At 12:04 PM 4/2/2002 -0800, Doug Ewell wrote: >This is probably true, but James is also correct. Several people have >complained on this list, loudly and frequently, that leakage of >ACE-encoded names to old software clients is unacceptable.
There are always people complaining about some design choice. For a working group to make progress, it needs to deal with those concerns when they are raised. Then it needs to move on. It is not productive to have those complaints constantly added to the discussion mix. It is also necessary to insist that complaints be accompanied by solutions. The simplistic UTF-8 advocacy is simply wrong. And we have covered that ground many times over the years. (I hope folks note the repeated references to years. This was an urgent project when it started and we still have not begun deployment.) It is worth noting that none of the UTF-8 advocacy produced a complete specification. In particular they failed to deal with transition. This is not surprise because it is the fundamental weakness of the "just start using binary" approach. >I believe James was simply trying to set up all the scenarios and >establish what "works" and what "doesn't work." Yes, such an effort is essential for understanding transition of the installed base. However it is never productive to look outside the group that is being transitioned. Components that do something other than Old Ascii DNS and/or New IDNA have nothing at all to do with this working group. And, yes, I mean nothing at all. They are a distraction. We might as well also worry about printers that only do ASCII text or that do graphics. Some can print only Old Ascii DNS strings and others can print extended characters. So they are part of the wide world of internationlized character strings. However they have nothing to do with this working group. d/ ---------- Dave Crocker <mailto:[EMAIL PROTECTED]> TribalWise, Inc. <http://www.tribalwise.com> tel +1.408.246.8253; fax +1.408.850.1850
