[XML-DEV Mailing List Archive Home] [By Thread] [By Date] [Recent Entries] [Reply To This Message]

RE: Best Practice for URI construction?

• To: 'Elliotte Harold' <elharo@m...>
• Subject: RE: Best Practice for URI construction?
• From: "Bullard, Claude L (Len)" <len.bullard@i...>
• Date: Mon, 19 Dec 2005 10:34:39 -0600
• Cc: 'Peter Hunsberger' <peter.hunsberger@g...>, "Costello, Roger L." <costello@m...>, XML Developers List <xml-dev@l...>

True statements all.   I use VML because the dll is sitting 
there.  OTOH, it is dead and that isn't good when it comes 
to figuring out its bugs.  

Although web services still win business where based on RFPs.

Market trends: good development companies being taken over by 
MBA managers are going to third parties for niche pieces that 
turn out to be critical.  This is a cyclic behavior.  If the 
pieces are solid, it is growth for the little guys at the 
expense of the big guys who erode their talent pool to meet 
commitments made by dealmakers who get all of their reward 
on the front end.   See the history of NASA.   The result is 
increased emphasis in contracts on liquidated damages clauses. 
If there are problems, the companies find themselves in damage 
control mode that never lets up as the customers become skilled 
at extracting more work for less money.  As the available pool 
of free resources shrink, growth and market momentum decrease 
proportionally.  Because the deals are forward-looking, a short 
burst of market share turns into an anchor of legacy commitments 
and the velocity needed to reach escape decreases.  Crash and 
burn or low-Earth orbit is the best that can be expected.

For that reason, pick simple solutions whereever possible that 
can be quickly ripped out and replaced.  Fast evolution of 
components increases the bugs unless one is paying a lot of 
attention to certification.  The integrator companies take on 
that responsibility as primes for the little guys.

len


From: Elliotte Harold [mailto:elharo@m...]

Bullard, Claude L (Len) wrote:
> I think there is a widening chasm between the folksonomic 
> application of web technologies and their specifications. 
> It begins to look like music theory: what the books say 
> is correct and appropriate versus what the performer and 
> composer know just works.
> 

Maybe, and maybe not. Lately I've noticed several cases where the 
Politburo triumphed over earlier, looser code-what-works approaches:

1. REST is slowly winning converts from the web services camp. Compare 
APP vs. the Blogger API. APP may be the protocol that finally shows the 
world what REST can do.

2. Well-formed RSS and Atom is far more common than it used to be. 
Escaped HTML is dying.

3. SVG trounced VML, and is making some small inroads against Flash.

4. CSS is finally preferred to tables for layouts among the 
professionals. This meme is spreading to the amateurs as well.

When the Politburo has good reasons for what they decide, they can often 
win in the long term. Of course sometimes they're just out to sea too 
(W3C XML Schemas) , as are we all at times. There are still some areas 
(Canvas vs. SVG, WebForms 2.0 vs. XForms) where I expect the wisdom of 
the Politburo will eventually be confirmed. However sometimes you have 
to let people dig themselves into a wet, muddy hole before they're 
willing to invest in your improved hole digger. :-)

• Follow-Ups:
  • FInding encoding
    • From: "Ravi Shankar Nair" <sujashankar@p...>

• Prev by Date: Re: Best Practice for URI construction?
• Next by Date: Re: Best Practice for URI construction?
• Previous by thread: Re: Best Practice for URI construction?
• Next by thread: FInding encoding
• Index(es):
  • Date
  • Thread