[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [syndication] Some suggestions for RSS .92

At 04:06 AM 10/15/00 -0700, you wrote:

To Tristan, I'm uncomfortable voting for the <date> element unless there's a
spec that says how it fits into the 0.91 spec. The actual proposed language
for an 0.92 spec would be great.
I was thinking that we would go through the different tags and vote on each one individually before going into more details and specing them out. That way, we can define what tags we want to look at and then spec them out in a very detailed fashion. I thought this might work best since it would allow every to have a say as to what tag goes in or not and then, once that is done, we can discuss the details. Technically, the first poll I should have suggested should have been: do we want to create an RSS .92 format.
Also, of note. Since RSS is moving to 1.0, should we rename the new effort (what we now call RSS .92) something different? I was thinking of something along the lines of one of the following: 

RLS: Really Light Syndication
VSS: Very Simple Syndication
SSS: Super Simple Syndication

Otherwise, we might find ourselves stuck when we try to go to 1.0 or later.


Jeff I'm not an expert on Dublin Core either, but I suggest we not go down
that slippery slope. Keep it simple.
Furthermore, categorization seems to be addressed by the efforts in RSS 1.0. Based on use of namespaces, they are working on a taxonomy that includes categorization through RDF and the Dublin Core. 


Also, taxonomy is a big subject. The last time there were serious open
discussions about next steps in RSS (in the May-June timeframe) this was put
on the table by Dale Dougherty at O'Reilly, and we had done some
experimentation on it at UserLand. One of my jobs for this morning will be
to dig up that thread and see what we did.


This is actually a hot subject right now on the RSS-dev list.


I also want to work out a simple XML-RPC/SOAP integration, I want to make a
channel subscribable, so that the publisher calls the subscriber when
there's an update.
Question: Wouldn't that add some complexity? And does it have to be within the RSS spec itself or can it be a tool that looks at the RSS changes and then pushes it. 


Finally, I had a meeting with the ICE Working Group a few weeks ago. I'm
going to help them come up with a "lite" version that is philosophically
compatible with RSS 0.91 and any successors (but not the
Namespaces-RDF-modular thread, which I don't see as being philosophically
compatible with RSS 0.91). They don't want to adopt RSS syntax, but they
want to offer current RSS syndicators an easy way to participate in their
networks. I think this is a good goal. They have an ICE "cookbook" which is
easy to follow, I've asked for an HTML version of this document, and expect
that we'll simplify it even more, and see where we can take it.


That sounds exciting. I can't wait to hear more about this.