BlogBridge

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

~ the obit ~

The announcement is still in SourceForge's news feed, timestamped 19:51 UTC on 10 November 2003: "BlogBridge open for business", describing a new kind of blog aggregator, an end-user application written in Java, currently in a pre-alpha state. The author was Pito Salas, who had invented the pivot table at Lotus in 1986, then built the collaboration company eRoom until Documentum bought it in 2002. A desktop feed reader was his next move, and he stayed with it for eight years. The logo in the source tree gives the team's address as Sebastopol, Ukraine and Boston, USA.

It was not built for somebody like me. BlogBridge was aimed at the reader who tracked hundreds of feeds as a job, which in 2005 meant journalists and PR people who needed to know what had been said about a client that morning. I was at university with maybe fifteen subscriptions, most of them tech blogs I could have checked by hand, and a Java desktop application was a heavy way to do that on a 512 MB machine. The people this was for had a different relationship to the firehose than I did.

What it did well was organize that firehose. From version 6 there was a memo tracker and a statistics view, so you could see which of your subscriptions had gone quiet without reading them. Smart feeds arrived late, and one of the last ones anybody wrote for it read Twitter lists instead of RSS. By 2011 that was where the syndication worth following had gone.

The last packaged release was 6.6.1 on 26 February 2009. Work continued in the open after that without producing another version number, and the final three commits in the repository landed on 2 November 2011. One of them adds a script for building and deploying JNLP releases.

~ the rap sheet ~

~ feeds for people who had too many ~

The free account is what lifted it above a plain desktop download. You uploaded your feed list to blogbridge.com, and your subscriptions followed you between machines and survived a reinstall. That was a genuine problem in 2005, when a feed list represented real curation work and lived in one OPML file on one hard disk.

It also meant the program had a server dependency, and server dependencies are what turn an abandoned desktop app into a dead one. A Java binary from 2009 will still start today if you can find a runtime old enough to humor it. It will not find its account, because the host stopped answering in 2018.

~ the shop stayed open ~

Wikipedia records that BlogBridge appears to have been unmaintained since 2013, and that is roughly right for the code. The website runs on a different clock. In the archive's index, blogbridge.com answers with its own page of around seven kilobytes continuously from 2010 into 2018, the last of them captured on 3 April 2018. Five weeks later, on 9 May, the same address returns 833 bytes, which is roughly what an empty page weighs.

By the end of 2019 the host was throwing a 502 and then a 403, and today the domain does not come back at all. Nobody wrote a farewell post. No announcement, no sunset date, just a download page that outlasted its own developer by six and a half years.

Most RSS readers died of Google Reader, either crushed by it while it ran or orphaned when it closed in 2013. BlogBridge was already finished by then, so it got neither death. Its ending came five years later, on whatever day the hosting bill went unpaid.

@}-->--

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

« previous  ·  random  ·  next »