It’s just data

Profile RSS

Dave Winer: We could establish a profile of RSS 2.0 and implement strict compliance with that profile in the major blogging tools.

Excellent!

I can make sure that the RSS validator checks for compliance.


RE: Profile RSS

Sam,
A profile implies a subset. Almost every feed I've seen uses a superset of Dave's spec via some module or the other. Right now I'm currently viewing your post as rich content via xhtml:body, can see a list of your posts sorted chronologically via dc:date, have a link to your comments feed plus count of how many comments you have been posted displayed in my reader via slash:comments, and I'm currently posting to your blog from the comfort of my RSS aggregator via wfw:comment.

I am completely disinterested in Dave Winer trying to specify or control exactly what conformance or compliance means with this plethora of options. As Don likes to say "May a thousand flowers bloom". Secondly if I actually decided to get behind something like this I'd prefer if it was someone whose spec writing chops were up to snuff (like Tim Bray, read http://www.tbray.org/ongoing/When/200x/2003/04/22/RSS-Problems )

Message from Dare Obasanjo at

Dare: if Tim Bray, Mark Nottingham or you were to step up to the plate and write a profile, then I would support it.

The question is: who is willing to do it?

Yes, a profile is a subset.  I don't see that as necessarily disallowing the use of extensions.  I do see it as saying that one should not use such extensions exclusively, i.e., *instead of* "Basic stuff like title, link, description".

Posted by Sam Ruby at

RE: Profile RSS

Sam,
  I'm not interested in being the target of Dave's flames or innuendo about Microsoft people subsetting his spec. At best I might more formally document the RSS items and modules supported by RSS Bandit instead of just leaving it in the code. However things are still in flux with the code which in combination with the fact that I'll be doing about 2 weeks worth of travelling means it'll be about a month before I can tackle something like that.

Message from Dare Obasanjo at

Without wanting to start up the old chestnut again, it might be worth pointing out that Movable Type, TypePad and the new upcoming version of Blogger are all standardising on RSS 1.0, and not 2.0.

But anyway, I'm confused - what's the difference between a profile and either spec's existing usage guidelines?

Posted by Ben Hammersley at

Ben,
I support the same stuff regardless of whether it is RSS 1.0 or 2.0. The difference between a profile and usage guidelines is that a profile is a [commonly used] subset of the spec that is semi-guaranteed to be supported across all implementations.

Thus I'd assume stuff like <textinput> and <enclosure> might not make it into the profile.

Posted by Dare Obasanjo at

Right, you brought it up, Ben: RSS 1.0 is vastly more in need of a profile than RSS 2.0 is: take a look at your spec, with its morass of "proposed" modules, some of which are in common usage despite being unapproved draft additions to core modules, some of which are never used, some of which are outdated or improperly formulated to actually take advantage of RDF. Now imagine that you are writing an aggregator to support it. How would you know that you should support content:encoded but not content:item?

Posted by Phil Ringnalda at

content:encoded seems popular in RSS 2.0 feeds too.  content:item not so much.

Posted by Sam Ruby at

Ben and everyone else, I'm emailing with Ben, Mena, Anil, and Joi and asking them to not "standardize" on RSS 1.0. My argument goes like this -- when I wrote the code for Trackback for UserLand I didn't "standardize" on some other way of doing it, I did it exactly as SixApart did it. I even implemented both methods, before and after they changed their minds about how it should work. I used the RDF discovery bits, even though, as you know, I'm not a big fan of RDF. In other words I respected that they went first, and didn't want to add gratuitous incompatibilities. Ultimately I think this argument will work with them. There's a logic to commercial development that transcends all the BS on the mail lists. If they want UserLand to support TrackBack, or if they ever want help getting another format or protocol started, they should respect the RSS that UserLand emits and understands, because in this case, UserLand went first (going back to 1999, as you know). I've also emailed with the Blogger folk on this. Eventually the argument will work there too, but I'm focusing on MT first.

Posted by Dave Winer at

Dare, this is the first time in memory that someone has accused me of flaming before I flame. ;->

May I help you parse this. I have known MS people going back to the early 80s, and there's always been this denial and disconnect betw the employees and the  management. I don't need to come between you. When I write about MS I am writing about the net-effect. I don't care how the company does it. I know that many people at MS are honest, hard-working, high-integrity people. I also know that the company itself has been convicted of antitrust, and after reading the trial transcripts, I believe that the top people would do anything to win, and I don't support that.

If you think that's a flame, please understand that it's not necessarily about you.

Posted by Dave Winer at

Interesting thread on RSS profile on Sam Ruby's blog....

Excerpt from Scripting News at

re: "I even implemented both methods, before and after they changed their minds about how it should work."  Well, they couldn't very well drop support for the old way, could they?  That would create a discontinuity.

Refusing to create discontinuities is better for current users, but it creates a barrier to entry for new implementors.  In the case of Trackback, that's you.

And now you know how new implementors of RSS feel.

Posted by Mark at

I would have preferred that we just had a specification that resolves the issues, but I'd favor of a RSS 2 profile. I proposed one when more of this forking and flaming was going on last fall: http://www.mplode.com/tima/archives/000126.html

The profile I proposed is not perfect, but it maintains a high level of backward compatibility with RSS 0.91 which is still the most widely used format in use.

I have gone on record as not being much of a fan of RSS 1.0, however it makes a lot of sense for TypePad to standardize on RSS 1.0. TrackBack is a major feature of MT (and I assume TypePad) and it uses RDF constructs. It would make sense to remain consistent by utilizing other formats like FOAF and RSS 1.0 that are RDF-based.

Posted by Timothy Appnel at

Phil, there may be crunchy bits around the RSS 1.0 modules, but the core RSS 1.0 as an RDF vocabulary is well defined : if the root element is <rdf:RDF> it says the document conforms to the RDF/XML Syntax Specification. In turn, this means that it is expressed in the language described in RDF Concepts and Abstract Syntax which has the precise formal semantics defined in RDF Semantics.

Given that this is the case I don't really see how Dare can "support the same stuff regardless of whether it is RSS 1.0 or 2.0.". It's simply not the same stuff.

For a module to be properly compatible with RSS 1.0 it must follow the same model. It isn't possible for MS or anyone else to write an RSS 1.0-compatible extension without leaving their extension open to use by other RDF applications. Lock-in can't happen.

RSS 2.0 modules on the other hand can currently be anything that has a namespace.

One way to prevent lock-in would be to insist on an RSS 2.0 module providing a mapping to the RDF model, keeping the RSS 2.0 syntax.

Posted by Danny Ayers at

Mark: actually, the TrackBack spec says that there will be (was going to be?) a discontinuity: MT was going to not accept GET after January 2003. Mine still seems to, though.

Danny: in theory, I agree completely. In practice, virtually nobody uses an RDF parser on RSS 1.0, so it's just XML with a sort of funny syntax, and Dare supports the same stuff because a <description> is a <description>. In theory, you can do all the RDFing you want in RSS 1.0 modules, stuffing a foaf:Person in your dc:creator. In practice, if MS Aggregator will only display plain text content from dc:creator, so that you can either be innovative or have it actually display in fifty million aggregators, you've got lockin.

Posted by Phil Ringnalda at

Wind'em up and watch'em go.

Dave Winer is calling for a profile to RSS 2.0 that blogging tools could strictly comply to in order to achieve interoperability. An RSS profile was something that I proposed last fall while a lot of flaming was going on over the release of RSS 2.0.... [more]

Trackback from tima thinking outloud.

at

Everyone assumes Microsoft operates in tightly-coordinated unison

In spite of many people referring to the company as the Borg, things are not even remotely so unified or so deliberate.... [more]

Trackback from Tommy Williams

at

There is a relationship between the defining of profiles and discontinuities.

IMHO, no progress will be made until there is some recognition that some valid feeds that exist today won't fit the profile.  Will these feeds remain valid?  Certainly.  But they won't be conformant to the profile.

Posted by Sam Ruby at

Phil, exactly right. And I believe it's better for us to work with each other before each of us ends up trying to work with Microsoft. The chances of it turning out that way are not good. It's never worked before, usually because each developer feels they've already won, or can cut a special deal with the BigCo. Also it wouldn't surprise me this time around if there's some astroturf around. If you don't know what that means, look it up. Watch out for people who try to keep the products from working together.

Posted by Dave Winer at

Tommy Williams's post is right-on. What he says about MS not moving in lock-step is absolutely true. I witnessed it myself, in awe, as they fought among themselves over SOAP vs WDDX in the late 90s. I couldn't believe what I was seeing.

Posted by Dave Winer at

Phil - your description of the current state of affairs isn't unreasonable. But I think the assumption that the newsreader is the ultimate application of syndication is a false one - you probably agree, but I've not heard this vocalised much among RSS 2.0 developers. Just because that's what most people do today, doesn't mean things have to be done that way, or will predominantly be done that way in the future.

People are building RDF applications that go way beyond this, and people that notice this and use RDF (like FOAF in MT) will have a headstart with their RSS apps, no matter what MS do. Those apps may not be exactly like the aggregators we see today, that's all.

Posted by Danny Ayers at

I looked it up : Astroturf is a "grassroots program that involves the instant manufacturing of public support for a point of view in which either uninformed activists are recruited or means of deception are used to recruit them."

Interesting. What makes anyone think that there is a threat from MS, or any need to opt for any single format right at this moment?

Posted by Danny Ayers at

I publish RSS 2.0 because I'm a comment slut, and it has a comments element that SharpReader displays as a prominent Comments link. I'm not going to publish RSS 1.0 because it doesn't have an equivalent (sorry, Ben, but I think you're misinterpreting mod_annotate).

Whether there's any actual value in controlling the RSS spec, or whether Dave just sees the Borg walking among us, dragging their trunks, is another question, and one we can't really answer for sure for a while yet.

Posted by Phil Ringnalda at

Actually Danny it's incredibly trivial to transform RSS 2.0 into RDF. The reverse--not so trivial.

As a code monkey, I don't really care about Microsoft. I'm not interested with competing with anybody in the high-yield newsreader market. All I know is I want to write newsreaders and other applications that use RSS (notification systems) for mobile devices and I know my life would be a hell of a lot easier if there was core RSS 2.0 profile that was implemented by pretty much everybody. (And if that profile enforced a namespace, hehe).

I'm all for a 2.0 profile. Let's get cracking. How soon can we something concrete?

This isn't a new idea, btw. Look at the only successful competitor to Microsoft--Java--and you'll understand the power of widely implemented, strictly enforced specifications.

Posted by Bo at

Bo, to be meaningful we need to at least try to get the Blogger and Movable Type folk to participate.

Posted by Dave Winer at

"What makes anyone think that there is a threat from MS, or any need to opt for any single format right at this moment?" Any time a large company takes notice of a "little" thing folks are gonna think there's a threat.

Posted by James Snell at

> ...to be meaningful we need to at least try to get the Blogger and Movable Type folk to participate.

Following this thread I have the feeling that the threat is not Microsoft, but the real possibility that Blogger and MT folks may choose to support RSS 1.0 instead of 2.0. I don't see this as a threat though.

Posted by Timothy Appnel at

Thinking more about Dare's "I support the same stuff regardless of whether it is RSS 1.0 or 2.0", I wonder if what we need is not a profile of RSS 2.0 or 1.0, so much as a BDG to RSS, that says "you'll want a name for the feed: look for channel/title or rss1:channel/rss1:title", "if you want the richest content for an item, look for xhtml:body, then content:encoded, then description or rss1:description", "item dates will be in pubDate with content that matches this regex or dc:date with content that matches this regex". To me, that's the reality of RSS: it's not this spec or that spec, or dreams of a semantic future, it's just "what elements can I put in my feed and have something actually understand and use them?"

Posted by Phil Ringnalda at

Dave: I would also recommend that consumers be included.

Phil: I would recommend that there also be guidelines for producers.

Here's an example of things that should be considered / recorded for posterity, and possibly tested for by a validator.

Posted by Sam Ruby at

RE: Profile RSS

Dave and Sam,
I don't see how profiling RSS protects anyone from the kind of embrace and extend Dave is worried about. Dave complains that Microsoft could enter the aggregator market and support only a subset of RSS but from what I've seen every popular aggregator only supports a subset of RSS 2.0. So this seems like a red herring especially since I haven't seen any official indication that Microsoft is interested in building RSS aggregators besides journalists putting 2 and 2 together and getting 5 based on work that myself and coworkers do in our free time. Also even if Microsoft or any other large vendor got into the aggregator market given that I am a non-programmer working in my free time and I wrote an app that supports most of RSS 2.0, I don't see why a major vendor with money to pay a professional developer to work full time cannot.

Secondly as I pointed out in my initial comment to the thread, the main value adds I get out of RSS are not the core RSS 2.0 specs but modules added by various third parties like xhtml:body, dc:date, wfw:comment and slash:comment. If I were a vendor building both a backend and a front end aggregator I don't see why I should be begrudged also creating an RSS module. Heck, I'm thinking of proposing the BlogX folks add modules that I can take advantage of in RSS Bandit. Does that make me an evil B0rg conspirator?

Message from Dare Obasanjo at

OK, Sam, I'm game.

http://www.gotdotnet.com/team/dbox/#nn2003-05-11T12:13:16Z

Flame away...

DB

Posted by Don Box at

Please direct comments specific to Don's proposal here.

Posted by Sam Ruby at

A Proposal: RSS for Weblogs

When we discuss a core profile for RSS, we need to define the context in which that profile applies. If we're going to talk about a set of namespaces and elements that define an RSS profile, we need to narrow down the scope of discussion--if we... [more]

Trackback from Six Log

at

RSS for Weblogs

C'est marrant le débat sur RSS 2 et le futur de RSS. On sent bien qu'il y a un consensus, tout le monde fait des propositions qui vont dans le même sens mais personne n'arrive à trancher, tout cela devrait être réglé depuis longtemps mais aucune... [more]

Trackback from Znarf News - le weblog

at

If I may be so bold as to speak for the masses of developers out there who have a great deal of respect for all of the lumnaries that have commented here, let me say this: it's disheartening and shameful to watch some of the innuendos and bruised egos in this thread. I'm not afraid to state that, professionally, I admire all of you as pioneers and master craftsmen of our trade. Please try to keep the flames out of this. At stake here is not just the future of RSS, something that we are all passionate about, but also the credibility of what I'd call open-source collaboration. You see, this is the first time in a long time that I've seen experts spontaneously get together and work on defining a rich solution to a problem. No formal standards body working group dominated by politics. The fact that some of you happen to work for the biggest names in the industry is irrelevant. You are contributing because you love RSS. Politics aside, I'd like to simply add that Dare's last post really resonated with me. You see, it's trivial to support any one feature and tedious but not impossible to support a whole swath of them. The problem for those of us who follow YOUR collective lead is that, barring a whole hell of a lot of free time to track down the 1000 different extension modules and their specs, it's nearly impossible to go to one single, solitary place to get the final word on what SHOULD really be implemented in a truly modern RSS tool. To illustrate my point: xhtml:body, dc:date, wfw:comment and slash:comment are all defined in different places. There is no single place to read/understand all these specs -- not to mention RSS and OPML over at UserLand. Profile or not, what I believe the RSS movement really needs to become truly ubiquitous - as ubiquitous as XML itself - is solidarity and a single place to go to get questions answered. When I have an XML question, I go to www.w3c.org. End of story. Thanks to each of you for your many contributions. Please keep it up.

Posted by Christian Romney at

Pingback from TIG's Corner

at

RSS

RSS and syndication is one of those simple little things-- well, simple for the people building it, at least-- that... [more]

Trackback from FultonChain

at

The battle for RSS

At SXSW I told Mena Trott that RSS 1.0 was dead or dying, because it was too complicated. Turns out I was partially wrong -- it's very much alive, but perhaps only because it's the default in Movable Type. Six Apart has signed up for the semantic...

Excerpt from Manton Reece at

Sam: I agree completely that there should be a Busy Producer's Guide to RSS as well. However, if at least some of the aggregator authors sign off on a BDG, then that makes it easy to write a BPG: "here's the stuff that will actually be used, so it's worth your while to produce it." I'm afraid I've gotten a bit bored with "if you build the data, the apps will come" so I'd rather say "do this, because it will be used" than "do this, because it's elegant and rich and might enable somebody to do something someday, and never mind how ephemeral your RSS feed is."

Posted by Phil Ringnalda at

RSS para Weblogs

Estoy notando que los fines de semana son bastante movidos en algunos weblogs de habla inglesa. Esta vez Dave Winer...... [more]

Trackback from El Refugio

at

Bo - transforming RSS 2.0 into 1.0 as syntax is easy, but doesn't make any sense unless the semantics are well-defined (see  http://purl.org/stuff/ssr). This is a problem especially for modules. From an internal RDF model to RSS 2.0 syntax is trivial, because RSS 2.0 is less rigidly defined.

James - what is there to suggest that MS has taken any particular notice of RSS then? Might it not be that this call for an RSS 2.0 profile is just an attempt to stop MT and Blogger going the RSS 1.0 path?

Christian hits on a good point : "..."it's trivial to support any one feature and tedious but not impossible to support a whole swath of them." The problem is in RSS 2.0 features (as modules) have to be designed individually and effectively from scratch because there is no framework to support them. If a framework was used then default behaviours could be defined, so that if the app didn't entirely understand a particular piece of markup, it could understand what role the data played still do something meaningful.

An RSS 2.0 core profile could be useful, but only if the terms are defined in RSS 2.0 in a form that can be mapped unambiguously to RSS 1.0, i.e. into the RDF model.

Posted by Danny Ayers at

Morning RSS links

Lots of interesting discussion has been going on in the realm of RSS. Dave kicked things off, they bounced over...... [more]

Trackback from life - listed chronologically

at

Maintenance

I've spread myself too thin.  Inspired by Tantek's "What to do with things to do", I have decided to prune.... [more]

Trackback from dive into mark

at

RSS: The Neverending Story

Hi, Evil Twin here! I'm not sure why, but the very mention of RSS tends to bring me out of my quiet corner, where I sit filing my nails while Burningbird does her thing. So, while she's off cleaning house and trying to get the next episode of that... [more]

Trackback from Burningbird

at

RSS und Weblogs

Eine neue schöne Diskussion wurde gerade entfacht: Was machen wir mit RSS 2.0 und Weblogs, wie kann verhindert werden, das...... [more]

Trackback from Noch'n Blogg.

at

Real-time Simple Standardisation

These are busy days in the RSS community. Dave Winer thinks Microsoft is going to fuck all of us and...... [more]

Trackback from Gotzeblogged

at

RSS Roundup

Craziness going on in the RSS world.  As usual, a good bit of it is happening over at Sam Ruby's blog: RSS Profile Don Box's RSS Profile RSS Identity [See also Ben Trott's entry]...

Excerpt from Matt Croydon::postneo at

Generic News Roundup

Roller 0.9.7.2 is out. Rick: "Just for S&G's I hacked Sam's Wx3pa.py script and wrote a 2 paned aggregator." Russ: "BlogPlanet is a new Java-based moblogging app that came out a week or so ago... What a kick ass little MIDlet!"  It looks...

Excerpt from Matt Croydon::postneo at

Can You Feel the Web Shifting?

Don proposes a profile for RSS. Discussion ensues over at Sam's blog, here and here. I just love watching the process evolve in real time - that's the greatest thing about blogs, in my opinion. ...

Excerpt from CraigBlog at

Blogger News Item

 On A Six Apart's Blog today: On Scripting News, Dave writes: We could establish a profile of RSS 2.0 and implement strict compliance with that profile in the major blogging tools. This has been followed by a discussion of the issues on Sam's...

Excerpt from Craig's War & Peace Blog at

open standards -- or just open?

Note: this argument has its fair bit of circular logic built in since I am "just talking" about how we find it hard to move quickly beyond the point of "just talking". Also, it's easy for me to say "oh, people should do this and that." Pontification...

Excerpt from d2r at

Too many cooks in the kitchen.

Here is my take on the whole thing.  I saw RSS a few years ago and considered using it.  However, I noticed a frightening similarity to HTML of early days (not too early, around 1995) when everyone was arguing over what features to support in their websites and browsers (NCSA Mosaic was on its last popular version).  I was younger and full of energy then, so I went through the evolution, adding standard and non-standard tags to my site.  I even survived the painful farewell to the only functional HTML tag around... <blink>BLINK</blink>.  But alas, I continued on, snubbing my nose at HotDog, FrontPage, and Dreamweaver because they introduced "non-standard" mess.  Not until Visual InterDev did I find a tool I could use effectively (well, unless you had script code within a <select/> tag).

Wanting to avoid the mess, I watched RSS carefully until I felt it was worth getting into.  That was last month with RSS 2.0 + Don Box's <xhtml:body/> recommendation.

I don't care if Don works for Microsoft or Mammaw's Kuntry Kabin, I like the idea.  It wasn't until I read his suggestion that I was finally able to write a useful blog management system that I could use.  I tried the public bloggers and nothing really worked well or consistently filled my needs.  So I wrote my own.  I still continue to refer to Dave's http://backend.userland.com/rss "spec" to ensure that I am compatible, but you will never find encoded markup in my title or description fields.  That defeats the purpose of an XML format and goes against the lessons we so painfully learned with HTML.

Consider this, while everyone is arguing over what to do with RSS, content continues to be created.  In my case, it's created to a generic format that is converted to RSS just before it hits the wire.  Why?  Because I know that RSS will come and go (at least its current invocation) and I'll be stuck with gobs of content that I cannot share if I write my system directly glued to any current "spec".

I say, "let the 'heathens' embrace and extend RSS, it will lead to a better web".  We can't fight it, we don't own it, and we'll have to live with it.  I'd rather it (the future of blogging tools [NOT CONTENT!]) be in the hands of a corporation with billions in cash reserves than in the hands of some company consisting of an MBA-turned-programmer-because-I-know-Word-macros development team (that was not meant to target anyone in particular).

Sorry if that's crass, I've been ill and I don't play well with others when I feel bad.

Oh, consider this a vote for Don's RSS:item profile.

Posted by Michael Earls at

An RSS 2.0 Profile

Do not implement this...

Excerpt from Don Box's Spoutlet at

Of interest

Focus du week-end, vite, vite quelques liens pour libérer la mémoire, parce qu'un weblog ce n'est pas que çà :...

Excerpt from .Conforme at

A RSS for Weblogs?

 On A Six Apart's Blog today: On Scripting News, Dave writes: We could establish a profile of RSS 2.0 and implement strict compliance with that profile in the major blogging tools. This has been followed by a discussion of the issues on Sam's...

Excerpt from Craig's War & Peace Blog at

Estándares blogueros "patas arriba"

Ayer comentaba las propuestas que se están realizando para unificar el API de acceso remoto a las bitácoras. Hoy me encuentro con la propuesta de unificación y simplificación del formato de intercambio de titulares. Los chicos de SixApart proponen...

Excerpt from Desarrollo de Blogalia at

A RSS for Weblogs?

 On A Six Apart's Blog today: On Scripting News, Dave writes: We could establish a profile of RSS 2.0 and implement strict compliance with that profile in the major blogging tools. This has been followed by a discussion of the issues on Sam's...

Excerpt from craig cline: My Day Job at

Setting a few things straight

Help me find better choice, and I will gladly choose it.  Until then, I plan to continue to respect the roadmaps.  Both of them. Dave is providing feedback.  Here's my responses:  I have a policy. This weblog will follow this... [more]

Trackback from Sam Ruby

at

A RSS for Weblogs?

... [more]

Trackback from Craig's War

at

The one principal concern that the sector humanity has as far as the writingis haunted is the position from where they can get the characterprinting done Cheap Prints only at cheap55printing.com

Posted by Joy at

Add your comment