On Apr 6, 2004, at 1:39 AM, Graham Klyne wrote: > > Hmmm... my concern is that the unqualified claim of "strictly avoiding > false positives" is also not factually true [1] and, if generic URI > processing software is developed on this basis, could lead to > incorrect results. Is there another way to address this? Hmm... I think that the goal of "strictly avoiding false positives" is indeed correctly stated. To the extent that if there are practices that would result in software that "climbs the ladder" in good faith getting one, then those by definition are bad practices. Today I'll dig up the latest draft and review > [1] http://lists.w3.org/Archives/Public/uri/2004Feb/0094.html > (Look for "Section 6 (concern):") and try to think of an improvement. False positives on URI comparison would be very damaging to webcrawlers and caches and so on and should really be avoided. -Tim
Received on Tuesday, 6 April 2004 10:46:01 UTC