Trying to improve x49gp

[3298]

(05-07-2018 06:33 PM)Claudio L. Wrote:  Thanks for your work, these are very welcome patches. What I value the most is your effort to "un-git" the project. Distributed version control has one big pitfall: unless there's a central maintainer ready to pull patches, projects fork ad-infinitum and nobody knows which one to use. To make it worse, github does a great job at making them look all the same. There should be a different template for pages of the "official" repository of each project, so people know where to push patches and users know where to fork from.

That wasn't my intention at all. In fact, merging the three best forks (in my opinion), using that as the base for patches, and submitting that to the official maintainer is very much Git-like. If he wasn't busy, you would probably have seen something similar to this patch series appear in his repository, including the merges, and this thread would have remained in the depths of the forum instead of resurfacing.
I agree that the lack of an active official maintainer (or group of maintainers in larger projects) can lead to fragmentation, but the beauty is that if somebody else wants to, they can get quasi-official by simply caring about a project in the absence of the maintainer, developing fixes and features and merging other forks into their own fork. In this case, I firmly believe chwdt will have time for x49gp again eventually; in the meantime we can run it ourselves...

(05-07-2018 06:33 PM)Claudio L. Wrote:  Thanks again, I'll pull your patches in my fork, hopefully we can make all existing forks consistent.

... by doing that, for example. Taking back the lead doesn't get much easier than a fast-forward merge, so with the tip of his fork merged into the new quasi-official branch, (and perhaps a pull request or message to notify him,) everyone should be happy. I for one would be, because it means this reaches a larger audience than this forum post can hope to. The significant influx of fresh patches compared to the number of past commits in this project should draw attention from some of the authors of the other forks too. I hope some of them will feel inspired to build on top of this work.
Judging from our private HPGCC3 conversation a few years back, I'm sure you can relate to the privacy objections that stop me from doing the GitHub push myself, so I really do appreciate the offer.