Indie Python project repositories to adopt pygame-ce

[KennedyRichard]

Hello, everyone,

A few months ago there was a split in the original pygame project. In summary, there are now 02 different pygame libraries in the wild:

pygame/pygame (the original instance) and the relatively new fork pygame-community/pygame-ce.

Both use the name pygame in the code and they are still compatible since the split was recent. This means they are interchangeable and changing from one to another requires no change in the source of any project, just that one is uninstalled and the other installed to avoid one shadowing the other. I even tested the new fork (pygame-ce) with Nodezator and Bionic Blue and both worked without a hitch, not a single change needed, just as expected.

As the time goes, the differences between them should naturally grow.

Given how recent the split is, I have been purposefully postponing dealing with its implications, in order to observe a little more before deciding what should be our response to it as the Indie Python community. However, last week Mr. Durk @durkisneer1 reached out to us on github (on Nodezator's repo) suggesting the adoption of pygame-ce. I think this is an opportunity for us in the Indie Python community to consider the matter.

I believe it is in the best interests of the Indie Python project and its community of users and contributors to adopt the new fork, pygame-ce, in the Indie Python project, that is, in its repositories that use pygame.

However, like most decisions in the community, specially those that have a larger impact, it is necessary to provide appropriate arguments/evidence in favor of the decision one wants to make and to give people the opportunity to share their opinion on the matter. Thus, let's try to understand what happened and why the new fork is most likely the best alternative.

I won't comment on the specifics of the split, since I myself didn't witness it. However, I deemed necessary to make a little research on the matter, in order to offer as much relevant information as possible to you. I searched all pygame-related channels/profiles I could think of: discord, reddit, twitter and mastodon.

After such research, I only managed to find an statement/explanation from the people representing the new fork, which you can find here. I found that explanation to be rather reasonable and think it dealt with the sensitivity of the matter gracefully.

The info was also corroborated by a prominent youtube channel, known for long and detailed videos/content on pygame.

Moreover, the owner of the original instance of pygame offered no official statement on the matter (at least not that I know of, despite the many channels/profiles I searched).

In light of all which I presented until this point, specially the aforementioned statement we linked, I do believe that the folks from the pygame-ce fork have a project that is better aligned to lead pygame development and manage the community of users and contributors in the pygame ecosystem.

But wait, I have only presented statements! Important decisions need to be backed by real data, hard evidence, etc., or at least coherent arguments, not just taking people words for it!

Don't worry, I won't present you any complex graph or several slides supporting my point. It is actually too soon for any kind of deep analysis. But, I'll present some extra info that I believe to be relevant.

First, I do know a lot of the developers that gathered under the pygame-ce project. No, I don't know them personally, but they are people with which I've been exchanging messages here and there, or observing their activity on reddit and discord for a long time. There's also people active in the development of other important projects related to the pygame ecosystem, including people with which I interacted in the Indie Python community. I don't know the actual numbers, but I'd say the great majority of the talent behind the original pygame instance is now working in the pygame-ce fork.

Second, try hanging out on pygame-ce discord a bit. I'm satisfied with what I've been seeing there from the past few months: people helping each other, several topics being discussed. I think the best thing about the pygame-ce fork is not that it has a lot of talented people there or that it has already surpassed the number of commits in the original pygame instance and outpaces it in development speed, according to this tool (the numbers below are for recent commits in the past few months; you'll have to add the names of the repos to perform the comparisons, pygame/pygame and pygame-community/pygame-ce):

Those numbers are cool, for sure, but they don't actually convey what is more important: they say nothing about the quality of the code or the design behind it, nor the quality of the process behind the development process. What I'm trying to say is that it is not the numbers that set pygame-ce apart (although, as we saw, the numbers are good), but rather the transparency of the process and the fact that the decisions are openly discussed and subject to inspection from a larger number of people.

I wish I had more data to show, but I think the arguments presented are more than enough to justify the decision to switch to the new pygame-ce fork.

I also wish I had more to say regarding the stance of the owner of the original pygame instance, but no comment was made until now regarding this and I think it should be taken as an indication that it should be left this way.

I'll just add that I'm glad that both projects seem to be moving on and I respect both parties involved and will keep cheering both on, and sharing and liking their work and content online across all Indie Python channels.

Also, thinking of pygame users alone, it is actually great that the technology split into 02 instances, since I expect competition between them to drive the technology to new heights. As long as the competition is handled with respect, fairness and ethic, users will greatly benefit from it.

I welcome all feedback/opinions on the matter, which why this discussion exists.

In a couple weeks, once this discussion is closed and the decision is cemented, I'll provide info on when exactly the switch to pygame-ce will be made and further instructions to users of Indie Python apps/games.

Peace.

[KennedyRichard]

With the recent v1.4 release of Nodezator, pygame-ce was finally adopted. Further instructions on how to deal with the change can be found in the release post: https://github.com/IndiePython/nodezator/releases/tag/v1.4.0

pygame-ce will also be adopted in the Bionic Blue game, when the next version is released.

Peace.