While I have no stance on the meme, I’d like to point out that just looking at raw numbers isn’t a sufficient counter argument for multiple reasons:
Many projects use PRs to merge feature branches from maintainers, while outside PRs are left lying around
I’ve seen people close their PRs in other projects after sitting around for a long time. How many of those 19,074 were withdrawn?
The Mastodon repo used to have the stale bot enabled (might still be), e.g. in this random example. How many of those 19,074 were automatically closed?
They also use the renovate bot. How many of those 19,074 were automatically submitted?
I actually went through and checked how many PRs were automatically submitted, we have:
dependabot-preview: 1,107
dependabot: 2,279
renovate: 1,126
github-actions: 270
There might be more that I’ve missed (e.g. dependabot-preview isn’t listed as an app in the authors filter), but this adds up to 4,782 PRs, just over a quarter of total PRs. That’s a rather large deviation.
Again, not saying that you’re incorrect, but the numbers you mention could also occur in a project that is as bad as the meme describes.
While I have no stance on the meme, I’d like to point out that just looking at raw numbers isn’t a sufficient counter argument for multiple reasons:
I actually went through and checked how many PRs were automatically submitted, we have:
dependabot-preview: 1,107dependabot: 2,279renovate: 1,126github-actions: 270There might be more that I’ve missed (e.g.
dependabot-previewisn’t listed as an app in the authors filter), but this adds up to 4,782 PRs, just over a quarter of total PRs. That’s a rather large deviation.Again, not saying that you’re incorrect, but the numbers you mention could also occur in a project that is as bad as the meme describes.