon can you claim that selling software to end-users sucks. I just bought Final Fantasy on the iPhone and it was the best under $10 purchase I have made all month. I spent $4 on a smoothie today so paying $7 for a classic turn based strategy game was a steal. I was also happy to support a company that works hard to make games that I love. It was a mutually positive transaction and only a loony Stallmanologist would take issue with it over Square-Enix not providing the source. I don’t expect the source when I buy software. I’m buying it to use it just like 99.99% of the population. I look at source all day long, I could care less about having the source to a freaking iphone video game. And at $7 I don’t expect them to hand over their R&D.

    “In summary the current way that software is developed and used in the free software community is the primary reason that I use this platform. If you took that away then I would see little reason in continuing to use it.”

    I think that is a good point that Lunduke should note. What he and sane people see as problems you see as benefits. Lunduke should probably ditch Linux and go back to the Mac where sane people are far more common. Or move to Android development at least where people don’t have issues over buying fucking 5 dollar software.

  30. danni says:

    People still try to push Linux as a desktop system? I gave up like ten years ago. Issues were not solved then, they’ll never be.

  31. Dan Douglas says:

    @.net jerkface

    “Oh you have to be kidding me. Linux application developers care about interoperability? Is that why most of them use GTK+ and only build for a single distro? In Linux you don’t even have interoperability between distros. That’s why package maintainers exist, to collect source and then re-build it for their respective distros. If there was interoperability then Linux developers could compile once and then have their application be deployable on any distro.

    Most Linux developers aren’t even making use of open source toolkits that encourage interoperability. GTK+ is lousy for porting to Windows and OS X compared to Qt. Most Linux developers use GTK+ and don’t care if works on other platforms.”

    Yes I agree, but don’t see how this conflicts with what I stated previously. The large collection of software that is this OS works well in large part due to the efforts of packagers and sometimes the general properties of unix such as the notion of programs as small composible functions etc. Interoperability between distros isn’t necessarily the main issue since they are essentially made up from that same collection of software with the main difference being the package manager. The difficulty is as I mentioned in the packaging itself, or more to the point in the build systems. If you’ve ever had to redo someone’s awful makefiles with hardcoded paths and variables you’d quickly realize this.

    “That model doesn’t work for consumer software. How many people would buy a $100 support contract for Photoshop? Especially when there is so much free help online? Who would buy a support contract for a game that they will beat in a month?

    Only on planet loon can you claim that selling software to end-users sucks. I just bought Final Fantasy on the iPhone and it was the best under $10 purchase I have made all month. I spent $4 on a smoothie today so paying $7 for a classic turn based strategy game was a steal. I was also happy to support a company that works hard to make games that I love. It was a mutually positive transaction and only a loony Stallmanologist would take issue with it over Square-Enix not providing the source. I don’t expect the source when I buy software. I’m buying it to use it just like 99.99% of the population. I look at source all day long, I could care less about having the source to a freaking iphone video game. And at $7 I don’t expect them to hand over their R&D.”

    The core point here is I think most of us realize there is a fundamental conflict between the notions of packaging up units of data as a commodity, and producing powerful maintainable software that’s useful rather than harmful due to it’s closed nature. I don’t think one has to be a Stallman Nazi to acknowledge the persistent boon that is ubiquitous yet unmaintained crap that nobody can fix and will not die such as IE and Flash. I do fully realize the difficulty in producing certain types of software through open-source means, but we’re talking about two entirely different things here. You don’t rely upon the ability to update and extend your iphone games. There is no major requirement for interoperability there or adherence to some standard. I’m also not against buying or selling that kind of software, supporting people who develop it, or porting it to other OSes. I don’t think it’s insane though to say that the traditional commercial method of producing software is non-ideal at the very least, and has the potential to create some dangerous beasts despite the good intentions of their developers.

    “I think that is a good point that Lunduke should note. What he and sane people see as problems you see as benefits. Lunduke should probably ditch Linux and go back to the Mac where sane people are far more common. Or move to Android development at least where people don’t have issues over buying fucking 5 dollar software.”

    Don’t let the door hit ya on the way out. I’ve no interest in fucking your 5 dollar software, or buying your 5 dollar fucking software, or…

  32. .net jerkface says:

    “Yes I agree, but don’t see how this conflicts with what I stated previously.”

    You stated that Linux application developers do not just dump their code on their platform without regard for interoperability. I see no evidence of this. I see most Linux developers building for a single distro and Windowing system without any regard for how it works on other distros or if it will work in Windows or OS X.

    “I don’t think it’s insane though to say that the traditional commercial method of producing software is non-ideal at the very least”

    I think it is insane that Stallmanologists take offense to mutually beneficial software transactions that involve closed source. If I create software for a business that saves them money according to Stallmanology I’m being unethical if I don’t provide the source, even if providing the source means nothing to the buyer and would only undercut my business by giving an advantage to my competitors. That’s insane. You can call proprietary development non-ideal but that is your subjective opinion while it is not subjective that the proprietary method has produced far more software.

    “I don’t think one has to be a Stallman Nazi to acknowledge the persistent boon that is ubiquitous yet unmaintained crap that nobody can fix and will not die such as IE and Flash.”

    There is still no open source alternative to Flash while IE6 should not be fixed but dumped and is indicative of businesses that resist upgrading software for a variety of reasons. If you think that web browsers need to be open source to follow W3C standards then have a look at Opera.

    “Don’t let the door hit ya on the way out. I’ve no interest in fucking your 5 dollar software, or buying your 5 dollar fucking software, or…”

    I left Linux quite a while ago and it had nothing to do with money. My main issue was with the Linux status quo that kept defending the Linux/Unix way even when it clearly needed to be reformed if it wanted to expand its userbase. Linux is a mess from a software engineering point of view and is filled with silly handshakes that only exist to stroke the egos of members. I couldn’t be a part of a system where illogical design decisions were defended on the basis of them being part of the Linux/Unix way. That’s a silly handshake club, not software engineering.

  33. Linux (Still) Sucks, and What We Can Do To Fix It. « Linuxgeek1 Blog says:

    […] (Still) Sucks, and What We Can Do To Fix It. April 29, 2010 — linuxgeek1 This is a presentation by Bryan Lundake about the present short comings of Linux in general and what is needed to move […]

  34. chris says:

    Windows 7 is really nice. OS X is okay but I think they lose this round. Linux has absolutely no compelling reason for me to switch to it, in fact many things I currently do would be impossible under linux, and a lot of the usability enhancements from win7 (that i use every day at work) would be inaccessible. I use stuff like visual studio, newsleecher, photoshop, quicken, ableton, FL studio, reaper, stuff that there is no linux equivalent that comes even close. I use more advanced features of MS Office, like macros, app extensions and outlook scripts. Switching to linux (again) would just be headaches from getting various bits of the OS working right and hours upon hours trying to find replacements for software that I already have and am comfortable with. Any replacements for those abovementionned programs would be way way underfeatured (speaking from personal experience). I am a consumer and I have no interest in lunatic fringe ideologies, my computer is just a tool and I want to make the most of it. It’s not a political platform. On top of all that, switching to Linux means dealing with the notoriously hostile Linux community. No, not every member is an arrogant my-way-or-the-highway type of jerk, but nobody can deny that the most vocal elements of the community tend to be self-righteous, obnoxious and largely misinformed.

    So I use windows, because windows is good. I’ll probably never use linux in a meaningful role ever again. There really needs to be a compelling reason to switch, and I can’t think of any. I’m an IT professional, the exact target market for Linux, and I would not in good conscience recommend it to anyone. I have nothing against linux, I just think it’s more of a project than a product. Consumers aren’t interested in projects, at least not to the capacity of using them for their day-to-day.

    Kudos to .net jerkface, I read your blog from time to time. A bit blunt sometimes, but you tell it like it is.

  35. Errko says:

    How is an “appstore” going to help anything, if even Ubuntu breaks software compatibility every six months?

    You’re still going to have to roll packages for all the umpteen versions of Ubuntu and Debian that are being used all over the place, each with a different set of libraries of different versions and frameworks. It’s not only package management you have to standardize, it’s the entire OS.

  36. bryan is right says:

    just +1

  37. nert says:

    I wonder why no one really looks at LGP here. I already wondered while watching the video. It is the *de facto* go-to-point for Linux games. I know the their PHB and if all of you would stick to what Bryan said, he would be a very happy PHB. :-)

    http://www.linuxgamepublishing.com/ for your gaming pleasure. Oh and Bryan: extend your list of commercial games for Linux please. :-)

  38. Carlton Banks says:

    Bryan,

    What do you mean “maybe we’re at 5 percent”? Desktop Linux is < 1% market share. It has never been anywhere near 5 percent, even after 19 years of being given away for free.

  39. aaa says:

    “Bryan Lundake has done a presentation about the present short comings of Linux…”

    Bryan LUNDAKE lol

    http://linuxgeek1.wordpress.com/2010/04/29/linux-still-sucks-and-what-we-can-do-to-fix-it/

  40. .net jerkface says:

    @Errko

    You could have an app store in Ubuntu but you’d have to keep it independent of the packaging system. Apps would have to be built with their own libraries. You could keep the shared library system in place for all the major open source programs.

  41. Michael says:

    @ .net jerkface – That was partly correct and partly bollocks. The shared libraries thing is completely true, what is not true is that it would have to be independent of the packaging system, there is nothing stopping a packager from bundling all the libraries an application needs in the same deb as the application, and it working alongside the shared library system. As Bryan says in the video Debs are just a tar file and a script there is nothing intrinsic about the Debian package system that forces it to be used in that way.
    Sometimes I wonder whether a distribution would work better if they dropped the shared package system for unsupported applications altogether. In the last iteration of Ubuntu I couldn’t install both Calibre and Sabnzbd+ together because the CherryPy2 library conflicted with the CherryPy3 library, so when I installed the application the other was uninstalled with little warning, leaving my very confused.
    I disagree with Bryan that settling on Deb is the answer here I think in the long run we need to be able to have multiple versions of the same library able to be installed without conflict and associated with it’s particular application as the upstream developer intended it. Which will have to include a rethink of the entire folder layout aswell.

  42. stunt says:

    Just to pick up on the games discussion here…

    Bryan, you are wrong! Commercially ported games do exist under Linux. Just check happypenguin or LGP. It’s not nearly the amount of Windows or Mac games of course but they do exist. And LGP is the only real supplier. So with one thing you are dead right: people don’t know where to go. Or how else could you explain why you don’t refer to it during your presentation. Linux gaming is not just Doom and Penny Arcade. And once upon a time, there was Loki games… look it up!

    So what happened to Loki and what is currently happpening to LGP? Dead simple: nobody is buying their stuff! Well, I did and still do BUT most of the time they got/get bitched at by freetards who just don’t want to pay money for quality. End of the story, sad as it is.

  43. Anton says:

    Great persentation! But what about a quick retwitter button?

  44. .net jerkface says:

    “there is nothing stopping a packager from bundling all the libraries an application needs in the same deb as the application, and it working alongside the shared library system.”

    No there isn’t but that wouldn’t be enough for an app store. You would need the distro to be able to install/uninstall those applications cleanly and also have them tied to a server account. There also needs to be an encryption system for purchased apps. The package manager can’t handle this functionality which is why such a framework should be built as an application like Steam.

    “I disagree with Bryan that settling on Deb is the answer here I think in the long run we need to be able to have multiple versions of the same library able to be installed without conflict and associated with it’s particular application as the upstream developer intended it. Which will have to include a rethink of the entire folder layout as well.”

    I agree, I think the shared library system needs to be scrapped or a new framework at least needs to be built along side it. Standardizing around Deb would certainly eliminate a lot of redundant work but there would still be library conflicts. The other problem here is that the LSB is already built around RPM. Both the LSB and Red Hat would be resistant to dropping RPM. Then you have distros like Arch that don’t care about widespread adoption and will use their own format regardless of what everyone else thinks.

    For standardization to occur in Linuxland there needs to be one distro that pulls far ahead of the others and gains inertia, thus attracting developers who only build for that distro which then attracts more users and so forth. I remember Bryan had some graphs which show a migration towards Ubuntu so he may get his wish. The kicker is that proprietary software would really help speed this inertia since it can’t be easily ported to minor distros like Arch. Open source often works against itself by forbidding exclusives that would normally help built inertia around a common platform.

    Though I think Lunduke is wasting his time by suggesting that distros voluntarily standardize he still has a better understanding of software economics and why Linux has stagnated on the desktop compared to anyone else at that conference. His presentation should have been 2 hours with a mandatory quiz afterwards. Pass the quiz or you don’t get any lunch :)

  45. Peter says:

    Linux users also have to eat. Basically everyone has to eat.

  46. Dan Douglas says:

    @chris

    Many of your comments sound along the lines of what I might have assumed prior to taking a serious look at Linux. After several years of using it I would now say about the same as you with regards to Windows – I don’t consider it a reliable platform for serious work. All of those programs have Linux equivalents that suit my needs and tend to be especially useful when I care about scripting. I couldn’t live without Gimp’s python modules for instance, though they are of course also usable on Windows.

    I wonder why people think using Linux is all about ideology and assume that the tools available for the platform are less feature-rich because my personal experience has been the very opposite. To me it’s purely practical and whatever philosophies I take seriously have a practical purpose. I really have to call bullshit on many who claim it would be impossible to do their work on any of the major platforms. That only happens when you really need to use a specific application that isn’t ported to your system – which for me is always some stupid arbitrary reason that I can usually find a way around.

    If you’re comfortable with Windows tools and can’t justify switching, that’s fine and you should continue using it especially if your primary purpose is .net development targeted to windows. I’d be more than happy to help if you’re having trouble doing something on Linux. Knowing how and where to ask questions helps. I don’t run in to much hostility unless people whine or start getting off-topic and preachy. Sometimes I run in to problems, take a quick peek at the code, hop on IRC, report the problem, and it’s fixed in the repo within seconds. I find developers are much more friendly to those who submit patches or at least good bugs with backtraces and such, than to those who whine. That’s always been the case.

  47. Mohamed Elsherif says:

    Great presentation Bryan, I agree on almost everything you’ve said, especially the marketing part even though you preferred not to discuss it, I still argue that marketing issues are more important than technical issues

  48. Rootz.de > Bryan Lunduke: Why Desktop Linux (Still) Sucks says:

    […] Jahr hat Bryan mit Why Linux (Still) Sucks nochmal eines draufgelegt. Auch wenn die Titel der Vorträge stark nach Linux-Basing klingen, […]

  49. igor says:

    Did he lose a teeth?He sound like he did.

  50. Perpetuum Immobile says:

    I can’t wait for next year’s video :), I’ve marked it in my calendar!

  51. Errko says:

    @Dan Douglas

    It’s nice that you find use for the scripting in GIMP, but it’s an image editor, not a programming environment, and as such it really sucks deep. Half of the functionality isn’t there, and what is there is mostly crap. Like working with selections, which simply doesn’t work. For an artist, GIMP is clearly an inferior alternative to just about anything there is unless you’re hell-bent on working around its quirks. It’s a “me too!” application that doesn’t really do what it’s supposed to do. It kinda pretends.

    Most Linux software is like that. Usable in many ways, even useful in some ways, but not quite there when it comes to what it’s actually supposed to accomplish. If you can build your workflow with the stuff, more power for you, but many people can’t because the software doesn’t work right.

    It’s like taking a cheap knock-off swiss army knife, and finding that the blades are dull, the corkscrew splits corks and the scissors don’t cut anything, but at leas the phillips screwdriver head fits the intended purpose, so it’s useful as a screwdriver! Great! That’s all you’ll ever need anyhow!

  52. .net jerkface says:

    “I really have to call bullshit on many who claim it would be impossible to do their work on any of the major platforms. That only happens when you really need to use a specific application that isn’t ported to your system – which for me is always some stupid arbitrary reason that I can usually find a way around.”

    Not being able to run a specific application is a stupid arbitrary reason? What are you going to suggest here, that someone run Create Suite or Autocad in a VM? That’s a waste of resources, especially when laptops are popular. Wine is not a viable answer when for so many applications it provides incomplete support.

    “If you’re comfortable with Windows tools and can’t justify switching, that’s fine and you should continue using it especially if your primary purpose is .net development targeted to windows. I’d be more than happy to help if you’re having trouble doing something on Linux.”

    And that attitude is what drives people away from Linux. I was providing suggestions on how the package management system can be fixed and you’re condescendingly offering me help? Do you think the 1970’s era shared library system makes sense in a modern consumer OS? Do you realize that the shared library system was designed to save resources in an era when gigabyte hard drives didn’t exist? How dare I question design decisions made over 20 years ago when the computing environment was completely different. It must be that I need help from a Linux guru.

    I have no problem with writing software for Linux nor would any developer worth a damn. The problem with Linux is that the porting and support costs for proprietary software are far too high relative to marketshare. This is mainly due to incompatible distros, incompatibilities between versions and distribution systems designed around open source. This leads to poor application support which keeps users on other platforms. I personally don’t care if Linux stays at 1% forever but this is a discussion on how Linux can be fixed.

    These are your people Lunduke. Have fun with them.

    Oh and give Elysian Immortal a try if you like IPAs.

  53. Dan Douglas says:

    @Errko

    I’ve gotta agree with you there. I find selections on Gimp to be one of the things I like but that might just be because I’m used to them. Gimp lacks all the layer effects and properties too, I tend to end up using a lot of layer masks to and scripting to compensate. You can still do a lot of the same things as photoshop, but it does involve more work. Linux port of Photoshop would rock.

    @.net jerkface

    “Not being able to run a specific application is a stupid arbitrary reason? ”

    Absolutely. No sane person would switch from an OS they like and suits their needs most of the time because of a few corner cases.

    As for the rest of your post, it hardly deserves response… and ROFL at the irony of your “condescending” remark. None of those comments were directed at you anyway. I seriously cracked up when you started talking about legacy and shared libs in Linux versus the “modern consumer OS” that is Windows. Proper package management alleviates the DLL hell concerns. Your assertion that the primary motivation behind shared libs is disk space is ludicrous. PIC on Linux actually works, and there are significant advantages in terms of memory usage and load time especially with –as-needed or even prelinking. Perhaps someday we can go back to having some real hardware-based architecture innovation so our computing environment is NOT stuck where it was 20 years ago thanks to Microsoft. Take your obvious generic fanboy trolling elsewhere.

  54. stunt says:

    @Dan Douglas:
    What the hell are you talking about? Millions of people have to use OS they would rather not use because those “few corner cases” is usually the shit they need to get their job done / food on the table. That’s the reality. Deal with it!

    As a Linux-User I can understand (and agree with) .net jerkfaces remarks completely. These are issues. They need to be fixed. He doesn’t look by a fanboy troll at all. You, on the other hand, are just repeating the same old claims without backing anything up.

  55. WinUsr says:

    @Dan Douglas : “Absolutely. No sane person would switch from an OS they like and suits their needs most of the time because of a few corner cases.”

    You have to be quite an ignorant troll to make this comment. Outlook + Exchange, Outlook Web Access (Didn’t have a rich interface in non-IE browsers until fairly recently, and only Firefox). Office + Sharepoint. Dreamweave, Photoshop, Sony Vegas. Games like World of Warcraft, EverQuest II, etc.

    People use Windows because of those “corner cases” because there is no real reason to use Linux. Almost every productivity software package for Linux also runs on Windows. There is absolutely no reason to run it. You have to be a troll to think otherwise.

    Microsoft doesn’t have us stuck where we were 20 years ago. from 92 to 2001(1 decade) we went from Windows 3.0/3.1 to Windows XP. Where has Linux gone? It’s still a shitty desktop operating system with crappy application dependencies, not decent user interface guidelines, and an installer that installs basically dozens upon dozens of overlapping applications by default.

    It took Apple to make UNIX friendly on the desktop, and Linux still can’t seem to learn from a pretty fucking obvious example staring them in the face. If they keep on with this status quo, I think Solaris can be poised to overtake them. There is less in-fighting in the OpenSolaris community, and they seem more interested in advancing the platform than duplicating work and confusing new users with a gazillion (terrible) choices.

    And do you really think a platform based on userland tools developed in the early 70 can really look at a Windows machine and claim it has kept us back two decades? Linux doesn’t even have stable API/ABIs yet. It breaks shit all the time. Even Sun Microsystems solved this issue in Solaris YEARS ago. The Linux community seems rather uninterested in things that actually matter to the wider spectrum of consumers.

    Choice is only good when there is benefit. Linux is ass on the desktop. That is why people aren’t choosing it. They rather pay 2k for an Apple Computer to get a decent UNIX box, than run Linux on an older machine which can give them equivalent performance. It’s terrible, and people are willing to pay for commercial OSes (and inflated hardware prices in the case of Apple) to avoid it.

    You need to stop trolling.

  56. .net jerkface says:

    @Dan

    ” I seriously cracked up when you started talking about legacy and shared libs in Linux versus the “modern consumer OS” that is Windows. Proper package management alleviates the DLL hell concerns. ”

    I didn’t mention Windows. We can compare the system to OS X if you would like, which actually has a cleaner implementation. When Apple engineers built OS X they wisely ditched the shared library system. It doesn’t make sense in a modern consumer OS. It creates unneeded interdependencies between applications.

    “Your assertion that the primary motivation behind shared libs is disk space is ludicrous. ”
    Disk space and memory actually. However when the typical laptop comes with 2 gb of RAM and a 250 gb hard drive the savings are negligible.

    You seem to think that there is nothing wrong with the current system. Try this link:
    http://www.google.com/search?hl=en&q=site:ubuntuforums.org+upgrade+broke

  57. Robert says:

    From an enterprise IT engineer point of view, this is presentation is really pathetic.
    The content is really good and objective, the pathetic bit is that it is like if Mr. Lunduke’s is trying to convince people to change stuff by making them feel cool.

    Basically it is like if he is addressing an audience of 12 years old boys trying to convince them to enroll in some project because it is cool.

    Maybe one day there will be an unified API for audio and one for video, but in the meantime I will keep using linux on a server and Windows/OSX on my desktop machines.

    A lot of wasted resources. What a pity. :/

  58. Børge A. Roum says:

    There is at least one website listing commercial apps and games for Linux: http://www.lin-app.com/

    It was mentioned by someone in the comments to your “Commercial Linux Software” page.

  59. goom says:

    Hi

    I don’t have read the odp presentation yet, but i have a question about the licence of the document. I would like to translate it into French and share it, can i ? Thanks, i go back to my reading of the presentation !

  60. Bryan says:

    goom: Absolutely. Feel free to translate it and give it to whomever you like. Consider it Creative Commons.

  61. Lunduke.com » Ubuntu 10.04 - Perfect says:

    […] often lamented about the lack of a good, easy to use software store application for modern Linux distributions. […]

  62. Vipul says:

    What about Donation!!
    who the money reaches the contributors to open source project.

    if i fix few bugs (contributed to )a big open source project.

    will i get any share out off donations flowing in to that project .

    i think if the money flow to everyone (all the contributors)then it will encourage them a lot

  63. mario says:

    The presentation is a lot about WHAT, but lacks on the WHY.

    Audio: This is a non-issue in 2010. Don’t talk it up just for the silly meme. The initial bugs of Pulseaudio have been hammered out. And it is a good _standardized_ software API for Unix audio. WHY: Its only purpose is to overcome the incompatibilization that the ALSA ABI fork of OSS caused.
    Good notion on just using gstreamer, though!

    Packaging: Again it should be analyzed why there is no standard in package formats. First off all, with package managers and a repository for every major distribution, you do not download apps like in Windows.
    Second, there is of course the snooty-nosedness of distributions. Everyone sticks to their NIH format, just so.

    Third, it’s not just the packaging format, stupid! Also the rpm/deb meta data are different between distributions. Package names are weird in Debian and in SuSE, package dependencies are therefore different. That’s why exchange and a standard is not possible. (The rpm/deb file format itself would not be a problem.)
    Package names are not going to standardize itself. If it ever happend, it would be a simple matter to patch dpkg or apt to read .rpm files as well as .deb. – Chop chop and the format wars are gone.

    In the meantime, I’d recommend “EPM” or similar tools. Generic package managers generate .rpm and .deb and .tgz and installers (and even Win32 .exe installers for EPM) from one source. If you package software, don’t bother with rpm or dpkg directly. Use a generic package manager. Just use epm if you want to be extra lazy.

  64. Amir says:

    Hi, You may wanna share this link:

    http://www.omgubuntu.co.uk/2010/05/how-much-would-you-pay-for-five-great.html

    It’s related to some of the things you said in the video!

  65. trojanflea says:

    Hey Bryan! Have you guys recorded the talk on options for developing linux applications? Will you post that soon?

  66. Darren Stewart says:

    Well, another year on.

    Mostly correct. The landscape is not that hard to see. Most was covered well in the video.

    Video and Sound APIs continue to be a rolling disaster. If anyone is actually serious about it being a ‘platform’, then to actually be a platform, you have to build on something, not shift the underpinnings round every damn 6 months.

    Therefore, someone somewhere has to create and lockdown, at least with a stateful set of planned changes, Video APIs and drivers, Sounds APIs and drivers, and a game API (OpenGL and something akin to Direct X), and application APIs – and dare I say this, it has to be done in a way that captures or brings everyone onboard.

    Package management. Simple, every distro should be on Apt. By all means support other package management, but default should be Apt (or a decent current and best version of it)

    Applications. Its almost insane to bother with applications unless the API/Sound/Video/Other API thing is fixed. Forget about it. No commercial or other company can continue to have to fix their shit all the damn time cos others break all the underpinnings again and again and again, and cover 500 distro’s all running their own flavours of every damn thing. That IS not a platform, that is a mess. Therefore unless the first thing actually gets fixed, you will never ever get this. Ever. Its never going to happen.

    You think Valve or Adobe is going to come in and face this garbage? Get real. You have to be a decent platform to bring people in. If you don’t, You kill yourself at every stage anyone looks at doing anything. Imagine Valve having to try and make their products work across 500 different broken, borked and messed up sound APIs. Imagine Adobe trying to work their way through the driver and API graphics mess if they did bring Photoshop. The end user complaints alone for program breakage would sink them.

    Fix the effing LSB. Its a very short agreement to not agree on very much. RPM? Give me a break. People talk up the LSB, but its so far short of what you need its not even funny.

    And I’ve just watched Ubuntu meddle and fiddle and screw around on their LTS version. Rushed? Not half.

    Despite making good progress, despite some good distro’s, people really need to get real on where things stand. Things are only superficially better today than 5 years ago. Underneath, many core problems are no better at all.

  67. Joe Ryan says:

    I watched the video and could not help but recall my issues just two weeks ago with my seemingly conquered Linux issues. I still love Linux though. Just in case anyone wants to hear my rant here it is. http://joehacker72.wordpress.com/2010/05/06/linux-is-frying-my-laptop/

    My main frustration are game breaking regressions, as you mentioned in your talk like audio etc.

  68. easybutton says:

    Steam has been officially announced for linux.

    http://www.phoronix.com/scan.php?page=article&item=valve_steam_announcement&num=1

  69. danni says:

    Two weeks later, where is the official Steam announcement? That’s right, nowhere because Phoroenix made it up. And as always, those wanting to believe, did so.

  70. Lunduke.com » Google Ads vs Apple Ads (and why I don’t care) says:

    […] yes, I know developers need to earn a living.  That is a topic I tend to talk quite a lot about.  And if including embedded advertisements within mobile applications is a way to do […]

  71. danni says:

    Dude, where’s my Steam?

  72. dan l says:

    I’m not making this up. I was about halfway through it and my audio blew up. I got a notification about pulseaudioalsophenomserver doing something or other.

  73. nate says:

    If anyone is still interested in Linux (no one should at this time and age), perhaps would want to know that Dell is dropping Linux, again.

    http://www.pcpro.co.uk/news/359740/dell-drops-ubuntu-pcs-from-website-for-now

  74. the_madman says:

    Canonical has released the Software Centre, and plan on creating a new, “repository” for user-facing software to keep up-to-date: a common example is making the latest versions of Firefox and OpenOffice available in a, “stable” set-up as soon as they are available.

    I think this should be extended to include new commercial applications that can be bought from the Software Centre directly.

    Finally, I believe Canonical could themselves spread word of, “new entries” to the Software Centre on a blog or RSS feed, for example. This way, instead of requiring the download of a .deb from the website or other such interaction, they could include an apt: URL that points to the software directly in the Software Centre. This would provide wide coverage, easy discovery, a single place to find all software mentioned in the New Entries page (the Software Centre) and the ease-of-mind that the software is coming from a trusted and secure repository in the Software Centre.

  75. yman says:

    Perhaps what you are looking for is something like NixOS with it’s Nix package manager, which is designed to not only support multiple concurrent versions of the same software, but also allow the system to be booted into different configurations, or be rolled back to the exact state it was in at any point in the past. NixOS is a research project, not a consumer oriented desktop, but still it might be worth a look:
    nixos.org

  76. Sean says:

    To me it’s incredibly frustrating to see such vast potential wasted the way it is in the Linux community. I am a System Specialist working for one of the top 10 retail websites in the world. I see every day the joys and the horrors of Linux as we use it for our webservers (but AIX for our appservers).

    The general consensus amongst our staff I would say is that no-one except for Linux developers want choice for the sake of choice. No-one wants 500 different distros. No-one wants 30 different IM clients, no-one wants the insane number of useless and broken visual effects on their desktop.

    But most of all NO-ONE except for Linux Developers wants to FIX THEIR OWN SOFTWARE. We want to USE our software.

    What percentage of the world’s population could write a basic shell script (or even have an interest in learning how)? Less than <1%??? Then that's where Linux's desktop market share will stay. Bearing in mind that even in our workplace where we want to work in a nix based environment, most of us choose a mac for the other 90% of the day when we need to GET STUFF DONE.

    The Linux community reminds me of watching a group of 5 year olds playing football (soccer). The ball goes one way, and en masse the whole clump of 22 kids runs that way and swarm all over the ball. Then suddenly it gets kicked the other way and off they all go to swarm all over it with no formation, tactical knowledge, no-one left to defend the goal, and no-one in space ready to accept a pass.
    Linux devs (generally) all swarm to whatever the hot new package, distro, theme of the day, is whilst leaving no-one behind to make sure the audio stack isn't screwed up whilst they are gone. There is no grand strategy (other than f**k Micro$oft lolllllz) and no building of a solid defence such as a single baseline of the critical components that need to be maintained a stable and way so that all developers can count on the critical stuff like video, audio, network, are handled in a consistent and stable way that doesn't get changed because everyone decides 6 months later that another equally broken way is better.

    Can you imagine the improvement in quality if all linux devs concentrated on a maximum of 3 variations of the same distro (perhaps split along the lines of business-focus, desktop-focus, and embedded-focus) instead of splitting their time over an infinite variety of clusterfucks as it does now?

  77. yaffare says:

    I totally disagree with the hole campaign.

    If I wanna pay for a great OS, I use apple. But I use linux because its free.

    Think about it, “Open source for free works!!”, it just takes some time.

    There will always be companies, rich persons who invest or spent money, and
    there will always be developers who do it for free or less.

    I mean just wait 3 years and gimp is there, where photoshop cs5 is today.

    I will throw all my penguins out the window, if this is the way linux goes.

    I hope there are more people with my opinion.

  78. undergroundman says:

    I wrote about this on my blog (http://samethoughts.blogspot.com/2010/11/linux-assessment-and-standardization.html), and on standards I agree but about purchasing commercial software I think this:

    I do not agree with Bryant that getting more commercial software available on Linux is important. Linux is the long-run (20-30 years from now) future as more and more people become programmers and the basic software functions that people need become programmed by these people as open-source. With the exception of games, it’s going to make most programmers unemployed and put most software companies out of business. Right now, I can do most of what I want on a Ubuntu laptop in my free-time without paying for anything. Now, that’s not technically true – I do want a better spreadsheet than OpenOffice and I may want to play a game (but not likely). At some point I might want a computer-aided design (CAD) program. But that’s really what’s left that Linux doesn’t have, and these specialized users are not a high percent of the market. Gamers are not a huge percent of computer users either and they will come when all the normal people can trust the OS to browse the web and play music start not buying Windows to save a few hundred bucks on their initial investment. After that they’ll start looking for games and play the Linux ones available (assuming WINE isn’t good enough to run the Windows games anyway).

  79. Bob says:

    I sort of like Windows and Mac better because when you are trying to get something to work, like a mysql gui, you just expect them to work in Windows and Mac. When it doesn’t work in linux or there are a hundred thousand dependencies to get the stupid thing to work you don’t even want to bother to ask for help because all the linux cultists are all about reading the unreadable man pages and other annoying things.

  80. Andy says:

    Brian, watch the LAS and watched this video. It’s refreshing to see someone who is a enthusiast, as is obvious by your presentation week after week on LAS, and yet a realist. I think more tree-hugging-hippy-hackers (not that there’s anything wrong with that per-say) need to realize the simple fact as you so succinctly put it: developers need to eat.

  81. Marc says:

    Also… Can someone slap (or punch) the guy who is playing with the video camera in the face. Zooming in and out moving up and down, right and left.
    Set the video once and for good before to start.
    Have a good week-end nevertheless.

  82. Lunduke.com » Why Linux Sucks (Less Than Before) says:

    […] This last weekend at Linux Fest NW, I had the pleasure of presenting my third anual “Why Linux Sucks” presentation (which I also presented in 2009 and 2010). […]

Leave a Reply