Tag: search

Humans and rats battle it out for second place in search

Posted on by Daniel J. McKeown

I’ve already tried to pour cold water all over Wikia, and I remain a skeptic on Spock. To the point: the machine powered search algorithms used by companies like Google and Technorati actually aggregate human decisions in a much more effective market-style system than small teams of web professionals judging what content is relevant. The search engines said to be community powered are interesting, but the idea that they will be able to give people what they want more effectively is elitism run amok, an attempt to build a new online “mainstream media” establishment when no such thing is needed, and an admirable but forlorn attempt to substitute labor for capital and taste for popularity.

Oh, I forgot, this is some well-intentioned attempt to save the world from SEO spam. What nonsense.

I had held out on writing a sour note that ripped on Mahalo because I like the idea that people believe that someone might seriously challenge Google in search in the next decade. It’s a silly belief, and certainly not an analysis of any sort, but it is amusing and maybe even a bit inspiring.

What a time to start an Internet business. It will take a very long time, warns Mahalo chief Jason Calcanis, before the search engine will really work well. Or make much money. How convenient.

As a practical matter, most searches end in Google results and then Google ads right now.  The site has an unserious florid look and even the established results pages are pretty weak when you click on the directory-style front page.  It is a new site with apparently a good amount of investment–here’s post from May about its launch–and for now people seem inclined to take it seriously.  But after some pretty positive press about his “human-powered” search engine from Read/Write Web, Jason Calcanis gave a conference speech this week that went sour with the attendees. Dave Winer explains on his blog:

‘Yesterday, and in all his previous marketing, he rails against advertising and spam, which ironically, was exactly what he was doing to the environment at this mostly non-commercial conference. What we said (and I wasn’t the only one speaking back to him, I wasn’t even the first) was a response to this. It didn’t come out of thin air. If he had given a similar speech to venture capitalists, if he offered them no way to win, they would have had the same response, but it probably wouldn’t have been as patient or polite. Now, clearly he doesn’t have the same respect for us that he has for VCs. But it seems that to some extent the success of his company depends on winning over the people here at Gnomedex. If it didn’t, he should have stayed home, because his pitch, as delivered, doesn’t work here, because he didn’t offer us anything we want. We get a better deal from Google, believe it or not.

Some of his argument against Google rings true, very few people love them as we did in their early days, but their proposition to web writers and podcasters is basically fair, it’s a win-win. We get flow from them, they get ad revenue. They also offer us a way to put ads on our sites, so we can profit financially from the relationship. Nothing in Jason’s pitch offers us anything like that. No flow, no money. And technically, it’s not a platform, so we can’t build on it.

We’re people, and we’re smart, Jason, just like you, just like your investors. If you come making a pitch, there should be something for us, or it’s not going to be well received.

So there’s a big bug in the concept behind his company and he tries to blow by it with an attack aimed at one person. That might convince really stupid people, but smart folk can see right through it.

Bottom-line, he needs to figure out a way to build the company so that many others can profit from it. Otherwise I don’t think it has a prayer against Google, which we like less and less as a company, but who basically offers an equitable proposition to the users of the Internet, who the Gnomedex crowd represent in a loose kind of way.

His pitch here failed. He can’t blame me for that. A good CEO goes back to the drawing board and figures out what works.’

Yes, a very, very good CEO finds a way to make the “human-powered” search engine idea work. Like, work in a small niche or something.  But a CEO with a bad idea will keep trying to sell his idea as long as its plentiful funding provides a constant prima facie reminder of what a success it is.

I don’t think I write hyperbolically when I say that you’re about as likely to beat Google’s machine-powered search system now with Mahalo or another “human-powered” system as with a “rat-powered” search where rodents sniff out the best-smelling morsels.

Will Spock.com, a social listings service, be worth using?

Posted on by Daniel J. McKeown

[Title updated 8/3/2007]

I read about Spock.com on O’Reilly Radar a few months ago, and couldn’t help but be intrigued. Mr. O’Reilly wrote:

‘Michael Arrington wrote the other day about spock, the new people search engine, but I have to say that I don’t think he did it justice. Spock is really cool, and performs a unique function that is well outside the range of capabilities of current search engines. What’s more, it’s got a fabulous interface for harvesting user contribution to improve its results.

You can search for a specific person — but you can do that on Google. More importantly, you can search for a class of person, say politicians, or people associated with a topic — say Ruby on Rails. The spock robot automatically creates tags for any person it finds (and it gathers information on people from Wikipedia, social networking sites like LinkedIn and Facebook), but it also lets users add tags of their own, and vote existing tags up or down to strengthen the associations between people and topics. Users can also identify relationships between people (friend, co-worker, etc.), upload pictures, and provide other types of information. This is definitely a site that will get better as more people use it — one of my key tests for Web 2.0. It also illustrates the heart of a new development paradigm: using programs to populate a database, and people to improve it.’

The service is a closed beta, but I finally found a way to check it out–I left my request for it at InviteShare and moments later, I was invited. [Check that service out if you are wondering what new web2.0 stuff is coming down the pike or want to join Spock or Pownce. In fact, InviteShare itself could actually develop into a pretty cool social networking site–at least until everyone in line gets their invite to the red-hot BitTorrent closed beta sites like SuperTorrents, Demonoid.com and Torrentleech.org.]

So I got my invite and went and looked at the Spock.com site, searched a few tags and a few people, and one thing that struck me is how dependent Spock seems on Myspace profiles–useful enough for lots of people but hardly a reliable main source. Also, some sites are based on LinkedIn profiles and entries about famous people rely heavily on Wikipedia.

I think at the end of the day, if nearly everyone joins and tends to their profile (although I don’t know how much control a user has over their own profile because when I was clicking the link to finish the “claim this profile” process it repeatedly gave an error message) it will basically become a social networking aggregator, a hot field at the moment but different than what the site seems intended for.

Why will it not work as a people listing? Basically, because I don’t know how it won’t become a major spam magnet once it opens up to public users. Besides that, it has all the potential to have all the problems that Wikipedia has with user credibility, self-promotion, grudges and personal attacks.

I went in and added my web site, so now people who look at my entry can find that page alongside my less-recently-updated Myspace page. That’s pretty cool. And since you can tag people, I guess I could have tagged myself “MBA” and “Midwesterner” or stuff like that–maybe I will. But then, what’s to stop someone from going to my profile and tagging me “capitalist pig” and “hillbilly” as well? Nothing, it seems. I guess maybe I can vote those down or something–but for now that’s not enough to convince me it will work, or that in the current social networking blizzard I’d really want to spend much time using this particular site.

It’s all kind of unclear how the site will look with heavy traffic, so my observations are quite preliminary. But does Spock match the hype so far? I don’t think so.

Ok another insignificant search engine changes hands

Posted on by Daniel J. McKeown

Here’s a news.com post about the acquisition of the distributed online search engine Grub by Wikia.com:

‘Grub was acquired from LookSmart under the open-source project Wikia. The platform, now available for downloading and testing, is built on users donating their personal computer power. It’s meant to operate through open protocol and community collaborative added functions combined with the wiki.

Last year, Wales claimed that Internet search as we know it is broken. Grub is one of his attempts to gather open-source technologies to organize free content on the Web.’

Well actually I think that Google and Technorati have gotten better rather than worse–and even Live search can be usable.  The only thing really broken about search is that people still use Wikipedia and Yahoo and Ask.

I’d say you’re better off letting your computer help the University of California find aliens than trying to help Jimmy Wales beat Google.  Anyway the first is more likely to happen.