Re: [syndication] Re: Thoughts, questions, and issues.

[Gary Teter <bigdog@bulldogbeach.com>]

on 8/21/00 3:28 PM, Aaron Swartz at aswartz@swartzfam.com wrote:

> Seth Russell <seth@robustai.net> wrote:
> 
>> Well i might be wrong here, but i don't see any reason why  RDF
>> couldn't deal with the old <channel><item/><item/></channel>
>> construct.  It struck me as very odd that the authors separated it
>> out.  Why did they do that?
> 
> This is becoming a bit of a FAQ. Once again, we want
> backwards-compatibility. This is the way it was done in RSS 0.9 (for no good
> reason) and we didn't want to change it. This was fixed in RSS 0.91 (which
> most of you use) but the new spec is based on 0.9. Does that make sense?

What doesn't make sense is why the spec was based on 0.9 instead of 0.91.
The vast majority of RSS sources are in 0.91, not 0.9. Honestly, this part I
don't understand at all, unless I put on my "geek hat" and realize that the
0.9 spec was in some ways cleaner than the 0.91 spec.

But if you're looking for backwards compatibility, I'm sorry, I don't see
it. Why not be backwards compatible with the version that's most widely
deployed?

RSS is in a lot of ways a "populist" spec. This is a point that's been made
more eloquently before by others, so I won't belabor it here, except to say
that I think the decision to base RSS 1.0 on version 0.9 instead of 0.91
feels kind of "ivory tower" to me.

(OK, I've chimed in enough without offering anything helpful, so here's a
suggestion: RSS 1.0 should be the same as RSS 0.91 except with a
quasi-standard set of namespaces. Those who want RDF can add it in "for
free", to paraphrase Ken. Oh yeah, and get rid of skipdays/skiphours. :-)

-- 
Gary Teter, Big Dog
Bulldog Beach Interactive http://www.bulldogbeach.com
"Fun will now commence." -- Seven of Nine