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

RE: [syndication] Compromise time....



> Okay, after looking over the past few days worth of discussion, it's
> clear that there are a few points of contention.  And I don't see a
> way of convincing either side to change their position.  I'd like to
> enumerate the big ones and propose a compromise:
>
>   1. Having a well-known fall-back URL, such as http://blah.com/feeds.xml
>
>   2. Using OPML to represent the feed list.
>
> I now agree that #1 is not worth it.  HOWEVER, I would like to pursue
> the possible use of robots.txt as suggested by Chad.  Does anyone here
> know any robots.txt "experts"--people who were involved with the
> discussions back then?  We should try to get an authoritative answer
> on whether it is (1) possible to do what we want there, and (2) not
> completely unreasonable to do it.  If the robots.txt option doesn't
> work out, then it doesn't work out.  We tried.
>
> Does anyone disagree with that?

It's hard to see what harm it could do if only considered a fallback. Having
to add a parser to deal with it should act as a bit of damping to stop
people getting carried away (folks already parsing robots.txt should be damp
enough already ;-)

Re. experts - Tim Bray and Norm Walsh have both been involved with
syndication technologies, and are on the W3C's Technical Architecture Group,
may be worth a ping.

I think it should be possible to support multiple formats, though there
ought to be a preferred format (I agree about OPML being a poor choice).
Something like OCS would suite me best, but as people are actively working
on alternatives, I couldn't go further.

> Using the list of required data that's there now, I hope to put a
> simple example file (as a proposal) on the Wiki later tonight if I
> have the time.  Might others do the same?  And maybe ping the list
> when there's a new one to see?
>
> I suspect that putting a simple example file together will help me
> better understand what we're creating.  I wonder if the same might not
> be true for others?

Sounds a good plan.

Cheers,
Danny.