It’s just data

RSS Profile Design Considerations

Tim Appnel: A very productive discussion of an RSS profile has continued throughout the weekend and into this morning. Enthusiastically many have dived in - myself included. I still maintain that an important consideration has not been discussed and needs to. What are our design goals with this profile?

My primary motivation is to discover the best practices that can be codified in the RSS validator.  IMHO, people don't read specs unless they have to.  And the RSS validator has already noticeably improved the overall level of quality of RSS feeds.

Implementation will likely commence next month Don's RSS Profile and the RSS Profile Wiki will be taken as input, as well as any others such efforts that we can find.


I agree people don't read specs unless they have to and that the validator has significantly improved RSS feeds over anything else. (I point people there all the time look when the mt-rssfeed plugin chokes on a feed and they contact me for help.) So I concur with what you are doing Sam. It's the right way to go with this.

The use of "design goals" and "design considerations" may not be the best choice of words. I don't want a specification or anything so formal.

In following the recent thread though, I was curious to what is the motivation behind all of the suggestions and +1/-1's. It seemed somewhat arbitrary at times. I don't think it has to though --I see a lot of common themes throughout the discussion so far. (Those where the points I tried to capture in my post.) Would the best practices become more apparent if we had a few higher-level "ideals" (context) in mind as we discuss this? I think so.

Posted by Timothy Appnel at

I agree with most of what you guys said, but if you're producing an RSS feed for the first time (or adding a new feature for the first time) you're going to go for the spec before you build and test. Specs don't have to be quite as complex as the IETF stuff to be useful. http://wellformedweb.org/story/9 is a perfect example. The fact that Sam will be adding the results of this open discussion on Don's RSS Profile and the RSS Profile Wiki is fantastic, but the developer trying to implement these best practices needs a bit more information and guidance.

Posted by Christian Romney at

Timothy: I actually wouldn't mind if the design goals and considerations were formal.  It would give the validator a place to point to when people ask "why?"...

Christian: my experience is that most people "view source" and then copy what they see... possibly referring to the spec when they have a specific question.

Posted by Sam Ruby at

Could be a bit of chicken v. egg at work, imo. Because the existing RSS specs are either loose and out of date with what gets practiced (or both, and plausibly, more), examining a good feed firsthand ends up providing better direction (or rounding out what the specs themselves might have been able to provide).

This was my experience personally, I can't say specifically at what point I realized that the specs were't enough on their own, but they were the first destination.

Posted by Grant Carpenter at

Could someone send the link to this RSS Validator to Cory? I'm not certain that the quality of RSS feeds has improved. Boingboing, one of if not the most subscribed feed is quite often not validating. Does anybody compute actual stats on the quality of RSS feeds?

Posted by Randy Charles Morin at

http://daringfireball.net/2003/02/short_and_curlies.html
http://www.datanation.com/fallacies/convacc.htm
http://radio.weblogs.com/
http://www.catb.org/~esr/faqs/smart-questions.html

Posted by Mark at

I don't want to post what the first feed on that list not to validate was.

Well-formed XML. We should start a profile for that.

Posted by Grant Carpenter at

http://boingboing.net/2003_02_01_archive.html#90352273
http://www.syndic8.com/feedlist.php?ShowStatus=AwaitingRepair
http://www.xml.com/pub/a/2003/01/22/dive-into-xml.html
http://diveintomark.org/archives/2003/04/01/in_brief_1_april_2003.html

Posted by Mark at

Good to see that the "most subscribed in Radio" list still includes a feed of Dilbert which last changed in November of 2001, with a 404ed image.

Posted by Phil Ringnalda at

Friday Feast #44: Standards, Site Optimization, RSS, and Weblogs

Today's Friday Feast covers more on the world of RSS feeds including innovative uses in education, new interviews and books on web standards and site optimization, and a handy tool for cross-browser, cross-platform testing. Designing with Web... [more]

Trackback from Brainstorms and Raves

at

Add your comment