The Block Album

Every few eons we get another RSS is dead swarm of stories, usually involving Dave Winer versus the rest of the universe. Sub-themes include dead calling is stupid, I found this post on RSS, and get off your porch grandpa. Typically Dave uses the event to launch yet another version of Radio 8 tricked out to convince us that his lack of business model business model beats traditional data silo roach motel closed software. It may sound like I am pursuing a personal vendetta.

I’ll admit to some mixed emotions about Winer and his attempts to regain control of what he calls RSS but is in fact a new socially adept layer dominated by Twitter. Starting with my post several years ago on TechCrunch, I’ve stated the obvious, that RSS has become at best a formative technology that has led to the development of realtime social streams of citations. When Winer led the development of XML-RPC, SOAP, RSS, blogging, and podcasting, he was often the single reason these technologies broke out of their original niche and expanded across mainstream media, tech platform players, and a broad coalition of individuals who could now make a difference. For Winer, success proved a difficult pill to swallow.

In calling the moment as I saw it, I in no small part was emulating what I view as Winer’s incredible sense of strategy. When Twitter took off, I teamed with Winer to organize BearHugCamp, so named as the result of a post where I described Winer’s strategy of bearhugging a similar technology to create more momentum and an adhoc standard. But bearhugging Winer himself only goes so far; he likes to be the hugger, less so the hugged.

This presents a problem when Winer’s motivation seems to be to absorb all post-RSS innovations under the blanket RSS model. When some Google engineers produced a service called PubSubHubbub, Winer attacked it as a bigco project even though it was created expressly as an open service that could and was adopted both by Google and non-Google companies and individuals. Then Winer produced RSS Cloud, which he pitched as being unencumbered by bigco politics. Certainly having a choice is a good thing, but I found the motivation for the effort to have more to do with Winer’s view of his legacy than any notable difference in technology or what is called openness.

But what I initially called out as a turning point had less to do with what RSS did and more to do with what it didn’t. Twitter’s social metaphor, the Follow, created a data model for extending and amplifying attention, something I had been exploring for some time beginning with attention.xml and continuing with the AttentionTrust and its attention recorder plug-in. With the ability to track and model the intersecting clouds of people, who they follow and are followed by, we could now begin to cultivate the full value of the realtime stream, its content, and its gestural metadata. RSS and its popular renderings such as Google Reader were succumbing to overload.

BearHugCamp succeeded in describing the possibilities for unifying around a defacto set of APIs that would speak across Twitter and its much smaller competitors. Perhaps too successful, because Twitter developed so much velocity that other sites such as Identica (now statusnet) using a clone of the API were vulnerable to Twitter gating realtime access to its full firehose. FriendFeed and even Google Buzz were also constrained from symmetrical access to and from the Twitter stream. Facebook’s acquisition of FriendFeed has still not precipitated realtime pushing of its updates into the Twitter stream. Bearhugging has passed its moment of opportunity, at least for now.

In the aftermath, Winer has shifted to attacking Twitter as closed while promoting what he sees as an open federated approach that uses his RSS Cloud work at the center. The biggest problem with trendjacking the open meme is that the cloud makes size irrelevant. You can operate in the realtime Twitterverse with a few well-chosen @mentions and be virtually indistinguishable from any Fortune 500 megaprise. In that climate, attacking big companies for proprietary manipulation is as good as attacking oneself. From there it’s a long slow slog trying to define what open is in such a way as to advance your particular proprietary interests.

In the RSS argument, Twitter is demonized for having control of the stream. Certainly the company has the ability to pull metadata out of the message body and establish a proprietary wrapper, as they’ve done with retweets. But to service their third-party client platform, they’ve made their proprietary wrapper available via their API. When you click the retweet button in the Twitter clients, the API produces this string — you: @retweeted: and the message body. In other words, your username followed by a colon (minus the @ symbol,) the @mention of the retweeted user followed by a colon, and the body with text, URL citations, and any other @mentions.

In effect, Twitter has established an open standard around the @mention syntax. When a Twitter retweet is consumed by a down-level client, the object is translated into you: RT @retweeted: message body. This format and its analog via the API are the open standard for retweets, since every other kind of manual retweet and Twitter retweets all conform to the same syntax and output. As long as Twitter retains this format, the standard persists. And it encourages other players to support the standard, thereby increasing the “lock in” for the standard. As with the way many defacto standards emerge, we see Twitter syntax being used in Facebook messages even though the @mention syntax is not officially recognized. Quora uses the @mention to push you back into its silo, rather than out to Twitter.

If the @Mention Cloud is open, then what does that say about RSS? Does RSS support the @mention extension for retweets? Probably, but if not, why not. If it’s supporting the open standard, then it’s supporting the broader micromessaging standard. Which makes sense for RSS, given micromessaging has garnering the lion’s share of attention. Which was the whole point of my original post about the death of RSS. That’s what happens when foundational technologies become oil. They live on as fuel for the new black.

This of course is hard to swallow when you’ve spent years struggling to get sufficient momentum to send RSS into orbit. Many of us have been doing that for years, not just Winer, but also a broad coalition of journalists, technologists, publishers, entrepreneurs, and competitors. We knew a game changer when we saw it and made it an imperative to support it until it could accelerate under its own power. And like all proud parents, at some point we have to kiss our children on the forehead and see them go out on their own.

What RSS did was create and nurture an information landscape based on the authority of the author. It allowed those authors to gain such authority both within and outside the media infrastructure of that time. But it also gave rise to a separate wave of authority, that of the reader, the analyst, the commentator, and so on. The blog post maintained its central role even as it absorbed every other document and media type, and it was joined with a seat at the table by a metadata stream of attention and gestures. Suddenly the root object was the @mention, whether a Tweet by the author, a retweet by a reader, an @mention of other such authorities, the implicit clouds of such signals.

Winer indirectly recognizes the power of the @mention. In a recent post, he acknowledges sitting on the @mention stream as his favorite view into the Twitter cloud. Of course, he mentions this (cough) in the context of explaining why he blocks anyone who spams that view by using @davewiner to force his attention. Blocking not only removes the blocked offender from his follow stream, but also from the @mention view.

Winer mentions one side effect that blocking produces, that the blocked person can no longer follow or view the blocker’s tweet stream. But interestingly, blocking an identity also blocks @mention updates of that user by other people that also include @davewiner. Retweeting a post with one name which then is retweeted by another produces a conversational thread opaque to the blocker but visible to the rest of the network. That may be a desirable effect, but until Track or some more flexible block mechanism is offered, Dave’s blockmention model is unlikely to gain traction.

The @Mention Cloud is the new black and nothing will stop it, just as nothing will slow the steady march toward the cloud in general. The @mention cloud scales beyond what I know to what we know, elastic in context and realtime relevance. RSS lacks the core components of the social web, and survives as a really simple reminder of how to get where we are going. Certainly it could and should be reworked to mine the @mention cloud precepts. Call it @RSS Cloud.