On Sun, Jan 22, 2012 at 08:52:52PM +0400, Michael Tokarev wrote: > think of a case when you created > the array with your clock was set wrong (say, 500 > years in the future) -- it will never be checked > in this case...
Yep. > Yes there are easy ways to work > around this very case - eg, to verify if creation > date is in the future and ignore it But we avoid from the another surprize (subject of this bugreport): running check for today created array. I can resend patch, but it's too easy to fix things this way. > but there, we'll > have another issue: 500 years from now we'll risck > skipping one check... Not for sure. There is a probability to have broken clocks or to do server setup during of the first week of month. Which one is bigger? > I think it isn't worth the effort fixing it at all, > principle of least surprize should work best here > (so that we know that arrays are checked at the > fixed date, always). That's a good reason to not overcomplicate things. You're right. > After all, the whole thing is very minor. > Do you not agree? Personally, I agree with this decision (wontfix). But I'm not bugreporter, nor a package maintainer. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
