💾 Archived View for gemini.susa.net › responses › re_spamtoberfest.gmi captured on 2020-10-31 at 00:50:02. Gemini links have been rewritten to link to archived content
-=-=-=-=-=-=-
Drew DeVault wrote about a marketing campaign of Digital Ocean and GitHub that encourages people to submit random unwarranted patches in exchange for a t-shirt.
Drew's experience seems to show that if someone has created a patch just to get a crappy t-shirt, there's a very high chance that patch will be almost, or entirely, worthless and a waste of mainainers time.
What this brought to mind was something I read relatively recently which was suggesting that there's some sort of 'accepted wisdom' for evaluating FOSS projects, based on project stats, and gave some handwavy heuristics on how to interpret these (e.g. ratio of open to closed issues, etc.).
It struck me that this would risk 'gamifying' the repositories - a source code repository turned into a fuckbook 'likes engine' instead of being a tool to store revisions of source code. Stars are already there, but gaming actual project stats would be a terrible thing.
Perhaps developers are being coerced into chasing pull-requests and issues like some people chase likes or karma.
As Drew points out, low quality pull requests and trivial issues are a sort of DDoS. I also wrote recently about the risks of developer burn-out being caused by things similar to this.
I think we should at least be on the lookout. If developers get sucked into gamifying repositories, they risk hanging themselves on their quest for vacuous project stats, and all for that enduring goal of increasing 'engagement'.