ZipcodeZoo

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

~ the obituary ~

type zipcodezoo.com into a browser today and you get a wordpress site reviewing pet products, with an editor's pick above a block for this week's hottest. the old name is still in the title tag and still in the schema markup, attached to none of what it used to mean. the encyclopedia that lived there is on the internet archive now and nowhere else.

it launched on 26 september 2004 with a goal that sounds impossible and mostly was not: one page for every living species and infraspecies known to science. it reached nearly 3.2 million of them. the method was the interesting part, because zipcodezoo generated almost nothing itself. the base species list came from catalogue of life and the distribution maps from the global biodiversity information facility, with photographs pulled off flickr, video off youtube, bird recordings from xeno-canto and conservation status from the iucn. it was a join across other institutions' databases, published under creative commons by whoever was running it.

that is the part nobody wrote down. the wikipedia article carries a conflict-of-interest tag from march 2015 and a primary-sources tag from the same month, which usually means the site's own operator was editing the article about it. no founder is named anywhere i could find. a project that cataloged three million species left no record of who did it.

~ the case file ~

~ what was in it ~

the raw numbers: over 800,000 photographs, 50,000 videos, 160,000 sound clips, 3.2 million maps. the search tools were better than the aggregation suggests. you could identify a bird from its color and shape plus where you saw it. you could ask for a list of plants or animals likely to be found near a zip code or a set of coordinates, and narrow it to reptiles or to fish. there was a sound trainer that played several bird recordings at once so you could learn to pull one out of the layer.

that last feature is the giveaway that a person built this rather than a content farm. nobody adds a simultaneous birdsong trainer to hit a page count. a sister site at lookup.life carried the same data, and a phone app called lookup life put it in your pocket, which treehugger wrote up in 2015 as a free way to identify the plants and animals around you.

~ what got buried with it ~

the archive shows the ending as a stutter rather than a shutdown. captures through 2016 and 2017 are steady, around thirteen kilobytes of homepage every month. from march 2018 the crawler starts getting 404s, then redirects, then a working page again in november, then a working page in january 2019, then more redirects through the spring. the last real capture is june 2019 and after that there is a two year gap. no closing notice and no farewell page, which is how most sites of this size actually end.

what got lost is not the species data, which still sits in catalogue of life and in the iucn red list, both of them running fine. what got lost is the join. zipcodezoo's value was that somebody had wired those sources to each other and given the result a url per organism. rebuilding that is not hard technically and nobody has, because there is no business in it.

~ the resale ~

the domain came back in 2021 as a content site and is now a pet-affiliate blog with the encyclopedia's own meta description still hanging off some of its pages. there is also a backup.zipcodezoo.com serving a similar thing. this happens to almost every dead site with a decent domain and it is worth naming plainly. the name is an asset and the archive is a cost, and only one of the two ever gets bought.

@}-->--

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

« previous  ·  random  ·  next »