Zuula

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

~ the eulogy ~

Boris Simkovich was an economist with a doctorate and about a decade of management consulting behind him when he sketched the idea in spring 2005. He and Tim Hunt fixed the feature list that summer, spent roughly eighteen months building it, and put Zuula out at the end of 2006 as a New Jersey limited liability company with no funding story attached to it. Search Engine Land reviewed it in the first week of January 2007 and the Star-Ledger wrote it up the next month as a local business.

Metasearch in 2006 was a well-worn shape. Dogpile and MetaCrawler and the rest queried several engines and merged the answers into one list using some scoring nobody outside the company could inspect. That merge was the product and it was also the problem: you got a list you could not attribute, from a blend you could not tune.

I was on TTNet broadband by then and typing into Google like everybody else, so Zuula was never mine. What holds up is the decision behind it. Somebody looked at a settled category in 2005 and judged the settlement wrong in a way that could be fixed with tabs.

~ the dossier ~

~ the tab ~

You typed once. Zuula showed you the results from whichever engine sat in your first tab, and the other engines waited behind their own tabs with the same query already run. Nothing was merged, nothing was rescored, and the source of every result was the label above it. You could drag the tabs into the order you liked, and the order was kept separately for each of the six search types, so your web ranking of engines did not have to be your blog ranking of engines.

The tab list was also where the small engines got a hearing. The majors were in there, and so were Gigablast and Mojeek, independent crawlers running their own indexes that were never going to survive a blended ranking against Google. Zuula gave them a tab, which is the only interface in that era that treated a small index as a thing you might deliberately choose rather than a thing to be averaged away.

~ March 2016 ~

It kept working for ten years, which for a two-person LLC with no visible revenue is the surprising part. ResearchBuzz noticed a substantial update in October 2010. After that the site is quiet in the record, answering month after month with the same page and the same tabs.

The last capture of the homepage is 9 March 2016. The crawler picked up robots.txt on the 12th. On the 13th it asked for pagenotfound.html and got nothing, and there is no capture of zuula.com after that day, which is unusual: dead domains normally keep getting archived because somebody parks them and the parking page gets crawled forever. Nothing was parked here. The name simply stopped resolving and the Internet Archive has had nothing to collect since.

@}-->--

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

« previous  ·  random  ·  next »