Hello Sam, On Thu 12 Aug 2021 at 05:35PM -06, Sam Hartman wrote:
> I thought I provided such an argument. > (you trimmed that part of my message when replying). > My argument was roughly that things like build systems, use of dh, > debian/rules interfaces etc might well need to apply to source packages > producing only udebs. > I think the issues are kind of complex, and I agree with you that we > didn't consider udebs properly. > > I do think though that you ignored the meat of my message and I think it > is worth a bit more consideration than that. Let me first apologise for seeming to dismiss the rest of your message. Let me explain more. In your message you expressed that you don't know a great deal about why udebs are udebs not debs, and I am in the very same position. Thus, I have been thinking about this bug mostly procedurally. We made S-V mandatory based on the usual reasons we have for making changes to Policy, such as prevalence of the S-V field in the archive. However, these reasons typically don't take account of udebs. So now it feels like Policy has unintentionally barged in upon the work of the d-i team when what we were actually trying to do was make a change that was only about the regular archive. It seems to me that we ought to back out the unintentional extension of the scope of Policy before we properly consider whether it would be useful to extend it in this way. Rather than dismissing them, I was just intending to delay consideration of your arguments until after this mistake has been undone, and we can have a proper discussion with input from those who know more about udebs. I should have said this earlier, so once again my apologies. -- Sean Whitton