My political FAQ for RSS 2.0

Is it a matter of just saying that the RSS name is not available to anyone who wants to hack something, effective or not? Has any legal trademark action occurred, or been contemplated?

If Userland does not control the name, or just does not enforce it's right to it, it risks becoming generic, and I can call this post RSS 9.11 for any good it does.

Aspirin was once a trademark. Is RSS? Oh, and an aspirin a day, they say, is good for what ails ya'.

The IETF license would be more accurately described as a Creative Commons Attribution-NoDerivs license, with the additional right to include just portions of the work (unaltered) in derivative works. IANAL.

The software developer doesn't want to reinvent the wheel, and wants to play fair with as many of the existing standards as possible. Where is the best place to get acquainted with the current standards? Where are the best specifications posted? How many standards are there now for accomplishing the above with XML? Is it possible to abstract from the politics and write something that will be useful two years from now or is it unavoidable in this case?

Human Observation: It isn't entirely a bad thing that an explanation in terms of People is necessary here and not just Specifications or Programs, as long as the People involved are more helpful than the inanimates.

Good to see IETF mentioned. IETF standards tend to get the infighting out of the way early, so that when a working group approves a standard, it's the *IETF*'s standard, so that people don't have to consider any one developer to be important to the *use* of the spec (beyond "oh, he wrote that RFC, cool"). Of course, the tradeoff is that they take longer or die entirely... Really, any spec that needs the author around to expand upon it has failed, significantly, to scale.

- Userland RSS descends from NetscapeRSS0.91 - NetscapeRSS0.90 descends from MCFinXML and AppleMCF and CDF and uses RDF. - NetscapeRSS0.91 descends from RSS0.90 and scriptingNews - scriptingNews descends from AppleMCF and CDF. - CDF descends partly from AppleMCF. - RDF descends partly from AppleMCF and MCFinXML (and various other proposals). - RSS 1.0 descends from NetscapeRSS0.90 and uses RDF.

FAQ type question: What would be a good place to look for correct example code?

At what point would you recommend developers look to innovation and new formats rather than RSS 2.0 ?

I do have what seems to be a FAQ regarding the recent RSS 2.0 discussion re Blogger and Movable Type. Dave claimed that MT created "funky" RSS. I'm guessing that's what he's referring to here when he says they refuse to do RSS Userland's way. But I've never found anywhere where Dave or anyone points out what is "funky" about MT's RSS. So my question is, Dave, could you explain what you meant by "funky"? Or could you explain what you mean now when you say Blogger and MT don't do RSS in Userland's way? Is this a technical issue? Some misplaced tags? Are they not following your published spec, or should they conform instead to RSS feeds that Userland software publishes if the spec and Userland's feeds are not in sync?

A feed is "funky" if it uses an extension to provide information that can be expressed by baseline elements, without providing the corresponding baseline element.

Or in other words, a feed is "funky" if it deliberately hides information it has access to from applications that only implement the baseline. Providers should strive to maximize interoperability, also with future applications, instead of just looking at what "most current clients" do.

And IIRC/BTW/FWIW, Ben's already said that he's aware of the problem, and that they're going to fix the "MT funkiness" in an upcoming release. All in order to give clients a choice, and maximize interoperability.

Also, to David, there are plenty of examples of cool RSS feeds. There are even some linked into the spec. I agree that having examples is very important, that's why I provided them. Confused.

Then what exactly was funky about your own feed, as noted here:

http://scriptingnews.userland.com/2003/06/17#When:2:40:38AM

Hmm, I just noticed that you did actually go ahead and remove those 'three bits of funk' from your feed. For the peanut gallery, they were a blogChannel namespace declaration, plus blogChannel:blogRoll and blogChannel:mySubscriptions elements, as defined here:

http://backend.userland.com/blogChannelModule

Unless I'm missing something, these 'three bits of funk' don't provide any information that can be expressed by core elements.

Interoperability between what and what? Using Dublin Core is surely a good thing when interoperating between diverse XML formats, since it is using a common and understood vocabulary. Is there actually a good technical reason why not using a common available vocabulary is "pro-interop"?

You've asked people before to read what you've written, and not twist what you say. Well, I did that, and your 'three bits of funk' statement definitely implied that you considered all namespaced elements 'funky'. You aknowledged that yourself, so I don't see why you're blaming me for drawing the obvious conclusion.

I explained how I came to this conclusion and asked (with respect) whether I had got anything wrong. I did all this several times, in several places. Each time you dropped out of the conversation without answering. I hope you can see why this caused me to be more persistent in wanting an answer, and not less, because I felt that what you were implying was important, and that your reluctance (refusal? unwillingness?) to answer my question was important as well.

I did not flame you, nor did I accuse you of anything. In fact, I took other people to task for flaming at you on several occasions. I only sought to understand, and explained what my current understanding was, my thoughts and feelings, and how I got there.

Now you're twisting *my* words, saying I asked you to check with a lawywer, threatening to sue you. Accusing me of having a hidden agenda and trying to make you seem an immoral or bad person. Not fair, Dave, not fair at all.

Feel free to reread all the email I've ever written you, the postings I've made to mailing lists and comments I've made on weblogs (yours and elsewhere). I don't think you'll find I deserve to be attacked this way by you, simply because I asked a question you didn't want to answer.

Between RSS 2.0 producers and RSS 2.0 consumers, of course.

(Besides, Dublin Core is a metadata model, not an XML file format. Conceptually, there's no difference between RSS 2.0's pubDate and a Dublin Core date. Both properties provide "A date of an event in the lifecycle of the resource". The syntax for the element value is different, but it's not like your computer really cares much about that...)

- language (vs XML's xml:lang) - lastBuildDate (vs HTTP's Last-Modified) - ttl (vs HTTP's Expire) - textInput (not necesarily funky, but what HTTP method is supposed to be used?) - skipHours, skipDays (vs HTTP's Retry-After) - length and type attributes of enclosure (eg., how does that work with HTTP's Content Negotiating?) - using character entity references without declaring them (vs XML's requirement to do so)

Is this "funk"?

Only if you think that an RSS 2.0 document always has to be transported over HTTP, and that all RSS 2.0 providers have access to other ways to provide metadata for their HTTP server.

If not, "language" is the only one that's somewhat funky. File under "historical warts".

That limited? No wonder its a mess then

I thought interoperating was about working together to share information where it is needed, not locking it within one closed system (as in RSS producer to RSS consumer only). The closed system you prefer gives me no scope without being labelled by you as funky and scorned at every opportunity.

If you were iterating and approximating your definition of funky, *it was not clear to any outside observer*. All anyone could tell was that you were being cryptic. Another way of accomplishing the same thing would have been to point to your feed and *ask* whether other people thought it was funky. You would have gotten feedback of an entirely different sort.

Part of taking risks is admitting when you were wrong. Doing so only reluctantly and in a nasty way ("so sue me") doesn't speak well of you.

Yes, I am a programmer. And a writer. And a designer. You obviously decided to pigeonhole me into a convenient category, to make your argument. Well, people don't fit into neat little categories Dave, we're messy and sloppy that way. I understand that about *you*, how about giving me the same f**kin courtesy?

As far as I am concerned, I can take anything you choose to dish out, but in the past I've apologized to you (and sincerely) for smaller things that you've called me on. Now you're making me a scapegoat for what other people have done to you. Not cool, Dave. I am not an object lesson.

Here's a clue: Hard questions are important questions, not an attack. Think about that for a while.

"Interoperability between what and what?: Between RSS 2.0 producers and RSS 2.0 consumers, of course."

For what it's worth (feel free to delete), I think the 'funky' ploy backfired. Big time.

HTTP is just an example, but an important one given that RSS over HTTP is far more popular than anything else. The point I was trying to make is that the issues lastBuildDate, ttl, skipHours, skipDays are concerned with (such as caching) should be dealt with at the level of the underlying *transfer protocol*.

That's what they have to do with each other.

Indeed, especially when the feed validates. Considering in the real XML world, usage of Dublin Core-based elements as a standard vocab for publishing related metadata, it is going too far to realistically expect an XML stream containing RSS to use RSS limited elements instead of an already existing Dublin Core equivalents - and to do so without duplicating the content. Its an odd philosophy to pursue - almost overbearing in its implicitness that RSS data shalt not be combined with any other data. Its odd, and limiting. So either I should just ignore this "funky" daftness and FUD, or use anything else other than RSS.

Its all probably bareable when RSS is all you have. But I don't like locked-in environments with no hope of sharing information and content.

Your answer to Dave reveals, however, that you were originally trying to make a point, not actually ask a question, and I think that qualifies as giving him a hard time (I suppose it's debatable, so I'll concede that this is my opinion). Peace.

This converstation is getting to be too 'meta' for me.

I don't see how asking if my assumption was correct that this is the sort of thing Dave describes as "funk" would be giving him a hard time.

You said "The point I was trying to make ..."

Asking a question to make a point is (IMO) giving someone a hard time.