RE: [xml-dev] RAND issues

["Bullard, Claude L (Len)" <clbullar@ingr.com>]

I mean things more like HTTP protocol support, 
or how many XML parsers do we need,  
nothing more sinister.  We definitely will have 
a semantic problem with "low level" and "high level". 
For example, is XML low-level or high-level?  SVG 
seems to me to be high-level but turning around 
and looking at GML, it would be low level compared 
to feature sets built up for specific location 
dependent services.   

The W3C can focus on the high level, but the numbers 
of viable implementations still tends to be driven 
by the numbers of viable companies that can implement 
them of the tenacity of a given open source group. 
Again, those with IP will have to make the decision 
to seek a standard or go their own way.  It certainly 
doesn't hurt Adobe or Macromedia which depend 
on colonization.  All of these strategies are viable 
and I wouldn't count on bad press, public pressure, 
or anything else to change that.  Even small 
vendors choose proprietary if it suits their business 
model.  There are politics and markets and when 
these overlap as standards, que bueno.  When they 
don't, don't expect the BigCos to fall on their 
swords.

len

-----Original Message-----
From: David Brownell [mailto:david-b@pacbell.net]
Sent: Friday, October 05, 2001 4:15 PM
To: Bullard, Claude L (Len); xml-dev@lists.xml.org
Subject: Re: [xml-dev] RAND issues


Len,

>     I hate to see the W3C reduced to creating 
> only specifications for low level technologies for 
> which there are only ever a few implementations needed.

I don't think anyone thinks that's happened, or is asking
for that to happen.  (Did you leave out a "would"?)

The concern is more that W3C should focus on
technologies for which multiple implementations
are expected and desirable ... like, for example,
many "low level" technologies, and not a small
number of higher level ones!

The "only a few implementations" is a business
strategy chosen by some vendors as a way to
preclude or minimize competition.  That's not
really what standards are all about.  Even if
folk advocating patents can licence patents
separately from implementations which use it.