Another endorsement of Dan Connolly's recent proposal

This conclusion to my final comments on ionfosets is more of a 
general comment, so I separated it out.

What I really hope is that some variation of Dan's most recent 
admirably concrete proposals will be accepted. I like them (even with 
the "undefined semantics") far better than the proposal to make the 
standard fundamentally vague with talk of "contexts" that allow any 
implementation to make an arbitrary decision about what namespaces 
are canonically equivalent. I'll freely admit that this vagueness is 
somewhat limited; because it must be supported by the creation of a 
new URI scheme with a specified comparison algorithm, but it still 
makes namespace code fragile and potentially in need of updates 
whenever a new URI scheme becomes popular.)

I like Dan's proposal because it answers almost all of the questions 
about what the code we know that we need right now should do. It also 
carefully avoids more grandiose, but contentious claims that will 
invite disagreements about the perennially controversial philosophy 
of identity and naming. While these issues are important for deciding 
a set of questions that are clearly poised to become very important, 
it's nice not to have to prejudge those questions now. I can disagree 
with Dan's philosophical premises and design priorities and still 
endorse his proposal.

That's what I'd call a good sign of a workable compromise. I wish I'd 
seen more favorable reactions to that posting, because if we could 
just agree at that level, then progress could continue, and the 
philosophy and the future could take care of themselves.

   -- David
-- 
_________________________________________
David Durand              dgd@cs.bu.edu  \  david@dynamicDiagrams.com
http://cs-people.bu.edu//dgd/             \  Chief Technical Officer
     Graduate Student no more!              \  Dynamic Diagrams
--------------------------------------------\  http://www.dynamicDiagrams.com/
                                              \__________________________

Received on Friday, 23 June 2000 19:50:15 UTC