1. 19
  1.  

  2. 7

    LibreOffice is such a big bloated project that it’s probably hard to find people interested in maintaining/developing it. It is by far the ebuild with the longest compilation time on Gentoo (even surpassing Firefox and Chromium), and even though I found a small GUI-bug last week, I’m not tempted to submit a bug for it, as I don’t feel up to debug it or even write a patch for it, and I hate submitting bugs without doing my homework.

    Considering the topic at hand, open source development can work this way, and the Document Foundation should look into paid support contracts to support their work.

    1. 6

      (fellow Gentoo user here) Honestly, I find Chromium’s build time exceeds that of LibreOffice.

    2. 7

      Over the years I have become increasingly weary of commercial involvement in open source.

      It seems as soon as businesses get involved, they slowly start alienating volunteers with their constant quest of “how do we make money with this”, and pretending that is an issue that volunteers have to be considerate to.

      Plus, all the god-damn behind-the-scenes corporate politics.

      I’d rather have projects that are developed (more slowly, with fewer features) by a few volunteers than some big-corp whose participation is only there as long as it is in their interest (or to make sure development happens according to their interest).

      For my projects, I’m going for “commercial-unfriendly” licenses, diluting copyright as widely as possible by not requiring a copyright assignment from contributors, and being careful to not accept contributions on behalf of corporations.

      Any additional ideas?

      1. 3

        I’d rather have projects that are developed (more slowly, with fewer features) by a few volunteers

        Who’s going to volunteer to hack on an office suite tho?

        It’s like … the dullest codebase imaginable.

        1. 1

          Maybe this introduced a hard complexity ceiling on software that is going to be written; wouldn’t be too bad I think.

          1. 2

            I always found abiword and gnumeric to be a lot more pleasant to use than LibreOffice personally.

            But big business customers want every feature that’s in MS Word and Excel, and yeah… maybe that’s just not sustainable.

            1. 3

              But big business customers want every feature that’s in MS Word and Excel, and yeah… maybe that’s just not sustainable.

              I think that’s pretty much the point I’m making – if big business needs every feature, they can go build it on their own, that’s simply not my issue as a volunteer – neither to develop, nor to maintain, nor to support.

        2. 1

          Well, my take on it is that:

          A growing user base and a growing compatibility-with-standards – necessarily introduced more complexity (and, presumably, more technical debt).

          For an open source project, it means more development time needs to be applied.

          Unless its developers work for non-profit or academia, or independently wealthy – they will, eventually, need to get paid to sustain the level of involvement that’s dictated by the growing complexity of the project.

          With regards to

          I’d rather have projects that are developed (more slowly, with fewer features) by a few volunteers

          I think that’s your answer of how to avoid commercialization for your project, if that’s one of your goals.

          Basically, your feature set and technical complexity needs to be such, that it can be managed by a ‘constant’ number of developer-hours. If you were able to master those (through contributions or by youself), then you have a good start, but, I think you would have to think ahead, to make sure the complexity does not grow.

          More practically, it means constantly ‘culling’ your features – dropping of capabilities/features, when you choose to introduce new ones (unless you find ways to constantly to introduce ‘superset’ features, that with the same development effort replace other ones)

        3. 5

          As long as LibreOffice keeps being maintained and some features are added, I’m pretty OK with that. It’s still much faster and has saner UIs than MS Office. (Try opening a standard conform CSV file with Excel 2016..)

          1. 2

            One would think that the demand is big enough in this space that there would be tons of great software to choose from. But neither LibreOffice nor Microsoft manage to create user friendly, non-bloat products. GSuite seems to do a rather decent job, but there are many other issues with depending on Google’s cloud..

            1. 1

              UI aside, I am glad that there’s a libre alternative for MS Office. One thing both miss is a good hosted solution (alternative to G Suite). I have Nextcloud with Onlyoffice and CODE and both are bloated, slow. I always end up using MS Office for larger documents.