"Operational approach" to web architecture

My hope is for TAG activities that improve the consistency and interoperability of the many  diverse specifications being produced within the W3C, and between W3C publications and  those of other standards organizations.

One way of improving consistency is to provide architectural guidelines that provide definitions of terms and describe the general principles by which the web works, in those terms, in sufficient detail to be useful guidelines to the working groups. The utility of the architecture in that role is a primary evaluation criteria to apply to TAG work. 

That's what I meant when I said I preferred an "operational approach": judge definitions by whether they're useful.

Recommendations are useful if they are implemented widely and result in interoperability. Definitions and architecture are useful if they help people write useful recommendations.

If you can't answer whether "it is correct for a web server to return a 200 OK response code in a response to a request for a URI which identifies a non-information resource."
then you don't have a good enough definition of "information resource" vs. "non-information resource". 

# I think Tim has something in mind when he says information resource
# and that job #1 is to craft a definition that accurately, and in a manner
# that it can be conveyed to others, captures the sense of it. 

I don't agree that this is job #1. 

Job #1 is to take brilliant inspiration and turn it into operational specifications that result in interoperable implementations; something I and many others have spent many years on. 

Merging the semantic web and the browser web is a brilliant inspiration; making it operational will require useful definitions of terms to make progress.

Meanwhile, there are many other architectural difficulties and irregularities surrounding the W3C, and I hope the TAG will prioritize its attentions wisely.
 
Larry
-- 
http://larry.masinter.net

Received on Wednesday, 28 January 2009 18:10:01 UTC