Thottbot

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

~ the obit ~

thottbot's peak years line up almost exactly with mine. 2004 to 2010 was ttnet broadband at home and the tail of university, and i was on a machine for most of every day. the distance here is not time or bandwidth. it is that thottbot only mattered if you played world of warcraft, and i did not. my mmo years had gone into turkish ultima online private servers, and by the time warcraft was the thing everyone meant when they said mmo, i was doing something else with my evenings.

bill dyess, who had run the everquest guild afterlife under the name aftathott, put the site up on 31 march 2001. it started as a general news aggregator for online games, which is what most of these things were before one game got big enough to deserve its own reference work. in august 2004 he rewrote it as a searchable database, and from then on it was about one game only.

the reason it deserves a grave is not the traffic, although the traffic was absurd. it is the collection method. thottbot did not have access to blizzard's data. nobody outside blizzard did. the site knew what an item dropped from because players had gone out and killed the thing themselves, then sent the result back.

~ the rap sheet ~

~ how the data got there ~

you installed a lua addon that sat inside the game client and quietly recorded what you saw. loot tables and vendor prices, down to the coordinates where a thing had spawned. when you were done playing you uploaded the log to the site, and your evening became somebody else's search result.

this had a strange consequence that people at the time understood instinctively. the database was only as complete as the population running the addon. an item nobody had bothered to upload simply did not exist as far as thottbot was concerned, and a drop rate was an average over however many kills had happened to be reported. the numbers on the page were not facts handed down from the developer. they were a poll of the playerbase, with all the sampling problems a poll has, and everybody used them anyway because there was nothing else.

~ the parallel years ~

by march 2008 alexa had thottbot at number 250 worldwide. that is a fan-run database for a single video game sitting inside the top three hundred sites on the internet, which says more about world of warcraft in 2008 than any subscriber figure does.

its owner by then was the zam network, and in june 2007 zam had spent a million dollars buying wowhead, the rival database that did the same job with a better interface. so one company now held both of the sites its users argued about. they ran in parallel for more than three years. people had genuine allegiances, mostly formed by which one they had bookmarked first, and mostly unaware that the argument had stopped having a corporate side to it in 2007.

~ the long tail ~

on 30 november 2010 zam announced that thottbot's backend was being folded into wowhead's. from that day a thottbot search returned wowhead results. the brand stayed on the door for another six years as a redirect with a skin, until the pages came off entirely on 16 november 2016.

the domain still resolves. it now points at wowhead classic, which is the reference site for a re-release of the game thottbot originally documented, so a url that once cataloged a living world now catalogs a museum copy of it. the addon model outlived all of it. every modern game database that asks you to install a client-side uploader is running the thing bill dyess shipped in 2004, and most of them do not know whose idea it was.

@}-->--

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

« previous  ·  random  ·  next »