Bookogs

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

~ the obit ~

Discogs spent twenty years proving something unglamorous: that if you let strangers type release details into a form, and you give them a credit system strict enough to argue over, you end up with a better music catalog than any label ever kept. Around the start of 2015 the company tried the same trick on other objects, spinning up Gearogs for instruments and Comicogs for comics alongside a film database and a book database. The book one was called Bibliogs.

Bibliogs had to change its name. In August 2017, with a little over 31,000 titles in the database, legal trouble over the old name forced a rebrand, and the site became Bookogs. That is a strange thing to survive in your third year and an early signal that nobody upstairs was watching very closely. It kept growing anyway. Twenty one thousand books by the end of 2016, a hundred thousand items by June 2019, and 155,120 by the time the lights went out.

The model was the thing people loved. A book on Bookogs was not a product listing, it was an edition, with a cover scan and a credit list that went past the author to the translator and the person who designed the jacket. Those are exactly the people that library catalogs tend to lose. For five years a few thousand volunteers put them back.

~ the rap sheet ~

~ the rise ~

By January 2016 Gearogs was out of beta and the other three were still in it, which is roughly how they stayed. A marketplace opened on Bookogs in 2017 so people could buy and sell the editions they had just finished describing. That was the moment it stopped being a hobby database and started looking like a business, and it is also the moment the comparison with Discogs proper became unkind. Discogs moves records at enormous volume. A used paperback is not a record, and the margin is not there.

~ the cracks ~

Comicogs went first, on 31 July 2020. A month later Bookogs, Filmogs, Gearogs and Posterogs all closed on the same day, 31 August. Discogs said it could not give the sibling sites adequate attention and was concentrating on the music database. It is an honest reason and a cold one. Nobody had done anything wrong. The work was simply worth less to the company than it was to the people doing it.

To its credit, Discogs did not just delete. A full data export and a collection of the uploaded images were prepared and published by 19 October 2020, so the catalog exists as a file somewhere even though the site does not. That puts it ahead of most things in this graveyard, where the usual exit is a 404 and a forwarding address.

~ what got buried with it ~

I never typed a book into Bookogs. By 2015 I was thirty one and working on computer vision, and the hours I used to spend organizing things for free had gone into work that paid. But I understand the people who did, because the same instinct built every reference site I relied on as a student, and almost none of those were maintained by anyone with a salary.

What died on 31 August 2020 was not the data. The export saw to that. What died was the argument: the discussion threads where two strangers worked out whether a reprint with a different cover counts as a new edition. That reasoning is the actual product of a crowdsourced catalog, and it does not survive a CSV.

@}-->--

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

« previous  ·  random  ·  next »