Slashdot Mirror

← Back to Stories (view on slashdot.org)

FLOSS 2013: the Survey For Open Source Contributors, a Decade Later

Posted by Soulskill on from the how-big-of-a-jerk-is-your-project's-maintainer? dept.

grex writes "In 2002, the first FLOSS survey was launched. With over 2,500 participants, it was the first large survey of open source developers around the world and had a major impact in the community, academia and politics. Over 10 years later, a group of researchers is replicating this survey in order to see how the community has changed. This time not only developers, but all kind of contributors to open source projects are asked to participate. How has the community changed in this last 10 years? Are the views the same? Is its composition and focus similar? These types of questions, among others, are the ones this survey is looking to answer (so far with over 1,000 respondents)."

21 of 27 comments (clear)

  1. Not enough application success stories by Camembert · · Score: 2, Insightful

    I am a regular user of open source. I am not a programmer. Obviously it has been successful in linux kernel stuff. But while there are a number of successful end user application projects, there could be many more. The one thing that is frustrating for me is how many interesting projects die when the primary programmer moves on. Also I find it a pity that several open source programmers work on competing projects, which often get left behind. Imagine if they had pulled forces together - like on the most succesful project such as VLC, it simply does not seem necessary to start your own variant project of these. At the risk of getting flak, I always found it such a waste to have both KDE and GNOME desktop and overlapping related apps projects. Both are of course rather succesful, but imagine what the current status would be if people had stayed with one project instead.

    1. Re:Not enough application success stories by varag · · Score: 1

      I completely disagree with your point. Part of the attraction for me is the ability to switch from one to another at will.

    2. Re:Not enough application success stories by Anonymous Coward · · Score: 2, Insightful

      To give you an idea of how open source software works and the forces at play, without getting into specifics I will tell you the history of a project of mine.

      Like all lazy programmers, I didn't just decide to write code. Instead I was led there because I was using some software and it didn't do things the way I wanted them done. I got involved in the mailing list and explained what I wanted to do, offering my services to do it. Nobody else was keen on doing things the way I wanted them done, so I decided not to argue and to set up a competing project. Eventually, my project had all the functionality of the first one and worked better (in my estimation).

      While I was working on my project, another project started up. They liked what I was doing, but wanted to work on a more portable platform. I wasn't interested in doing a rewrite, but wished them well. Eventually, their project had all the functionality that my project had. They also had more programmers on their project because their project was more popular (being more portable). I shut down my project and recommended that people using my code move to the new project.

      The original project that I proposed changes to still exists, but almost everyone has moved to the newest system. I suspect that the person running the original project only keeps it going because he likes to tinker with it and it does everything that he needs. Everyone is happy and we have a much better ecosystem because new ideas could be tried without having to convince the old guard. I would argue strongly that the friendly competition that existed between our projects -- using the best ideas and ignoring the bad ones -- greatly enhanced the exploration of the problem space.

      I would likewise argue that if we had all pitched in on the original project that we would never have gotten such a good system in the end. This situation is far from uncommon. You see it in virtually every open source project around. It is one of the best things about it.

    3. Re:Not enough application success stories by Raumkraut · · Score: 2

      If everyone worked together on a single tool for each job, we'd likely have a bunch of tools which are bloated, complicated, try to be everything to everyone, and end up being useful to nobody.
      Some projects might be wildly more successful than others, but that doesn't mean that one is a fundamentally better solution than the other, or that the less-successful one is pointless or useless. It just means that more people prefer one over the other.

      Why is it seen as good that we have choice and competition in mobile operating systems and devices, for example, but so often not that there exists a choice in desktop software?

    4. Re:Not enough application success stories by blackest_k · · Score: 2

      Choice is good and not having to choose is also good.
      Thats the best aspect of KDE and Gnome. As a user you might choose one of these as a base desktop environment others are available but your still free to pick up the applications that you want to use.

      I'd bet most users (who have a choice) use a mixture of Kde and Gnome applications.

      Kde seems to be making some advances in supporting touch and making the view part of MVC more flexible and why shouldn't it be this way wimp and touch interfaces have differing requirements. Gnome seems to not be moving much in this direction if at all but thats fine by me only my tablet has a touch-screen so far.

      With Microsoft backing touch with windows 8 (love it or hate it) and the kinect bringing gestures into play its likely our hardware is going to be supporting touch and gestures in our not too distant futures. Voice too if we can overcome the embarrassment and the recognition becoming accurate enough to not wind the user up.

      With 2 or more teams with alternative view points quite different applications and desktops are developed. One teams idea's may work better than another which may inspire another team to steal, as in take that idea and turn it into their own.

      Imagine a world where the choice wasn't between kde and gnome but say unity was the one and only linux desktop whats your option then windows?

      Even with windows you have windows 7 and windows 8 even Microsoft gets the need to have choice. They might prefer everyone to move to windows 8 but they still have to support 7 for a good while yet.

           

      --
      Blarney Quality Restaurant, Plants
    5. Re:Not enough application success stories by Camembert · · Score: 2

      Oh, I know that, it is a nice theory. However in many cases we end up with many half-finished applications for roughly the same functionality instead of one neat app. Gnome and KDE are each quite succesful but they and their apps could have been 4 years further in the evolution if there was one desktop. In my opinion!

    6. Re:Not enough application success stories by aiadot · · Score: 1

      I partially agree with you but partially agree with the parent post as well. Competition is good but fragmentation is not. Developer time is precious and limited. For one project to simultaneously develop for many similar APIs and ecosystems can be really costly and inefficient in many aspects. I personally would just prefer that each team focused on getting something that works well first on limited number of configurations first and then expand in to other APIs/devices/environments/OS later. Successful closed source projects/corporate world functions, I see no reason for open source not to adopt similar policies.

    7. Re:Not enough application success stories by __aaltlg1547 · · Score: 2

      To me, this is a big problem. There's a great deal less development of applications. There are tons of productivity projects that are years, sometimes a decade or more, out of date compared to commercial applications. Frequently they've been left in a halfassed, marginally functional state. Largely, the open source community has failed to deliver on FSF's vision of free software for the masses, unless the free software you want happens to a Linux desktop or an Android phone or some works-pretty-well-but-not-as-well-as-Microsoft office software.

      The reason seems obvious: open-source programmers work on whatever they want. They want to work on what's hot so they all gravitate to current hot projects and everything else gets left by the wayside. Meanwhile commercial companies with customers who pay work on software they can sell to people who need software to make complicated tasks easy to do.

      But seriously folks, how many programming languages do we need? Do we have enough already? Did we have enough already 15 years ago?

      I think it will be this way for a long time. Maybe forever. What might change is that we'll see commercial software companies take more of an interest in making their programs work on Linux and FreeBSD desktops instead of or in addition to Windows and Macintosh.

    8. Re:Not enough application success stories by Will.Woodhull · · Score: 1

      I may be feeding a troll here. I usually do not respond to ACs unless they have gotten some mod points.

      Pooling people to do one thing well is not good? You'd rather have those dozen non working programs, which you can switch between?

      The AC is incorrect in assuming that everyone agrees on how the goodness of a program should be judged. The number of current ways of judging a given program's goodness is about the same as the number of current users of that program. That there are successful alternatives in some application areas reflects the different needs of different groups of users.

      The AC is rather arrogant in assuming that everyone agrees that his way of measuring quality is the only way.

      I have been using FOSS from app to OS in all my work (writing, illustrating, some web site administration) for more than a decade, typically working on the computer 7+ hours / day, 7 days per week (when you combine favorite hobbies with work, you end up looking like a workaholic when you're just having fun). So I know a little bit about long term use of FOSS products.

      --
      Will
    9. Re:Not enough application success stories by Will.Woodhull · · Score: 1

      Successful in the not-so-humble opinion of author of parent post. Others may use a different yardstick to measure success.

      Gnome and KDE have very different, and basically contradictory, approaches to desktop management. You could probably combine the two: you could probably take the design for the biggest cement mixer truck ever and combine it with the design for the fastest Ferrari ever. But you would end up with shite.

      The same goes for other places where there appears to be competition between FOSS products. As soon as you look deeper than the surface, you find that the products differ in very significant ways to meet the needs of very different uses.

      Apache is a good counter-example. There is basically only one way to be the Best Web Server Ever, so Apache has no significant FOSS competitors. Persons who see a way to make a server better contribute to the Apache project.

      --
      Will
    10. Re:Not enough application success stories by Camembert · · Score: 1

      Thanks for your comment. This is exactly what I meant in my original post.

    11. Re:Not enough application success stories by Vyse+of+Arcadia · · Score: 2

      At the risk of getting flak, I always found it such a waste to have both KDE and GNOME desktop and overlapping related apps projects. Both are of course rather succesful, but imagine what the current status would be if people had stayed with one project instead.

      Well, the reason there's both KDE and GNOME needs a little historical context. KDE relied (and still does) heavily on the then-closed Qt toolkit. The authors of GNOME wanted to build something basically like KDE, except with entirely free software components. Naturally, they also needed to write replacements for KDE applications too, because they also relied on Qt.

      Of course years later Trolltech relicensed Qt under the LGPL so there was no longer any fear of Qt vanishing and KDE having to scramble to find a replacement. But by then, KDE and GNOME had taken different paths, KDE focusing on extreme configurability, GNOME focusing on user-friendliness. And then there were all those people who didn't like either desktop and decided to roll their own to fit their own needs better, and that's fine too.

    12. Re:Not enough application success stories by Kinwolf · · Score: 1

      Apache is a good counter-example. There is basically only one way to be the Best Web Server Ever, so Apache has no significant FOSS competitors. Persons who see a way to make a server better contribute to the Apache project.

      Nginx maybe?

    13. Re:Not enough application success stories by Will.Woodhull · · Score: 1

      I have definitely been feeding a troll.

      So sorry.

      --
      Will
    14. Re:Not enough application success stories by bregmata · · Score: 1

      That damn free market.

      Obviously the only successful way to run the show is to have a central committee decide what everyone needs and wants, and an effective 5-year plan to meet those objectives.

    15. Re:Not enough application success stories by UnknownSoldier · · Score: 1

      There are 2 relevant cliches:

      * You can please some of the people some of the time but you can't please all the people all of the time
      * Show me your strength and I'll show you your weakness

      Note: Capitalism has the same strengths and weaknesses as Open Source. Let me explains ...

      The strength of having diversity in "competing" open source programs is that they help feed off one another to make themselves better. Think of a little friendly healthy competition. The user in this case benefits as "vendors" try to "market" to the user by providing a solution that fits the user's needs.

      Unfortunately it also leads to wasted and duplicated effort as everyone (re) struggles to re-invent the wheel. This is caused by different user requirements. The user in this case is marginalized, and/or frustrated.

      There is a balance between a small program that does one thing well, and a big program that does lots of things but only adequate. This balance of S/N is extremely difficult to get right.

      The "Barrier to Entry" makes it somewhat difficult for outsiders to start contributing -- code, docs, test cases, etc. So people get frustrated and start their own project. It is a vicious cycle.

      Open source has the benefit that you are given the freedom to actually fix things. It is extremely empowering as a programmer that you have the source to step into when debugging. You also never have to power about money grabbing nickel-and-dime licensing issues as you do with proprietary software.

      Eventually open source will be "good enough" for the masses (we're almost there!), and we'll look back on closed source software as archaic: Someone who thought more about what they could get, then what they could give to the world. If businesses were smart they would financially support open source software by sharing the burden instead of bitching about paying the Microsoft "Tax" or other extortion licensing fees.

      --
      If you can't duplicate the experiment it is NOT real Science!.
      A thought experiment is an oxymoron and teaches you nothing new that you didn't already know.

  2. OMG by Big+Hairy+Ian · · Score: 1

    Even in Open Source you can't get away from the census!

    --

    Build a Man a Fire, and He'll Be Warm for a Day. Set a Man on Fire, and He'll Be Warm for the Rest of His Life.

  3. Two significant trends by Anonymous Coward · · Score: 1

    Trend #1: Reinventing the wheel. We don't need yet another programming language or MVC framework. We don't need two incompatible and competing versions of Python. We don't need Google churning out new programming languages. The amount of time and effort wasted on reinventing wheels is significantly damaging the software industry, which needs to go back to standards. Pick a language and use it, and develop that language, rather than inventing new ones every month. Fragmentation is making progress difficult.

    Trend #2: Change for the sake of change. If a project runs out of innovations, they change things gratuitously, like FireFox removing the status bar or Unity/Gnome 3. No one asks for these changes, and no one wants them. It's like a project has to make changes to justify its own existence, even if they're not good ones. Churning out gratuitous changes is significantly harming open source, because people like me are about to say nuts to open source and use a Mac.

  4. Fragmentation and Exploitation by Anonymous Coward · · Score: 2, Insightful

    The responses here are rather depressing. As a(n extremely humble) coder, to have a half dozen posts complaining of fragmentation is rather depressing. Somewhere there must be a non-trivial program which was written purely for the sake of doing so, but for the general case we may say that software is written neither accidentally nor arbitrarily, but to remedy a lack in existing software. It is in exceedingly bad taste to complain about any article offered gratis, but to say that the work involved was counterproductive is quite offensive. As a remark applied to a large and popular software project, as more than one of my fellow commentors have, it is a slur against the entire profession. The software is gratis even to ingrates, but -- as a programmer, you should consider the beam in thine eye. And if there is someone who instructs a computer but will not lay claim to being a programmer, well...perhaps silence is better than 'removing all doubt.'

    Touching more directly on the subject at hand, the question which most interested me was the one on exploitation. My opinion is that F/LOSS is indeed exploitative, especially if you aren't paid for working on it. Closed source may be just as exploitative, with the added sting that you may not be compensated commesurably with the commercial exploitation of your work. However, with the understanding that the fruits of your labors are often not directly monetizable, the value of the F/LOSS network may be presumed to be proportional to the number of participants, and reaping the benefits of this is limited only by your ability to consume them. That in itself may not generally be exchanged for a cup of coffee, but for many people it is worth an excellent salary, and ends up being a good value proposition. Discounting the ecosystem, "self-exploitative" isn't a bad description of contributing to F/LOSS projects, and many other walks of life besides. Taken on the whole though, it's abundantly clear that an open development model is highly beneficial to both its participants and users.

  5. FreeCiv by unixisc · · Score: 1

    I like FreeCiv - that's one FOSS that I use regularly, and which would enable me to switch to something other than Windows (aside from the usual Firefox/Chrome & the rest)

  6. Re:Javascript Trap by unixisc · · Score: 1

    Mr Stallman, I presume?