Re: [syndication] Re: Total confusion in RSS-Land

[Mark Nottingham <mnot@mnot.net>]

My .02 - 

Another reason it's hard to get something done with RSS is that there is a very
poor sense of who "we" are. The Syndication list isn't meant to have
deliverables like RSS; it's just a discussion forum where interested parties can
talk and get an idea of what's happening in this space.

I participate in a number of standardization efforts, both public and private.
Getting different companies as well as people to agree on something and then
publicly endorse it is a very difficult thing. In my experience, ad hoc efforts
and on-the-fly consortia don't get very far, because they're poorly defined and
don't have well-considered proceedures for obtaining consensus or publishing the
results of what they do.

RSS seems to suffer from all of this. It is a prime example of an effort that
needs some form of endorsed, well-considered standard; if there are conflicting
efforts, it makes it much more difficult for any of them to succeed, as they
drain energy from each other.

Please consider this carefully, considering the history of RSS. I know that some
will resist taking RSS to an established, public standards body, but IMHO it's
the surest way to success - broad adoption. RSS is only really useful if I can
go to my favorite Web sites and get a summary file from each to aggregate into a
convenient interface.

To me, then, that leaves two reasonable possibilities; the IETF or W3C. I'd lean
towards the W3C; although you have to be a (paying) member, the fees are
reasonable, and many people here are already there (outside experts can also be
invited to participate, without incurring a fee). Additionally, this is W3C's
traditional turf, being made up of mostly application-level interests
(publishing, content, etc.). IETF, while more open, is much more technical,
slower and would probably consider this out of scope.



Quoting Aaron Swartz <aswartz@swartzfam.com>:

> Dave Winer <dave@userland.com> wrote:
> 
> > Well, here's the deal, I'm very confused about what's going on here.
> I
> > suppose many other people are as well.
> 
> I think a large part of the confusion stems from the fact that RSS is
> very
> ambiguous. Is it a format? A process? A standard? A style? Who really
> owns
> it? _Can_ someone own it? What does ownership mean in this context? It
> is
> very confusing.
> 
> > I've been reading the archives of
> > RSS-DEV, after participating and asking directly (a couple of months
> ago)
> > where they're going, and never got an answer.
> 
> If that's true, I'm sorry. I'll try my best to answer here. (Please
> correct
> me if there are inaccuracies in this account. I try my best, but I am
> only
> human, and sometimes make mistakes.) First background:
> 
> Rael Dornfest began using RSS in several of his projects including
> Meerkat,
> and some internal things in the O'Reilly Network. He liked the format,
> and
> saw that it already had a wide adoption, but was disappointed that
> there
> seemed to be no real way to extend the format. He proposed a modular
> system
> using namespaces to extend RSS to this list. Many people expressed
> approval
> of this system, and Dave repeatedly asked for a spec.
> 
> Not much seemed to happen after that until sometime in August, when
> Rael
> announced the publication of the RSS 1.0 proposal, and the creation of
> the
> RSS-DEV list to go with it.
> 
> Now, here's where I see us going:
> 
> First, we're trying to finish up a RSS 1.0 spec which:
>     1) is backwards-compatible with 0.9
>     2) uses RDF (although this point is currently under discussion)
>     3) allows extensibility through namespaces
>     4) is very similar to 0.9 in syntax and model
> 
> Once this is done, we're planning on doing work on RSS modules to allow
> people to add additional information/functionality to RSS. Meanwhile,
> some
> of us will continue to work on the core spec, simplifying it, fixing
> issues
> that have been raised with it, and perhaps modifying it to allow for
> more
> extension/functionality as needed.
> 
> What do we plan to use the format for? A lot, too much, you might say.
> Our
> goal is spelled out in our title: RDF Site Summary. In my eyes, this
> means
> that we're providing a way to distribute semantically-enhanced (like
> XML)
> content for websites. That way, we can get content and information from
> websites, without having to deal with "site scraping". Sort of like
> SOAP,
> but only one-way. The key is getting more value and information out of
> web
> sites. 
> 
> The RSS-DEV group will be a home for site publisher to meet together
> and
> discuss the model and syntax of various modules, so their sites can
> provide
> useful information, but also work together.
> 
> However, this is only my vision, others may see it differently.
> 
> Does this answer your question?
> 
> > RSS already is something. To have RSS go through this artificial
> identity
> > crisis is a waste of a lot of people's hard work and for naught.
> 
> I don't understand this. How does work on a new version of RSS affect
> work
> that has already been done on the old one? We're not taking away 0.9
> and
> 0.91 from you! My.UserLand is just as useful and relevant as it always
> was.
> Perhaps you could explain to me why the publication of RSS 1.0p is a
> waste
> of a lot of people's hard work.
> 
> > In the end, the identity crisis is a no-op because RSS already is
> something.
> 
> I don't quite understand what you mean by no-op, but if you mean that
> we
> won't get anywhere by fighting over who owns RSS, I agree. I'm sick and
> tired of going around in circles and fighting about who gets to decide
> who
> does and says what.
> 
> > It has an
> > identity and a purpose. Look at My.UserLand and My.Netscape. RSS is
> the
> > format behind these aggregators. Look at all the sources of RSS.
> They're
> > publications.
> > 
> > Anyone who wants to make a real contribution, get busy living on the
> other
> > side of the fence, as writers and editors, and understand the issues
> from
> > our point of view. Then the format issues that are discussed with
> such
> > passion fade into the background, they'll be feature-driven, instead
> of
> > trying to bring RDF to world domination, which is something that RSS
> is
> > *not* about, the goal will be to enable the Web to have its own
> syndication
> > system that follows the grain of the Web.
> 
> I agree. This is totally right -- we want a system that is
> feature-driven,
> not a standard no one will use. We want web publishers to work together
> to
> build a system to empower the Web. The goal is what's important -- the
> technology is only a way to get us there.
> 
> > http://davenet.userland.com/2000/09/29/strangeBedfellows
> > 
> > As long as the format is easy to understand, format issues matter not
> one
> > bit. I'm totally unexcited about what's going on here. The only thing
> that
> > keeps me here is the investment. But that's fading now too. I have
> much
> > bigger things I want to do. This argument that keeps going on focuses
> all
> > the attention on some ideas that are way way off-topic.
> 
> Then let's stop arguing and start working. If you have an important
> issue,
> raise it. If you have a puzzling question, ask it. But let's not argue
> about
> issues that don't matter, or don't help get anything resolved.
> 
> > My fallback is to work on ICE. A community that has professionalism
> and
> > wants to go somewhere, and cares what I think. Clearly the RSS folk,
> whoever
> > they are, don't care one whit. I'll put my efforts into ICE and learn
> the
> > lessons from what happened here.
> 
> Dave, I care what you think, but it is only one opinion. I try and
> balance
> with others. Sometimes people will disagree with you, people often
> disagree
> with me -- even after I've explained and argued my point of view more
> than
> was needed. When they do, I either give in and go with their way, or, if
> I
> really care, go off and do things my way. If you want to go work with
> ICE,
> that's fine. If you want to help RSS-DEV, that's also OK. If you want
> to
> work on formats like RSS 0.92, that's great too.
> 
> -- 
> Aaron Swartz | me@aaronsw.com | http://www.aaronsw.com
> 
> 
> 
> 
>