March 21, 2006

Dave Bonaparte

He he! I get some link-love from Dave Winer. There goes my credibility ;-)

Actually, I am a trifle embarrassed, because I've just finished writing a response to something Danny said, where I've been, generally, complimentary of Dave. So now this is gonna look like reciprocation. It isn't. It was mainly written before I saw that link, and it hasn't been substantially changed since. (Y'all believe me, right?)

Actually, who knows what Dave will think of this? ;-)

For the record, I'm not a friend of Dave's, I only know the virtual Winer through reading DaveNet and Scripting News, and nothing of the real one. In the past I've admitted to being a "fan". And intellectually I'm an admirer. I think he's consistently smart and right about a lot of things to do with the internet and software industry; although there's plenty of stuff I disagree with too.

But most of all, I'd claim to be a "student" of Winer. I watch him carefully. Like everyone else, I have an interpretation of what he thinks and does - which is obviously open to criticism, comments are welcome - and I have an interest in trying to keep the story as accurate as I can, for my own benefit. In many ways, he's a role-model I want to learn from. In some ways, he's a model of what to avoid.

Anyway, here's what Danny said.

If I interpret Phil's this line correctly in the context of the rest of his comment ... he's placing Winer on the side of pragmatic tool builders.

Which is odd given Winer's history of extremely idealistic format building.


Not exactly. What I'd suggest is that Winer understands, maybe better than anyone, how these things work as an "ecosystem". Primarily he's a "platform" builder. It's not that any particular piece of his platform is necessarily all that great, but he knows that he has to produce all the bits and he knows how put them together.

Look carefully. Whatever Winer is promoting, he always has a tool and a format and some kind of hosting or central server and he's "dog-fooding" it and talking it up on his blog and he's finding new, quick-win, applications to extend the platform and he's making new connections into a user community and he's on the offensive, smacking down any potential rivals or threats to his authority.

(That last bit isn't particularly pretty, but who's to say it isn't part of the success of the things he's promoting?)

The point is that other people don't have this perspective. They just make a format or a tool and hope the other bits fall into place. Or they dream of an entire ecosystem that's way beyond their capacity to build. Or they are corporations pouring money into a bloated, late and unwanted attempt to enclose this new platform for their own profit.

Dave is fascinating, and probably unique, as an individual who's smart (and rich) enough to put together an entire platform pretty much by himself. (OK, by definition an ecosystem needs other people; but Dave puts it together : he co-ordinates, leads, bullies, entices, flatters, does deals etc. He makes it happen.)

And his hubris is spectacular. From a standing start last year, he's co-ordinated all his resources to launch an assault on the "permanent, trusted, hierarchical data-repository" market. He has a simple little editor which people are using and already seem to love. He has partners creating various OPML viewers. He has people talking about OPML for reading lists, OPML for attention data (in the "year of attention"), an OPML network of world outlines, OPML as a replacement for traditional CMS, live outlining for intra-corporate communications etc. He knows who his enemies are (DMOZ, Wikipedia, and to a lesser extend Google) and he's already spinning against them. He knows who his friends are (librarians, a significant alliance with WordPress which essentially turns it into a generic hierarchical database server). You can even see Edgio as a cheeky attempt to steal Paul Ford's vision from under Google and the SemWeb's nose. How long before Edgio is accepting "stores" written in OPML?

Compare this with what the SemWeb community has achieved over the last (what?) 7-8 years? This is a project / community invented and led by THE guy who created the web. (How much better publicity can you get than that?) It's full of super-smart, super-enthusiastic people. They've written gallons of code. Libraries for every language I can think of. Dozens of programs that let you enter, query and manipulate RDF data. There've been several attempts to build real popular applications for RDF (eg. FOAF).

And they're still promising that the tools, applications and users are coming "real soon now". At some point you have to suspect it's not lack of effort or tactical ability which is holding them back, but lack of a plausible strategic objective.

[Ed - that's as far as I got before seeing Dave's link. The rest was written later. So you can start assuming some sycophancy creeping into the mix.]

I'll stick my neck out and make a few predictions. (And you have permission to quote them and laugh if they all turn out to be wrong in 5 years ;-)

I think the SemWeb has got very little time left before it finds itself side-lined. Of course, it's got a lot of momentum so it's not going to disappear over-night, it will even continue to grow and gather new users, but as the amount of machine-treatable metadata in "web 2.0" explodes over the next couple of years, the proportion that's part of the SemWeb (ie. in RDF, marked-up with URIs who's "meaning" is defined in OWL ontologies) is going to be infinitesimal.

That's why smart pragmatists like Danny are, quite sensibly, trying to make bridges between their technology and the users who actually have the data. But these attempts are being sabotaged (eg. by Shelley who is missing one of the golden rules of establishing a platform), and it looks like Danny's too polite to fight his corner. ("Fair enough" indeed!)

Instead, this is Winer's game (again). Publicly hosted OPML hierarchies are going to roll right over DMOZ this year, by which I mean they'll exceed DMOZ in popularity. They'll give other silos of expertise (About.com, Squidoo) a serious fright. OTOH, they won't make a dent on Wikipedia. (Aside : I don't think Winer really "gets" wiki. If you want to take him on, that's the front to do it.)

Within two years, most major blogging clients will be exchanging reading-lists, blogrolls and attention data with each other and online aggregators, memetrackers etc. in OPML; scutters will be crawling raw OPML for "who-likes-who" data; and FOAF will be gone. And FOAF was the greatest hope for a genuine popular RDF application.

It's not that I think everything is all going Winer's way. My guess is that the most popular OPML Editor by then probably won't be Dave's. He doesn't seem interested in cultivating a community of developers for the editor itself, which strikes me as a strategic error. Someone will take his code (or write their own) and pretty-it-up for the mass-market. [Note that Dave hasn't made a connection with the whole Getting Things Done cult. (Unless you count moving to the Mac.) There's no great OPML-love on 43 Folders (I can't even find the Editor in its wiki). And these guys are obsessive about outliners. But they're much more driven by aesthetics.]

On second thoughts, maybe Dave doesn't really care about the Editor. In practice, Dave's business model seems to be "create platforms, sell ping-servers". (Which casts some light on the whole Winer / Cadenhead fallout, doesn't it?)

But that's another story. The point I want to make here is that Winer is Napoleon, and the poor SemWeb people don't even know they're at war.

Or rather, they know something is up. They know they're hurting, and that people seem to keep getting inexplicably excited by stuff they consider obviously inferior to their grand project. They complain about the crappiness of Winer's format, or the obnoxiousness of his personality, or that it's "only" because of the free publicity he gets on his blog, or his infantile fan-club, or that he's rich. What they don't see is the master strategist playing the game.

Let's be clear. Winer isn't mainly attacking the SemWeb - it's incidental to him. He's likely to be far more perturbed by more practical rival formats like Atom, defending himself against losing the server business (like he lost the podcast-server business), and fighting off the big companies.

The damage he's doing to the SemWeb is a side-effect of him being a smart SynWeb operator in the inevitable conflict between SemWeb and SynWeb. But there are plenty of others willing to play a SynWeb strategy if Winer were to retire.
Post a Comment