One of the sessions on Sunday morning at FOO Camp was a
brainstorming session on how a site could provide a list of
feeds.
This session was initiated by Jeremy Zawodny. I didn't
capture a list of participants, but from memory some of the more
vocal people were Tim Bray, Jeff Barr, Dirk-Willem van Gulik, and
Dave Sifry and, of course, Jeremy himself.
A working name for this effort is "FDML". The stands for
Feed (Discovery / Directory / Detailing) Markup Language, depending
on who you ask. ;-)
We did not get to the point where we created a concrete
proposal, but here are some of the use cases that were
discussed:
A single site provides a number of feeds that subset the
information by subject
A single site provides a number of feeds that contain the same
information, but in a variety of formats
A site provides a list of feeds that reside on a variety of
hosts that are somehow related (example given was "Microsoft
bloggers")
For extremely large lists, this information may need to be
organized hierarchically
Here is some of the requirements that I captured for a description
of the list itself:
Version information
Title of the list
Description of the list (perhaps optional?)
Parent (where to find a list which contains this list)
Children (a way to organize extremely large lists)
Generator (information on the tool that produced this
list)
And here is a list of requirements for the description for each
feed:
title of the feed
link (where to find the feed, relative to the list)
description of the feed
unique id of the feed (feeds that differ only in format or
version would have the same id)
In order to further this discussion, I have created
a wiki and
seeded it with this information. Jeremy also indicated that
he was going to set up a mailing list.
Update: Apparently between the time I took the notes and
posted them, Dave Winer posted a
myPublicFeeds.opml
RFC.
An observation. Seems like you're doing it again. First it was Atom because you didn't like RSS. Now it is FDML because you don't like OPML. Isn't is possible to extend the existing stuff out there? Do we have to start over?
Today, both Sam Ruby and Dave Winer release two different proposed solutions to the same problem (A standard way for a site to refer to a list of feeds), both on behalf of Yahoo's Jeremy Zawodny. From Sam: FDML. From......
[more]
I would be interested to know how FDML might correlate to or even utilize some ideas of the Open Archives Harvesting Protocol (OAI-PMH), http://www.openarchives.org/.
In the case of OPML, yes. It would be lovely to have something serving the purpose of OPML that can actually be run through a validator. Something that is actually XML. Something that isn't from the school of "hey guys, just make up your own tags if you don't like ours".
Try writing a DTD or an XML schema for OPML. It's not as easy as it may first seem.
Then again, don't you think "a list of feeds" should be integrated with "a list of services" that the weblog software provides? You could see the feeds as a service as well, after all.
So I woke up this morning to discover that Sam Ruby and Dave Winer are talking about the same thing, in a different manner. Guys, I already wrote about this! Sam is inventing a new format (FDML; wiki here); Dave......
[more]
Just so it is absolutely clear: all I have done is listed a set of requirements, many if not most of which are directly from Jeremy himself. The acronym was suggested by David Sifry.
People are welcome to question, refine, or add to the requirements, or present proposals on how these requirements can best be implemented. Perhaps even with OPML.
FDML: At O'Reilly's recent Foo Camp, some bloggers started fleshing out an idea for RSS feed self-discovery. Kind of like WDSL for RSS — you can get an XML document that explains all the feeds that a particualr site offers.......
[more]
Yesterday afternoon, Dave Winer published a request for comments (RFC) on how to best provide a list of feeds that are available on your site. The result of this RFC? The generation of another OPML file, somewhat akin to the use of mySubscriptions.opml. The name of this new file should be myPublicFeeds.opml. I like Dave's proposal for at least three......
[more]
Every time I see one of these proposals I remember the sating "He who writes the code, makes the rules".
I suspect that it is more important to get the authors of the more popular aggregators like NetNewsWire, SharpReader and Newsgator behind an idea than posting competing RFCs and starting up wikis. Then again looking at your list all I see is a bunch of features but no description of the functionality that you want. In my short years working on software this is typically an alarm bell. You start with functional requirements and use cases not details like what kind of information you want in some XML format.
Dare: Yahoo, Syndic8 and Technorati intend to be a producers of this (an possibly consumers). They got together to discuss a format. It sounded interesting, so I chose to listen in. I endeavored to capture the use cases and requirements, and publish this for all to see. Those that wish to participate, can. Those that don't don't.
One thing that was stated in passing that I didn't capture or follow up on - while this could be used on sites that contain blogs, that was not necessarily the primary intended audience. I gathered that MSDN was more of what they were thinking of.
Technorati, in particular, is tracking over a million feeds. The challenge is how to organize this information.
In any case, you are welcome to participate, if you so chose. Or not, if that is your inclination.
Sam,
The question is who is the intended audience? If this is just so large directories can share information with each other that's one thing if on the other it is for end users to be able to use it to import a list of feeds then it is another. If it is the former I can see why you'd want to go down the complicated route you describe if it is the latter I don't see why Dave Winer's myPublicFeeds.opml isn't good enough.
If the problem is categorizing information, this seems to be different from a syndication problem and something more akin to what the folks at DMOZ use (http://rdf.dmoz.org/) may be more acceptable.
This may or may not be of interest to me depending on what the intended use cases are. So far I haven't really seen any.
As my day starts to wind down, everyone else is waking up: Dave Winer feeds back sensibly on comments that I and others left on his original RFC; Sam Ruby clarifies that he wasn't necessarily proposing a new markup; Dare......
[more]
If users could just point the aggregator at the main page and start selecting which feeds they want, I think that makes their experience a lot better than it is at present.
Dare, a DMOZ like categorization scheme was definately along the lines of what Jeff Barr was thinking of. However, one where the owners of the feeds suggested the categories. One of the ideas tossed out was that creators of feeds be encouraged to add feed level subject information for categorization.
I'm not sure this is any differen than OCS(open content syndication). Ross Rader gave me a demo of it, it's a self discovering XML feed that locates (heirarchically) the RSS feeds in your site. Take a look:
During last weekend's, Foo Camp a group of luminaries got together to discuss feed auto-discovery. Sam Ruby took notes and what they are calling FDML (Feed Discovery/Directory/Detailing) Markup Language) and has posted them on a wiki. (My favorite.) I revised [my previous example] to bring it closer in-line with the requirements they listed....
[more]
OK, so it seems he wants a RSS autodiscovery that points to a list of links instead of a single RSS feed. Seems simple enough and I'd say that Dave Winer's proposal hits the sweet spot for that except for it uses a well known URI instead of the <code>link</code> tag like RSS autodiscovery does.
I could probably implement this overnight and it seems fairly useful enough. I'd assume this could be added to Step 2 in Mark Pilgrim's Liberal RSS Autodiscovery algorithm[0] which RSS Bandit already uses.
I agree with Derek Balling [who criticized Foo Camp], and when you come back to earth, I bet you will too Jeremy. Did I read that you guys had meetings about RSS? At a private invitation-only event? Do you realize how WRONG that is? Source:Dave...
Yesterday afternoon, Dave Winer published a request for comments (RFC) on how to best provide a list of feeds that are available on your site. The result of this RFC? The generation of another OPML file, somewhat akin to the use of...
Okee, let op. Er zijn mensen bezig een standaard te verzinnen voor lijsten van feeds. Een manier van aangeven wat voor publieke (RSS, etc) feeds een site beschikbaar heeft. Dave Winer werkt aan een specificatie voor myPublicFeeds.opml, (zie ook...
Dave has noted the beginnings of his spec for RSS autodiscovery. He has gone the OPML way, and basically his discussion mirrors what I wrote regarding the topic and the choices we faced about a month ago. Jeez. A month! A month! What is up with time...
When one of the other developers says "Ahh, it's all coming together!", why does my brain make me reply with "Like the stars of a Penthouse Forum letter?" Why? Because it is against me.Meanwhile, I had to make myself read the...
Dare wrote a post about the Top 3 Features I Want To Add To RSS Bandit, which all sound good, but there's a quote at the end I want to pick out: I'm completely bored with doing the same old shit or even worse trying to reinvent the same old shit...
I originally covered this subject back on February 27th, 2003, the same day that I announced ongoing to the world. I think it’s worth revisiting, because it sure would be handy if there were such a thing as a Web Site, as Dave Winer, Sam Ruby, and...
A lot of buzz about enabling feed autodiscovery. The idea being that a newsreader/aggregator could be pointed at site (e.g. a blog) and it would be able find any non-HTML data files such as Atom feeds without leaving buttons all......
Kalsey Consulting Group acquires the BlogSnob text ad exchange; Jacques Distler offers some relief to Movable Type users hit by comment spam; Max Design demystifies CSS floats in Floatutorial; emusic.com annouces that they’ll no longer offer...
The conversations inspiring (and inspired by) FDML, a proposed “list of feeds” format, are quite interesting, especially those debates concerning feed auto-discovery. In September Jeremy Zawodny noted two possible auto-discovery...
Furthering the discussion involving Tim Bray (where I picked it up), Dave Winer, Sam Ruby, Jeremy Zawodny, and Joe Gregorio, I started picking at the issue Tim brought up: The problem is that all the Web knows about is URIs, and the Web can’t...
Furthering the discussion involving Tim Bray (where I picked it up), Dave Winer, Sam Ruby, Jeremy Zawodny, and Joe Gregorio, I started picking at the issue Tim brought up: The problem is that all the Web knows about is URIs, and the Web can’t...
Furthering the discussion involving Tim Bray (where I picked it up), Dave Winer, Sam Ruby, Jeremy Zawodny, and Joe Gregorio, I started picking at the issue Tim brought up: The problem is that all the Web knows about is URIs, and the Web can’t...