RE: [xml-dev] "Introducing MicroXML, Part 1: Explore the basic principle

["Len Bullard" <Len.Bullard@ses-i.com>]

The difficulty, Peter, is you are in the well-known trap of classical
liberalism vs social liberalism.  To preserve the greatest amount of
flexibility for the one or few the governing authority makes the fewest
rules possible (aka, smallest government) rather than pursue the social
liberalism approach of the rules all must obey for the greatest
efficiency.  The outcome is given the fewest rules the most effective
group takes over making the rules.   On a pirate ship (where pure
democracy is the ideal), the secret crew takes over once the rest of the
crew has obtained the prize.  It's inevitable:  an elite takes the
spoils.   This isn't unnatural.  It is exactly what one can illustrate
in the majority of civilizations where common resources have to be
managed and distributed by cooperative consent.  See the Mayans.

Technically, what is now current is right.  Unless XML is kept free of
application semantics then the semantics chosen govern all applications
that use XML.  In practice, the difference is moot.  XHTML rules the
roost until HTML5 does away with XHTML and thus, XML.  And then someone
will knock off HTML5.

As long as the web itself IS an application (which in ISO layer
architecture, it is), whoever governs the web sets those rules.  This is
not TimBL, the W3C or any of the others you mention but the mostly
anonymous browser designers at a handful of companies and their moneyed
customers. 

The very people you are asking to vote have the least interest in making
changes and really, not much power to make changes that large.  They can
tinker but that's about it.  Your scenario between Roy and Tim is ironic
because the W3C and its membership have organized it to prevent exactly
that scenario from occurring.  Spend time on the Web Architecture
Working Group list.  An organization that avowed to perform in Internet
Time discovered that Internet Time is a myth.  It enabled them to reduce
ISO influence and thus obtain some political objectives, but once an
application is fielded in the large, it becomes a honeyed treacle
through which any fast march is considerably reduced in speed.  The W3C
succumbed to the very forces of market treacle it said it could
overcome.  These aren't hurt feelings.  This is market reality. 

The likely way this changes is for one of the current big companies to
discover a not too risky technical change that enables them to take
market share.  In game theory, it is a stag hunt and they have to decide
to chase a rabbit because the short term rewards outweigh the immediate
or long term advantages of continuing to cooperate.  The proprietary web
clients used by publishers have been attempting this but so far enjoy
only limited success.  The way they would succeed would be to replace
the web itself (build and field a different browser that doesn't use
HTTP even if it is technically a REST model) and that has a very small
likelihood of succeeding.

len

-----Original Message-----
From: Andrew Welch [mailto:andrew.j.welch@gmail.com] 
Sent: Monday, July 02, 2012 8:47 AM
To: Rushforth, Peter
Cc: liam@w3.org; David Carlisle; Len Bullard; xml-dev@lists.xml.org
Subject: Re: [xml-dev] "Introducing MicroXML, Part 1: Explore the basic
principles of MicroXML"

> I believe that one of the problems with XML 1.0 is the lack of support
for the web, so this will eventually kill any child if it's not
addressed.

XML has 'support for the web' in the form of XHTML, or any other
specific dialect.  There is nothing that needs to be added to XML in
general.  No one feeds their data bearing xml straight to the browser,
it always needs some processing before getting displayed.

The need for a simplified XML didn't come from the web, it came from
fellow developers struggling and getting annoyed with parsing XML
config files, or trying to write out some XML from a set of fields, or
trying to insert some nodes into some XML.

Nothing Peter, nothing.. to do with linking.  :)



-- 
Andrew Welch
http://andrewjwelch.com