RE: [xml-dev] Expertise and Innovation - was Re: [xml-dev] Non-Borg servers can authenticate Borg clients (Was Re: [xml-dev] Re: Cookies at XML Europe 2004 -- Call for Participation)

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

Killer content would be better but again, look at the 
top ten search terms and despair the rule by power law.  

As to expertise and innovation, I remember the old saying 
that "One knows God loves fools because he made so many  
of them."   Innovations from experts are everywhere but 
no one gives them much press.  When the average idiot 
makes good on an idea, that's a great news day because 
no one bets on an idiot until the idea is obvious. It 
isn't 80/20ness; it isn't implementation; it isn't the 
source or the destination:  it is the obviousness.  

Simple is the hardest work we do.  Musicians know it and 
probably most artists; less is more. But you can't do less 
until you have done too much.  Most of my time in the 
studio is spent throwing away parts until I have only 
the ones I can't afford to throw away.  When the test  
audience says, "very nice but obvious", I'm done. The 
rare case is the untrained amateur songwriter who 
copies a chord progression and has a hit.  Quick, how 
many hit songs have the same chord progression as 
"Happy Birthday"? (For the literate:  I V I V I IV iv I V I)  

As I said, the average idiot can be successful and there are 
lots of them. I won't bet on one until I hear the song. 
A hit is well.... obvious.

The question is if a success will keep on being successful. 
Beethoven was the pop art of his day. Today he is a 
museum piece or at least, playing his work at a club 
will get me fired.  Jazz did well when the blues was too 
black.   Today?

Are technical innovations free of the fates where 
power laws spin the threads, pundits measure them 
but trends snip them?  Are the daughters of necessity 
also the masters of the technical lifecycle?
  
Mosaic and the web browser in general made the web go, 
and the pundits said it would ever be so.  But the heyday of the 
web browser is ending or so it is said.   Long live 
the PC Client for where there might be one icon before, 
now there will be many each with its own non-interoperating 
XML format.  Standards schmandards, I want candy.

What isn't waning?  HTTP and URLs.  

So you have a case for the architecture over the implementation. 
I have to wonder if the web arch document would have 
resulted in a success in 1991-93.  I doubt it.  A URL 
made a lot more sense than a URI but 20/20 hindsight 
and all that implies.

The only predictor of long term success is that an idea 
or innovation reproduces into the next generation.  The 85% of 
XML that works unarguably is all SGML, so SGML is wildly 
successful even if only mildly innovative. As to money, 
no one made money on jet technology until its main innovators 
lost the war and changed their citizenships.

Meanwhile, those VRML people keep trying.  
Clotho, Lachesis and Atropos laugh at our feebleness.

len



From: Joshua Allen [mailto:joshuaa@microsoft.com]

Yeah, that's right.  I think these architectural principles behind URIs
were just as important as having a killer app.