WebCite

» see webcitation.org as it lived, on the wayback machine

~ the goodbye ~

WebCite was Gunther Eysenbach's answer to a problem almost nobody was treating as urgent in 1997. A paper cites a web page. The paper is permanent and the web page is not, and a few years later the citation points at a 404 or at something that has been rewritten since the author read it. He described the idea publicly in 1998 and ran a pilot at webcite.net the year after.

The mechanism was plain. You handed it a URL, it took a snapshot right then, and it handed back a permanent address you could print in a reference list. It stored the HTML along with the images and stylesheets that made the page mean what it meant, plus metadata recording the moment of capture. It did not crawl. It only archived what somebody specifically pointed it at, which in 1999 made it the only service of its kind anywhere.

That idea is now standard equipment. The Wayback Machine has a Save Page Now button, and Perma.cc exists because law reviews needed exactly this. Both of them arrived a decade or more after WebCite did it first.

~ the chronicle ~

~ how it worked ~

For an individual scholar it was free and required no account at all. The money was supposed to come from the publishing side: a journal could hand WebCite its back catalog as XML and have a crawler walk the references retrospectively, archiving whatever was still standing. Eysenbach was open about retrospective archiving being the inferior option, since there is no guarantee that what the crawler finds in 2008 is what the author saw in 2003.

The legal position was argued carefully. WebCite held that its copies were fair use and covered by an implied license, and it honored no-archive metadata and robots.txt exclusions, which is the behavior that generates the implied license in the first place. On receiving a valid DMCA notice it did not delete anything, it moved the snapshot into a dark archive, where a copy could still be produced for a legal proceeding at $200 for up to five snapshots and $100 for each further ten. Lawrence Lessig cited WebCite copies in his amicus brief for MGM v. Grokster.

~ the money ~

There never was any. In 2012, asked on Twitter why WebCite had dropped out of the International Internet Preservation Consortium, Eysenbach answered that the consortium charged four thousand euros a year in membership fees and WebCite had no funding.

In January 2013 it ran a FundRazr campaign asking for $22,500, the sum it said it needed to move onto Amazon EC2 and cover legal support past the end of that year. The campaign page could not say whether WebCite would stay non-profit or turn into a commercial service, because that had not been decided.

~ what it is now ~

Sometime between 9 and 17 July 2019 it stopped accepting new archiving requests, announced by nothing louder than a banner appearing on the front page. Archive Team went after the S3 bucket that same month and pulled 31 million items out of it, 648 GB compressed. A significant fraction of what they captured has since been removed from the bucket at the source.

Then it went dark properly. From around 29 October 2021 until 24 June 2023 nothing worked except the front page, which reported "DB Connection failed" for a year and eight months. It came back. webcitation.org answers today and the old snapshots resolve, and the submission form is still sitting on the page under the banner saying requests are not being accepted. The page source still carries a Google Analytics urchin.js tag with a comment noting the day it was installed, 16 January 2008.

From the reading side this was always invisible. WebCite appeared as a webcitation.org link in somebody else's reference list and you either clicked it or you did not. That is the part worth sitting with: those reference lists are printed and permanent and still in circulation, and what they point at has been held open for more than twenty years by one person with no money, currently answering requests it will not fulfill.

@}-->--

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

« previous  ·  random  ·  next »