Truveo

» see truveo.com as it lived, on the wayback machine

~ the goodbye ~

Truveo's best years line up almost exactly with mine online. By 2006 I had TTNet broadband at home and watching video on a computer had stopped being an event. What I never did, and what I do not remember anyone around me doing, is go to a search engine for video. You went to YouTube and typed in the box that was already there. That habit formed within about eighteen months and it is the entire reason this company is a grave.

Timothy Tuttle and Adam Beguelin started Truveo in the spring of 2004 and shipped a commercial service in September 2005, and the technical problem they picked was genuine. Video in 2005 sat behind JavaScript and player widgets, invisible to a normal crawler, scattered across news sites and network portals with no shared format between any two of them. Truveo's crawler was built to interpret the page the way a browser would rather than the way a spider did, so it could find the clip a conventional index could not see.

That was good enough that AOL bought the company in January 2006, for a figure reported around fifty million dollars. Truveo went on to power video search across AOL's properties, and in August 2007 it was relaunched as a destination of its own, later in seventeen countries with syndication deals attached. At the top the claims were an index beyond six hundred million videos and traffic in the tens of millions a month across the sites it fed.

~ the chronicle ~

~ the thing it got wrong ~

The assumption underneath Truveo was that video would stay spread out the way web pages had. Under that assumption a neutral crawler is obviously valuable, because somebody has to see across all of it, which is the position Google took over text and never gave up.

Video went the other way. One site accumulated most of it and became both the library and the search box, so the crawler that could find video anywhere was solving a problem that had shrunk to almost nothing. AOL owned a very good answer to a question that stopped being asked around 2008.

~ the postmortem ~

Quietly and in stages. AOL stopped relying on it well before saying so, and by the time the announcement came the technology had been on life support for months. On 13 February 2012 AOL confirmed a deal with Blinkx to power video search across its sites and said the Truveo platform was being retired, with the engineering team moved to content tools instead.

The site itself lingered for years afterward with nobody home, which is the usual arrangement for a domain owned by a large company. Eventually it went to a redirect, and then to nothing. There was no shutdown notice for users because by then there were not enough of them for a notice to reach.

~ what remains ~

The crawler idea did not die with the company. Reading a page by rendering it, rather than by parsing the HTML a server hands over, is now how everything from search indexing to scraping works. Truveo was doing it early and doing it for one media type, and being early to a technique is not a business.

@}-->--

              )
             (_)
            .-'-.
            |   |
            |   |
            |   |
            |   |
          __|   |__   .-.
       .-'  |   |  `-:   :
      :     `---'     :-'
       `-._       _.-'
           '""""""

« previous  ·  random  ·  next »