On Monday, April 30, 2018 06:52:30 AM Chris Lamb wrote: > [adding Matthias to CC as he filed #883581] > [adding Holger to CC as he filed #886259] > > Hi Scott, > > Thanks for opening this! > > > I understand (and agree with) the intent of this check, but in practice, > > it's harming the release. I don't think it is salvageable. > > This tag has quite a history: > > - https://bugs.debian.org/870822 — Added as P: > - https://bugs.debian.org/883581 — P: → W: > - https://bugs.debian.org/886259 — W: → I: > - https://bugs.debian.org/897213 — removal requested (this bug) > > I suppose one question would be: why not downgrade it and/or make > it more explicit that you shouldn't blindly remove it? :) > > > There has been a lot of discussion on debian-devel about the future of > > python2.7 in Buster. The > > dependency-on-python-version-marked-for-end-of-life flag is confusing > > people into thinking they should drop python2.7 support from their > > packages because it isn't going to be in Buster. > > (Just out of interest, do you have any concrete examples? The > thread itself talks a lot about packages being removed with scant > citations.) jbicha already gave these, so I'll skip over it. I like the idea of pushing towards python3, but my impression of the near-term utility of this tag has changed. Specifically, I see people being confused about if python2.7 will be in buster (it will) and if it they should drop python2.7 support from packages that still support it (they should not). While I originally didn't think so, I've become convinced that there's no wording short of something like "Ignore this tag for buster" that would fix it. At that point, I considered if the tag was useful no and concluded it is not. That's not what I thought originally. People pay attention to lintian results and so the tags should be actionable. The action most people seem to read into this tag is get rid of python2.7 support and we're one release too early for that. Scott K
Attachment:
signature.asc
Description: This is a digitally signed message part.