Robert Scoble, the former Microsoft evangelist and die-hard PodTech videoblogger, has ended his brief departure from the Web. Clearly he thinks he's "adding value" with his bold theory that "Mahalo, TechMeme, and Facebook are going to kick Google's butt in four years." You won't be able to read his theory, of course, since he has, tiresomely, recorded it on video. But you can see the sincerity in his eyes, hear it in his voice, and watch him pull out the whiteboard and three, count 'em: three, colored markers! In truth, he's just revealing what he has always been: a confused evangelist who doesn't understand the underlying technology, doesn't have his facts straight, and can't keep his story consistent. But, boy, is he enthusiastic about it! Why? I think he's lobbying for his next job.

Scoble's "theory" is that Facebook, a social network; Techmeme, a very small tech news aggregator; and Mahalo, the umpteenth attempt to organize the Web with human editors rather than robotic "crawlers," can unseat Google, the king of search. His argument centers around the boogie monster of search engine optimization, or SEO. His arguments, such as they are, have several obvious flaws:

• SEO is not inherently evil. It's also how quality sites get relevant results as top links in search results. Google works hand in hand with SEO practitioners to both combat bad SEO and to promote good SEO.
• Scoble ignores link quality, relevance, breadth of results, breadth of search terms, timeliness (how quickly a link appears as well as how long it remains valuable, a problem also known as "link rot") and other factors affecting search quality just as much if not more so than SEO.
• Mahalo, Techmeme, and Facebook are not immune to SEO. (An SEO expert comments that he's gotten search-optimized listings accepted to Mahalo.)
• Even if Mahalo, Techmeme, and Facebook were immune to SEO, Scoble doesn't link these very disparate systems together -- aside from drawing some arrows on his whiteboard.
• Even if Mahalo, Techmeme, and Facebook were truly immune to SEO and were plausibly linked together, they would just represent one of many alternative search engines attempting to use social features, voting, human-editing, and several other methods to augment search while trying to compete with Google. Thus far, these efforts have failed.

His theory isn't even backed up by facts. He makes several errors:

• He begins by saying, of his own video, "You couldn't have found this from search." Of course, Google and Techmeme, both algorithm-based search engines, listed his page shortly after posting.
• Then he claims that Google can't compete with Mahalo because their algorithms are "stuck in sand... in cement ... If they changed their algorithms, there would be too much of an outcry." Scoble appears laughably unaware that Google is constantly tweaking its algorithms, routinely defeating SEO attempts to game the system, despite the outcry from search-engine marketers.
• Scoble fabricates the number of search results Mahalo can generate -- this, even as he mentions that Calacanis has provided "actual" numbers. Never mind that Calacanis's idealized projection was for the generation of 50,000 results per year by 100 editors, or the updating of 25,000 results using 1,000 volunteers in 15 days. Scoble thinks Mahalo can create 20,000 results a day! Yes, a day.

And he's inconsistent:

• Mahalo is, according to Scoble, immune to SEO, but "Jason [Calacanis] can spam the system all day long."
• Mahalo's results are better, but they are "very incomplete right now."
• Scoble has criticized Techmeme for being a pale imitator of Google News; now he feels it's a Google-beater?
• Scoble actually says a commenter is correct, that Mahalo doesn't scale: "This stuff doesn't scale, and he's right." He then pulls the above Mahalo numbers out of his ass to prove, somehow, that Mahalo does scale.
• He then says to be successful they need to "keep it small." What?
• Despite his argument being predicated on Google's supposed inability to deal with SEO, he refers to algorithms throughout: Facebook's algorithms, Techmeme's algorithms, Mahalo's algorithms, the algorithms to tie these systems together. At one point Scoble shows a glimmer of self-awareness, as he stumbles over this contradiction: "What if the algorithm... what if the algorithm... what if everyone had their own little Mahalo..." What if we all had a pony, Robert?

Scoble then erases his entire argument off the whiteboard and whispers that Yahoo will be the real winner because they bought some social sites and because they have "weird mojo." Okay then!

Techmeme is a niche news aggregator using a small subset of sources to provide a daily flow of news to a small audience. Facebook is a hot social network drawing its own set of "Facebook optimizers." Mahalo is an already-failed experiment being rehashed by Calacanis. Scoble thinks he can conjure up the vision of a Google search killer simply by drawing meaningless lines on a whiteboard between these three sites and putting it on video. But he can't, and he doesn't.

All he's assembled is evidence, on video, that he doesn't understand tech, doesn't understand video, and doesn't understand himself. He's an expert, in other words, on exactly none of his three favorite topics.

At one point, Scoble says: "Where are we going with this? I don't know!" No one else does either. The most sensible explanation? Scoble's conducting a public job interview. Rumor has it that he's already tried to land a job at Facebook. But Scoble working for Jason Calacanis at Mahalo? The two will make a perfect pair: misinformed, contradictory, self-promoting evangelists who will do anything for the next buck.