get GNU/Linux! community

Github is problematic

Roy Schestowitz calls to avoid GitHub since a long time. Here is a recent article:
No Excuses Left, Time to Leave the GitHub Monopoly

The libreCMC project doesn’t want it software get forked on GitHub and has put together What is wrong with GitHub?

In the long run I think we need to move away from GitHub. The website is small enough that we can recreate it with any other framework on any other plattform. And it will always be small. The work lies in the well thought-through explainations and calls to action.

The GNU website get translated page for page. The current approach with is meant for software. It seems to me like a big framework for little straightforward tasks. No need to seperate each paragraph and have several suggestions for a translation and to introduce the category unreviewed translations.

I’ve spent a couple of hours reading through Roy Schestowitz’s posts as well as other online content regarding this subject. I must admit that I am new to many of these recent developments regarding GitHub and I’m only just now reading about these. However, I find some of the arguments they use to avoid GitHub vague and not really compelling.

GitHub itself shows effort to defend the developer. GitHub does keep part of their source code closed in order to have a competitive edge, as is their right to do so. But they also open sourced many parts of their code. I am skeptical of some of Schestowitz’s claims regarding Microsoft’s motives for buying GitHub, but then again, I am not well read on this subject and may be a little naive.

At this moment I don’t see compelling reasons to avoid GitHub. To me GitHub is a powerful tool that allows us to share our project with the rest of the world, and it does a good job at this. I could self-host the source code, but that has many other disadvantages. But I am open to suggestions for GitHub alternatives.

I love free software, but even I use Ubuntu even though it contains proprietary drivers, but it does make my life easier. Sometimes we have to make concessions.

I don’t think the current approach is meant for software per se. GNU gettext, the software that we use for translation, doesn’t care whether the translatable strings are from software or from a website. When you say “big framework” are you referring to Transifex?

Making the content translatable at the paragraph level has advantages. One doesn’t need to review the entire page when just one paragraph needs updating, for example. If the same paragraph happens to exist on another page, one doesn’t need to translate it twice.

The suggestions are there to help the translator and make her life easier. The translator can choose to completely ignore these suggestions, if she so pleases. They are just that, suggestions.

The “Unreviewed translations” category is just another helpful tool for translators which allows them to review each others translations, just like we do with code in software development. We as human beings make mistakes, and this is just a tool to help spot mistakes and fix them. But it is by no means obligatory. I don’t reject unreviewed translations.

I’ve found a nice alternative to Transifex. It’s less bloated and works very well.

Thank for sharing. It’s helpful for others.