NZBMatrix

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

~ the obit ~

NZBMatrix ran its whole life inside years I was completely online for. It still never touched me, and the reason sits one layer below it. Usenet binaries were a habit that never took here, and what formed instead was torrent-shaped and needed nothing beyond a client and a tracker page you could reach from a browser. The machines around me in that period had bittorrent on them. Nobody I knew was buying newsgroup retention.

No source I can find gives a founding date for the site. The honest floor is the first archived capture of nzbmatrix.com, on 15 July 2007, which is what the years above are built on. By the early 2010s it was the largest Usenet index most people could name, serving millions of visitors a month, with its heaviest user bases in Denmark and the UK.

It died on a Sunday morning, 9 December 2012, by its own hand. The notice left on the front page was written by people who had spent years being scrupulously compliant and had just finished the arithmetic on what continued compliance would cost them.

~ the rap sheet ~

~ what it actually did ~

A film posted to Usenet does not arrive as a film. It arrives as hundreds of encoded parts spread across newsgroups under filenames that mean nothing to a human, and the parts age off the servers on a retention clock. An NZB file is a short piece of XML listing exactly which message IDs to pull and in what order. NZBMatrix ran bots that watched the raw feed and generated those lists automatically, which is what separated it from the earlier generation of hand-edited indexes.

The distinction matters legally. The site held no copyrighted bytes at any point in its life. What it held was the knowledge of where the bytes were, which by 2012 several courts had already decided was enough.

~ the notice ~

The takedown that ended it came from Wiggin, acting for the Federation Against Copyright Theft together with six film companies running from Paramount to Warner Bros. It was not the first notice the site had received. It was the first one large enough to make the operators think past it.

Their closing statement made the reasoning plain. They had always honored takedowns and always would. Once this one was processed they would be left policing their own indexing bots forever, because the bots would keep finding the same material the moment somebody reposted it, and an indexer whose content no longer exists is not an indexer. The payment side had been failing underneath all of this for months. In their own words they had been through pretty much everyone out there, and in the end they all pull out.

~ the long tail ~

The format outlived the site by more than a decade. Every Usenet automation stack still speaks NZB, and the indexers that filled the gap mostly cloned the API rather than the interface. Usenet Crawler had already launched that year and took a share of the refugees. An unrelated forum picked up a near-identical name within weeks, which is its own kind of epitaph.

What the 2012 wave established is that the index is the most exposed thing in a distribution chain. The storage sat in Amsterdam and Virginia and nobody sued it. The search box was the part with an address on it.

@}-->--

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

« previous  ·  random  ·  next »