RE: [xml-dev] When parsing speed matters (was Re: [xml-dev] No XML Binar

["Len Bullard" <cbullard@hiwaay.net>]

SL is taking care of that with a marketing budget and a zealot at CNet.  XML
is dull.  It needs to become more animated. :-)  As I said to Rob, people
post to their interests in XML.  The vast majority of people on XML-Dev
happen to be database/document wonks which is historically where markup has
thrived.  I migrated away from that and lo and behold, XML turns out to be a
marginal choice for some data intensive applications because speed really
does matter and so does verbosity where it impacts that if it does.  The
case for binaries in graphics pipelines is said to be secure.  I'm waiting
to see the results of the working group.

If XML hardware can speed up the pipes, that is worth doing, but so far, I
don't see that happening unless the game console folks decide to do it.  The
current specialty tasks don't seem to apply from what I can make of this
thread, and a game environment won't be ubiquitous enough soon enough to
make XML board commoditization of any value there.  Still, anyone who deals
with the effects of lag on real-time web-distributed simulations is looking
for any advance or advantage they can find now that online sims are a fast
growing business.

The realization that HTML browsers are slowing down the web is dawning on
the people who create persistent games.  If the HTML browser is getting in
the way of faster rendering, that is worth doing away with altogether.  If
3D games are the wave of the future for social networks and business online
work, then HTML starts becoming more marginalized in the GUI.  It takes
awhile to build up, but eventually someone releases a client that is a good
3D GUI but doesn't use HTML for anything significant.

One fun bit about online games:  up to now, gamers could blithely code to
the metal for every release.  Once a game goes online and uses the web
infrastructure, that quits being affordable.  Suddenly, game programmers
have to experience the *real* software business and they hate it.

len


From: James Fuller [mailto:james.fuller.2007@gmail.com] 
 
apart from performance aspects, isnt the point of these boxes simply
'commoditization'...e.g. eventually I will want to put my 'translate
to chinese' 'box in series.....adding value to the information being
transported (or 'encode all my xml into encrypted content',
etc...etc).

adding value ala pipelines is an easy story for marketeers to sell....

cheers, Jim Fuller

ps: I think Len is on a mission to somehow game google with 3d terms