Why Linux Sucks & Why Linux Does Not Suck – THE VIDEOS

Note: The 2014 edition is available here.  Below is information on the 2012 edition.

Over the past few years, I’ve given a presentation that I call “Why Linux Sucks” at Linux Fest Northwest.

In that session I go over a whole mess of reasons why I feel Linux Sucks… and what steps we can take to fix those sucky parts.

This year I decided to shake it up by doing two, back-to-back, sessions:

You can watch youtube videos, taken from the audience, at those links above (I’ll also embed them below for good measure).

Side note: This year was a ton of fun with a great crowd (about 200-ish, I believe, for each one).  Linux Fest Northwest remains one of my favorite technology conferences – a very relaxed and friendly environment, while still being large enough (this year was pretty packed) to have some amazing content.  [More thoughts and videos on Linux Fest as soon as I get my head back on straight.]

 

Share Button

33 Responses to “Why Linux Sucks & Why Linux Does Not Suck – THE VIDEOS”

  1. Velkro says:

    Truly an INSPIRATION and food for thought.

    Watched both back to back and am still smiling.

    You guys are AWESOME

    Keep up the good work!!!!

  2. Sicp says:

    precise and all-encompassing; great effort, thanks!

  3. Tony says:

    Chris, you REALLY need to use a tripod. Set up the tripod, frame the shot, and leave it alone for the entire talk.

    The shakey zoomy wobbly cam makes the video extremely difficult to watch.

  4. alboNica says:

    Hey lunduke:
    http://en.wikipedia.org/wiki/Quetzal

  5. Tom says:

    Hey Bryan,

    Great lectures, but I wanted to point out that there are a number of professional CAD solutions available on Linux. In the transportation industry, the two main contenders are ANSA and Catia, both of which run on Linux. For most major manufacturers the Linux versions are actually preferred!

    Cheers,
    Tom

  6. Why Desktop Linux Sucks & Why Desktop Linux Does Not Suck | Cerebrux says:

    […] “Why Linux Sucks” and “Why Linux Does Not Suck” videos from Linux Fest (lunduke.com) […]

  7. JA12 says:

    +1 for Tom. I didn’t know those existed.

    But I know car manufacturers like VW (who also owns) Audi/Lamborghini/Bugatti, Mercedes-Benz, BMW, to give an example, run Linux, have done so for a long time, and use professional tools on their servers and desktops.
    There are CAD tools, casting simulation, finite element analysis tools, software for diagnostics, collision simulation and so on. They’re just not very loud about these things and so we know quite a little about them.

    As far as I can tell, such big companies run Linux because it’s powerful, scalable and open so they can better build their own environments and tools. Obvious really, but it doesn’t hurt to mention that.
    They use a lot of expensive proprietary software. You probably can’t spend so much money just to educate yourself and/or use them for hobby, which sucks, but at least something flows back to Linux. I did some digging to write this comment and VW has donated CAN (controller area network) code in the past. Makes sense that these companies would want changes to happen in the upstream, that’s how it usually goes, so there might be more examples. These companies might also give monetary support but I’m not going to dig more to confirm that.

    Some sources and conclusion in parenthesis:
    http://archive09.linux.com/feature/59200.html (Audi runs Linux)
    http://www.docstoc.com/docs/23933185/Case-Study-Crash-Simulation-Linux-Cluster-at-VW (VW runs Linux)
    http://www.computerweekly.com/news/1280097128/How-BMW-virtualised-with-SuSE-Linux-and-Xen (BMW runs Linux)
    http://www.redhat.com/support/resources/basics/linux_intro.html (Mercedes runs Linux)

    Some of software they use:
    http://www.mscsoftware.com/Products/CAE-Tools/MSC-Nastran.aspx
    http://www.magmasoft.de/ms/home_en/index.php?rubric=Home+North+America
    http://en.wikipedia.org/wiki/Abaqus

  8. Dan Douglas says:

    That $992/mo figure for Ardour only considers “monthly subscribers” (those who automatically donate a fixed amount every month). Actual income is typically a few times that. Not sure if You (Bryan) were aware of this. If I’m not mistaken, they do still typically meet their monthly goals. The point is still valid, of course – it’s still far lower than the income a typical developer would expect, and that’s one of the very most successful purely donor-supported free software projects.

  9. John Cartwright says:

    The Linux distributions in use today are doing a good job on the desktop, but they need a unified desktop interface, and this is not Unity. They need something professional looking and presentable as well as fast. Something that corporate users can deploy out of the box. That is what Linux really needs right now.

  10. Will says:

    Wow! this guy is a great speaker.

  11. Mike says:

    Many people have tried to address the fractured (or at least poor) packaging ecosystem by introducing… yet another packaging format. As you said, it’s a religious war.

    Here’s what it would take to successfully make a unified packaging system: 1.) you need to support all the features. See Arch vs. package signing and Debian vs. incremental packages. 2.) Don’t constrain yourself to be compatible with existing package systems. 3.) Most importantly, your new package should be able to be *automatically convertible* to some other package.

    I really don’t think it would be that hard. Distros tend to have certain ways of doing things; those rules could be packaged as some sort of template mechanism. Dependency names could be mapped on a per-target-distro basis. Let’s call this new system Universal Package Manager (UPM). So you write your UPM, provide all the necessary data, and then convert to Ubuntu .deb, Debian .deb, Fedora 17 RPM, openSUSE RPM… whatever you have configurations for.

    The final step is to write yum/apt-like tools for UPMs and enable distros to ditch their legacy formats.

  12. al says:

    awesome talks :D

    keep it up!

  13. enimem says:

    And for a fourth year I wonder why does the camera is hand held and using the built-in microphone ?!

    You’re a public geek figure, I’m sure you can setup something to professionally film a lecture.

  14. Stefan says:

    Agreed with Tom and JA12.

    There are some electronics programs that run on Linux well too.

    Blender is an excellent example of a great open source project. They are really developing quickly.

    Also, a correction, GIMP doesn’t strive to be a photoshop clone. Their developers hate it if you say that.

  15. lafka says:

    “XMonad is hardcore awsome, it doesnt have things like close buttons. It’s just awsome. Now is that for most people? No, infact it would be the dumbest idea in the world to recomend it to almost anybody; but it’s awsome.”

    So true :)

  16. Anonymous says:

    Excellent videos, as to why Linux sucks even more than the host showed, there’s another regularly updated list by some Russian guy:

    http://linuxfonts.narod.ru/why.linux.is.not.ready.for.the.desktop.current.html

    Strangely or coincidentally the video “Why Linux sucks” repeats a fair number of points of this list.

  17. Sonny says:

    I would have very much appreciated the Why Linux Sucks slides to be downloadable. The video is fun, but trying to folow your slides is impossible with the non-tripod supported video.

  18. A.I says:

    The presentations would require references for support the claims, otherwise they are just for entertainment at best :-|

    If Linux distributions has to choose just one package format they would need to choose the best not the most popular one.

    I don’t think the release names are so big deal. For example Fedora 16 “Verne” or openSUSE 12.1 “Asparagus” are hardly seen anywhere. Not even in their websites or release notes *. They are just using release numbers.

    I hope LAS would use release notes in upcoming reviews.

    * http://docs.fedoraproject.org/en-US/Fedora/16/html/Release_Notes/
    http://www.novell.com/linux/releasenotes/x86_64/openSUSE/12.1/

  19. Chris says:

    Enjoyed both these talks. Honestly there are a lot more reasons Linux sucks though. Nonetheless I enjoy the versatility it gives me so I continue to use it despite the fact that it kinda sucks.

  20. helsinkiharbour says:

    hmmm, even agreeing on one package format would not solve all desktop linux app distribution problems … only a subset of them.

    One step deeper, the very core problem is the tight integration of applications with the OS itself. Which is exactly what package management is about, keeping the OS with the apps integrated and in sync.

    This is fundamentally wrong, (desktop) apps should be clearly separated from the core system/OS.

    For the application producer/ISVs it’s also a “man in the middle” problem, direct distribution to the customers gets more complicated and slower, update cycles are dictated by the distro(s) when it should be the ISV (“push”) or the user (“pull”).

    e.g. Ian Murdock formulated it like that:
    “moving everything into the distribution is not a very good option. Remember that one of the key tenets of open source is decentralization, so if the only solution is to centralize everything, there’s something fundamentally wrong with this picture”
    2006: link

    also, this problem was noticed by many persons of the community before…
    2006: Mike Hearn, creator of autopackage
    2006: Mozilla developer about the approach of integration of apps with the ubuntu system
    2009: link
    2010: current situation accepted as bug by Ubuntu
    2010: Matthew Paul Thomas Ubuntu (software center is a first but not sufficient step) http://www.youtube.com/watch?v=GT5fUcMUfYg
    2012: Ingo Molnar mentioned this recently in the context of missing scaling of the distro repository approach for software distribution

    Solution could be imho only approaches like CDE, autopackage (RIP), listaller, zeroinstall, portable linuxapps… approaches which goes beyond the distro centralized package management approach.

    thx for reading, still hoping …

  21. helsinkiharbour says:

    @Stefan:
    “Also, a correction, GIMP doesn’t strive to be a photoshop clone. Their developers hate it if you say that.”

    Yes, they mentioning this often… but this is part of the problem of GIMP. From user point of view it’s pretty irritating to be confronted with policies like: “yeah, we know there is some kind of (defacto) industry standard in usage patterns, functionality, GUI design etc. but we don’t care… it’s more fun to make everything different (not even better), users will adapt. You know, it’s open source, if you don’t like it change it yourself!” Also, this is disrespectful to the competences the users might have aquired in a specific application (e.g. photshop), which is propabaly also the competence which pays the bills and the food for the kids. Therefore, the offer sounds very risky to them if someone states “oh, forget this propritary software and his usage patterns, it’s fun to start learning doing things in a complete different way! Use Gimp, its free in all dimensions!”, maybe to risky. For professionals it’s frightening like hell to be forced in a situation where they loss (at least in the adaption time) the ability to fullfill their daily tasks in the most productive way. What is sometimes great and fun in hobbyist domain, to tinker around with new tools, might be to risky, scary and time consuming in the professional domain.

  22. theGrg says:

    Very interesting talk!
    I think there’s one thing from the “Why Linux Sucks” part that you haven’t addressed in the “Why Linux Doesn’t Suck” part, though:
    That large projects like professional-grade video editors, etc. require large investments, reliable funding, and full time work. Which means there may always be some large professional commercial tools that may not be available on Linux due to the company not having interest in the Linux market, and at the same time it not being feasible for the open source model to produce alternatives of the same quality.
    Please tell me if this has been addressed this in the 2nd video because if it was, I missed it.

  23. Serg says:

    Hey Bryan, I am a little disappointed that you somehow managed to ignore the biggest Linux gaming news in recent memory. A few years ago you could jump out the window and survive on will alone if this were to become reality, and now that Steam is officially heading to Linux, you don’t even mention it anywhere?

  24. Bryan says:

    Serg: Steam is not officially coming to Linux. Valve has yet to confirm this. It may happen, and it would be great, but we have no confirmation that this is officially going to happen.

  25. Serg says:

    I seriously doubt Phoronix would make up a story about visiting Valve and speaking to Gabe Newell. Pretty sure it isn’t even legal to do something like that. The time for jokes was one month ago, to the day.

  26. Bryan says:

    Serg: I’m not saying Phoronix made anything up. But Valve has not officially confirmed that Steam is going to be released for Linux. There’s no public announcement or press release. Right now all we have to go on is a conversation one guy had with one other guy. It looks potentially promising… but that is all.

  27. helsinkiharbour says:

    Bryan, one question about “Linux Tycoon” and your cross-distro distribution solution tar.gz. As you pointed convincingly out in your great second presentation (I enjoyed your approach with the almost identical slides ;) ), a big part of the strength and power of linux is the innovativity, the openess for better and non-standard solutions.

    Have you considered new and innovative approaches like e.g. Peter Simons “PortableLinux” (http://portablelinuxapps.org) Apps as cross distro solution for distributing “LinuxTycoon”?

    The community and the creator there are very active http://portablelinuxapps.org/forum/
    I know also that the creator would be glad in helping you because he is desperatly looking for first class applications which use his technology.

    Also, more general, I would love to see an LAS episode discussing & comparing actual innovative non-repository based software distributing approaches like Philip Guo’s CDE (http://www.pgbovine.net/cde.html), listaller (http://listaller.tenstral.net/), Zeroinstaller (http://0install.net/), Ryan C. Gordons Mojosetup (https://launchpad.net/mojosetup) and PortableLinuxApps

    cheers and a nice first may!
    PS: thanks for fixing some of the link formating quirks I introduced in the fomer post

  28. Beluga says:

    On The Linux Action Show s12e10 there were repeated exclamations of “awesome!” regarding OpenSUSE Build Service (now: Open Build Service). Any new insights? I see Tizen is one big project of late using OBS: http://www.open-build-service.org/2012/04/14/obs-developers-tizen-developer-conference/

  29. Lunduke.com says:

    […] “Why Linux Sucks” and “Why Linux Does Not Suck” – My sessions from this year. […]

  30. Simply Bryan Lunduke | Unixmen says:

    […] you feel you like them, or hack with lunduke lang. Also don’t forget to watch his 50 min session Why Linux Sucks (As Usual) at Linux Fest […]

  31. The problem of selling software on Linux « mike.williamson says:

    […] I listened Bryan Lunduke’s talk “Why Linux Sucks“. One of the arguments he made was that the open source/free software development model has […]

  32. Somebody says:

    Howdy howdy howdy ho!

    I had never heard of you before seeing this Linux Sucks talk. But I kinda like you. You bring the cool to Linux. Srsly. (It’s the shoals.)

    Before anything can be fuckin awesome, one has to understand that it sucks.

    I’ll just throw it at you:

    After many years of absence from Linux I just recently installed Ubuntu. Why? Bitcoin, that’s why.

    I knew Linux sucked and having to relearn what the hell sudo was sucked even more. But I needed Linux. I needed a safe environment to do online banking in, and mostly because of Bitcoin, I needed to know my system was absolutely virus free. A pre-installed browser, maybe a text editor and easy means to establish an internet connection for mobile access even without software from the service provider. Linux/Ubuntu did all that. Kudos!

    It runs installed on my Windows partition and I dualboot whenever I need it. Apart from the occasional crash and a screwed up display it serves me just fine. The 2GB I had allocated to it for file storage turned out to have been outperformed by the Bitcoin blockchain size. Reinstall ahead.

    Those shoals man… Why do the most obvious things work? Like Hawaii shirts and Porsches.. anyway.

    Knowing about why Linux sucks is one thing, knowing why it’s gonna ROCK is another!

    This guy in the communication self-help seminar I was forced to attend said “You should always have three(3) options at hand. That gives you true freedom. One option is better than nothing. Two is fine, but having just two will keep you trapped. Three is the magic number. Three answers to a question, three possible solutions to a problem, three women, etc..” You get the drift.

    Watching your Linux Sucks video made me think “What would make Linux truly kickass? What does Linux need? What do the people need? What’s the future of Linux?” No wait. The question was “Why would I want to use Linux myself?”

    And here’s the three(3) Linuxes I have a use for (you may only pick one though in the end):

    Linux 1) The above scenario^^^^^. A stripped down, minimalist alternative desktop OS that gives me a safe environment for sensitive stuff. Easy internet connection (good driver support), basic communication needs, a browser etc. Just so I can safely log in to MtGox knowing there is no trojans and stuffs installed and move my precious coins over to satoshidice.

    Linux 2) A Linux that incorporated everything you mentioned in your video: Photoshop (finally!), big MOFO funding, well-paid developers, kickass games and all that shit, all the time, all the way!

    That reminded me of an idea I had about a year ago. I suffer from the occasional fantastic daydream in which I foresee the future where we put aside the “That won’t happen” philosophy for just a minute and say “Yea! This is totally possible! It’s been done before!”.

    Imagine the Commodore.

    Are you imagining?

    Not the bird, of course. And I don’t mean that C-64 thing either. I’m more thinking C-64000k or something. Or Commodore X. Or Commodore Ultra. Or Commodore One. Or Commodore Revolution, if we must. That thing everybody wants. That thing everybody has, or you suck.

    OK. Now try imagining that again. See the glowing keyboard? It’s mobile, too, right? But not just another notebook. See the kickass display? Good. Have you ever seen a design like that? Fuckin Airwolf! See the revolutionary keyboard layout? Like built in dvorak 2.0 and shit. Maybe foreign character support? Customizable? Fuck if I know. Can you see the lights slowly pulsating like it was breathing? See that array of plugs for all sorts of peripherals? Hear the sound? Bang&Olufssen style. Pretty decent. Gamers love it! I don’t know what the hell it’s running on.. but it’s running pretty fucking fast! There’s even a USB-driven Twinkie toaster for it. Under certain circumstances it has the ability to go airborne, too. What about the operating system? Oh yeah! The operating system! Hmm.. It would need an operating system right?

    But not just any operating system. It would have to be a kickass operating system. Like… One specifically designed for the system. 100% hardware compatibility. Perfectly designed for what it’s running on. Like a Mac. No not like a Mac. Like a MACK! Sheeesh! See that user interface design? How slick it is, how super-professional, how it jumps right in your face? How it redefines interacting with a computer?

    Can you hear that sexy voice saying “System ready.” “Connection established.” “Initiating system shutdown.” Can you?

    Now we’re going somewhere.

    See the windows? I don’t. Instead, I see the system, it’s background services and connections to the virtual sprawl unfold before me like it was space that I’m moving through. Just seamless. Unlike anything I’ve seen before.

    Kind of like a Porsche driver needs a Porsche dashboard. And a Porsche dashboard needs to sit on a Porsche itself.

    Oh, and it has Photoshop!

    Linux 3) A complete and easy-to-use anonymization suite for secure online communication that bundles all those hard to get, hard to understand and hard to setup Tor and GPG encryption things, etc. into one operating system that can easily be used in conjunction with existing ones.

    Pick your favourite!

    Oh, and have you ever thought about renaming yourself to Linduke?

    Regards

  33. Abraham Limpo says:

    Love the presentations but you talk a bit faster and english is not my first language… do you have a transcription of the conference?

Leave a Reply