Re: Are the chapters of webarch in the wrong order

On Mon, 2002-12-09 at 16:14, Tim Bray wrote:
> 
> Working on my slides for my speech at XML2002 (the mainline plenary one 
> whch has a lot of web architecture, not our TAG event) it dawns on me 
> that the current order in the webarch draft (identification, 
> representation, interaction) might work better if recast as 
> identification, interaction, representation; here's what there is, 
> here's how you get it, here's what you get when you get it. -Tim

On the flip side, you can look at most of the identification
and representation issues in Web Architecture by pretending
the whole web is on a local CD-ROM; i.e.
no time-varying stuff, no distributed computing issues;
but you do have the issue that
you'd like to be able to put the same CD rom into
computers that run different platforms and such.

Then you can sorta "look under the hood" and see
that the CD-ROM actually consists of this world-wide
network of machines and people and such.

That was a rationale that TimBL gave me for putting
the protocol/interaction issues last. I find it
appealing, though not so compelling as to 
rule out other approaches.

-- 
Dan Connolly, W3C http://www.w3.org/People/Connolly/

Received on Wednesday, 11 December 2002 13:20:27 UTC