That model is not how we have worked as a community, nor do I think it is a productive way to run a smaller developer community such as ours. Doing things that way gives me a "all animals are equal, but some animals are more equal than others" feeling.
If you want to deploy things quickly, make them optional that those who want to use them can try them out to gain experience. If it is useful, then people will naturally take it up and want it to be standard. However, I still want this removed outright from the experience I already have from seeing it, and I see no way to gain any benefits. At the very least, can we agree to not make these labels automatically added. Let those who want to use them use them to see if people find a benefit to them. Travis On Friday, May 10, 2024 at 11:02:19 PM UTC+9 Matthias Koeppe wrote: > FWIW, I suggested to implement this feature in > https://github.com/sagemath/sage/issues/37254; I'm thankful to Aman Moon > for implementing this feature and Sebastian Oehms for his help with it. > > Obviously a metric such as the number of lines of changes is only a > one-dimensional way to express the complexity of a PR. > When I suggested the feature, I explained the possible positive effects: > - A size label "tiny" could encourage quick reviews of trivial changes. > - A size label "huge" could help flag problematic PRs. > Personally I think that the size labels for "medium-sized" PRs do not add > much and could be removed. > > But I'll note that "developer experience" improvements like this one are > really best developed exactly as it was done here: By deploying them early, > the developer community can gain concrete experience with them -- and then > suggest and implement refinements based on the experience. Harsh dismissals > of the whole features, on the other hand, are not very helpful. > > Matthias > > On Thursday, May 9, 2024 at 2:46:17 PM UTC-7 Travis Scrimshaw wrote: > >> I am *very* strongly opposed to these tags. Their cutoffs are arbitrary >> nor they serve no useful purpose as far as I can tell. To this point, they >> do not reflect the difficulty of a review; in fact, they are at best >> counterproductive to finding reviewers because it might deter people from >> reviewing "large" or "huge" changes as they can include lots of trivial >> doctest changes. At best it is just additional clutter in all of the >> information for PRs. >> >> From a community perspective, I feel such changes should have been >> brought to the attention of sage-devel once the PR was at a positive >> review. Specifically, *before* the PR was merged. Not everyone has time >> to read every PR, and a small consensus of developers might not reflect the >> development community at-large when making changes like this. >> >> Best, >> Travis >> >> >> On Tuesday, May 7, 2024 at 3:12:27 PM UTC+9 seb....@gmail.com wrote: >> >>> Dear Sage developers, >>> >>> You may have noticed that since yesterday a new type of labels with the >>> `v:` prefix has appeared on our PRs. These are automatically set to >>> classify PRs based on their size. For more information, see #37262 >>> <https://github.com/sagemath/sage/pull/37262>. >>> >>> Sebastian >>> >> -- You received this message because you are subscribed to the Google Groups "sage-devel" group. To unsubscribe from this group and stop receiving emails from it, send an email to sage-devel+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/sage-devel/85cc0f0f-e719-45ec-95fa-012906a3567dn%40googlegroups.com.
