xml-dev - Re: A certain difficulty

[Len Bullard <cbullard@hiwaay.net>]

Michael Champion wrote:
> 
> On the other hand, if this stuff is so powerful, how come nobody out there
> has used it in a fabulously successful project/product that makes us stand
> up and take notice?

Dr Newcomb provided a list of successful applications of AFs.  Now this
is 
an idea that has been around for awhile.  I can say with certainty that 
had we used at least the simpler parts of groves and the grove
methodology 
to start the X3D project, we could have saved ourselves a very nasty set 
of encounters.  

o  There is probably truth in the NIH charges in all of the tribes

o  He may be right that you aren't meant to take notice.  

Given the bizarre economics of the web (the so called, populist market) 
if I were an investor and thought you had an edge, I would personally 
brain you for talking on these lists about it.  Rising tides sink  
boats if they are tied to the pier.  Poverty and debt start knife
fights.

Weirdly, it IS easier to build a complicated spec that meets all the 
requirements, then create the "simpler/easy to learn subsets" if one 
has the time and some hack doesn't take the field in the meantime.  
Again, the phrase, "internet time" like all "hurry up before the 
competition gets traction" phrases is foolish.   Bottom up or top down, 
at some point, the bloody thing evolves as the requirements do.  
What I see in most W3C specs is requirements creep.  I sympathize. 

RDF and X-schemas:  talk among yourselves and figure out just for 
grins, what the absolute simplest subsets are and dink with those.
But make sure before you do that, you have a matching and very 
precise set of requirements.  In effect, make a table with three 
columns:

Requirement    SpecFeature   Example

Do that for awhile and pretty soon you will figure out if the 
spec is too complicated.  Because you have to give an example, 
fuzzy requirements are exposed.  If you add a fourth and fifth column

Alternative   Example

where you put other features from other specs, proprietary 
or open, you will also see some simplifying solutions.

len



***************************************************************************
This is xml-dev, the mailing list for XML developers.
To unsubscribe, mailto:majordomo@xml.org&BODY=unsubscribe%20xml-dev
List archives are available at http://xml.org/archives/xml-dev/threads.html
***************************************************************************