Dave Winer claims on Scripting News today that Google is playing dirty with RSS in favor of Atom:
... Google is going to start reading feeds, but if I understand correctly, they're going to ignore the billions of RSS feeds out there, and ask everyone to convert to Atom to get more currency in search. You can imagine that I don't like this. I wouldn't like it even if I didn't play a big role in getting those billions of feeds out there. I wouldn't like because I have thousands of RSS feeds on my servers, and believe me -- they are not changing to Atom anytime in the next few decades. I don't think I'm alone in that.
Now a little preaching. Big companies always feel they can push the rest of us around, but I gotta say -- I've never seen it work. Usually the lesson they learn is that they would be better off if they would just Go With The Flow, and let the users guide them. Nothing wrong with reading Atom feeds, but to ignore RSS, well guys that's just plain dumb.
Give up the fight Google. You don't have to acknowlege me, but RSS -- that's a force of nature. That's why I did rssCloud -- for you -- to give you the impetus to do what you should have done naturally, support the formats that the users have chosen. It's not too late to get our relationship back on track. I'm not your enemy, I'm just one guy in an apartment in the West Village writing on my blog.
He understands incorrectly.
If he's talking about the news that Google may use PubSubHubbub (PuSH) to allow web publishers to submit new content to the search engine, there's no reason that this development would exclude "billions of RSS feeds." The PuSH protocol does not make feed publishers or software developers choose Atom instead of RSS. The protocol works equally well with feeds in both formats. If a hub is monitoring an RSS feed, it sends RSS data out to interested clients. If it monitors an Atom feed, it sends Atom.
There was some early confusion because the PuSH specification was not clear on this point. To address the issue, I made some spec suggestions in September and Brett Slatkin incorporated them into the current draft of the specification. The spec leaves no doubt that PuSH is designed for both formats.
This blog is proof of that. I upgraded my blog a few months ago to send out updates using the protocol. Although my feed is in RSS format, PuSH has no trouble transmitting updates. People who are reading my blog in Google Reader or Google Buzz -- two of the first popular clients to support PuSH -- will get this blog entry a few seconds after I publish it.
PuSH is the best way to deliver real-time updates to RSS or Atom feeds. Now that WordPress supports the format on all 7.5 million blogs on WordPress.Com, all of the leading blog platforms have adopted the format.
The alternative, RSSCloud, still lacks a specification seven months after Winer revived it. There's only some rough implementation notes and no process in place to enable interested parties to decide what features the protocol will contain or how the spec will be written.
Google, if you're reading this, I'm concerned about our relationship. Why don't you call me any more? Things can be good again, baby. I'm sorry I got so angry before. I love you so much sometimes it just makes me crazy.
You're deliberately misunderstanding (and misrepresenting) my post. I've got feeds that don't have any of this stuff in them. Google should read them. That's what my post says. They should have been reading them for the last ten years. Stop being so competitive Rogers and just read stuff at face value.
I linked to your post, so people can read it and figure out what you meant. I described it as clearly as I understood it, and to me it's not accurate. I've been corresponding with the PubSubHubbub creators for a while. They're not trying to screw over RSS.
To me, it seemed like Winer was talking about indexing RSS for inclusion into search. What does this have to do with PuSH?
If that's what he's talking about, Google has indexed RSS for search since 2005, as I noted in my comment.
Cadenhead you are being a jerk putting words in the mouth of Dave Winer --again-- as those of us who used to read the RSS mailing lists can attest; so herein I speak for myself in this regard.
Secondly, if you were as professional as you imply Cadenhead --and-- if you were an all-around decent kind of fair play fella (which you are having a problem with) you would use your ill-deserved name recognition to expose the fact that the feed validator at feedvalidator.org has been coded by the sleazy-weasel(s) Sam Ruby et al. to undermine RSS.
The validator has been corrupted to favor the use of Atom by compelling the use of an atom element in the channel element of an RSS file in a manner which actually intentionally obstructs the use of RSS channel element attributes.
Failing to abide by the polluted and corrupt bully-pulpit-bullsh!t imposed by the validator by using specified RSS elements and attributes consistent with the specification now flags the RSS file using color and language the same color and language used when the file is actually invalid (mis)leading those who do not read very, very carefully to be misled into misbelieving that the developer of the file has failed to generate quality results.
As I pointed this out in your little boys club email list I was banned because I called you under-handed sleazy worms out for what you are: vandals.
The rssCloud element was specified first years ago (almost day one) and Winer does not have the resources to build a feed generator the way Google has these days.
Furthermore, Winer was way ahead of his time and had no idea when the rest of the development community would learn to adopt web services which are required to use the cloud in near-real-time.
It is only your dishonesty, the lack of information, the lack of coding skills of the typical web developer and the bully pulpit that causes the rssCloud element to now appear as if must be relegated to the back seat.
I'd like to see rssCloud have a fair chance, the developers of WordPress agree and support rssCloud too so gfy Roger.