The RSS Blog

To be fair, the real test will be to revisit these tests in a few months and see who's incorporated fixes.

I'm not naming names, but there are a few developers in this chart who maintained that Atom was just RSS with a search and a replace on the tag names -- it'll be interesting to see if their support evolves...

   -d.w.

Atom did solve this problem.  Phil can unequivocally point to the Atom specification and say that implementations that do THIS are right, and implementations that do THAT are wrong.  An equivalent discussion about title elements in RSS feeds would lead to a neverending discussion about whether Phil's test cases were correct.  By contrast, there has been no discussion of whether his Atom test cases are correct; everyone agrees that they are.

anonymous, the ability to say, this is right and this is wrong, is less important than to say, this works.

d.w., support for RSS evolved too!

Randy

Anonymous has a good point, though.  Notice that we're not arguing about what the spec says -- that's crystal clear.  Fixes get incorporated that much more quickly if we no longer have to have arguments about what the spec actually says, or doesn't say.

  -d.w.

The RFC number was assigned barely two weeks ago... is that really enough time to declare that it is time to "decide on Atom's success or failure"?

At the moment, the wiki shows seven implementations which not only do this according to spec - but perhaps more importantly, they each do it identically to each other.

Two others have already indicated that they are planning to make fixes.

In the upcoming weeks ahead when the ground is again littered by broken aggregators dealing with Tim making a posts like "Drop the <!DOCTYPE>" in the title, we can not only point to what the spec says but the number of people who actually implement the spec correctly.

I see this as forward motion.

- Sam Ruby

Sam, this is where the community failed. I'm gonna put a large part of that blame on Dave Winer, but he's not the only person at fault. Two years ago, we were in the position Atom is in today. We could have said, <title>'s are encoded like this and anybody not in line is a broken. Fix it! We would have saved 2 years.

Randy

That there is still crime doesn't mean that we should give up on having police.

re: "We would have saved 2 years."

Ah, the wonders of selective memory. Those who were around at the time will no doubt recall that the principle supporters of what became Atom DID try to clarify parts of the RSS spec that we felt were ambiguous. And we tried. And we tried. Much virtual ink was spilled on blogs, wikis, and mailing lists trying to do exactly that.

Eventually we gave up, frustrated at the lack of progress, tired of the he-said-she-said politics of RSS extensions, and we all decided that it would actually take less time and less effort to start over. I won't waste any time trying to place blame, but there it is, and here we are. With a spec that people can actually point to as the starting point for further discussion, instead of the object of discussion itself. That was one of Atom's primary goals, and the discussion at Phil's place is proof that the Atom community succeeded.

-Mark Pilgrim (I was the accidentally anonymous commenter earlier)