9 Nov 2009 federico   » (Master)

Mon 2009/Nov/09

  • Zeitgeist Hackfest, Monday

    The Zeitgeist team and yours truly are in an energetic hackfest in Bolzano, Italy. We are busy cooking the future awesome.

    Busy cooks

    Next door to our hacking room in the school, the Sugar people are having their own hackfest. Sugar has a Journal similar to Zeitgeist's, so we need to share ideas and if possible an implementation of the journal-of-your-stuff.

    We have three sub-teams within the hackfest: the Zeitgeist engine, the Tracker metadata storage, and the end-user GUI tools.

    Zeitgeist engine

    The core of Zeitgeist is a daemon that logs events that get produced as you work on your computer. When you open a file, that makes an event. When you view a web page, that's an event. Having an IM conversation, looking at a video, and playing music are all events.

    Zeitgeist logs these events and provides an API to query those events for useful purposes. The Journal, of course, shows you a stream of your most recent events. More sophisticated clients can ask the engine questions like, "what applications have I launched within the past week?", "what files have I used for the longest periods of time within the last month?", "which web pages did I visit while editing my-research-paper.odt?".

    Seif Lotfy, Ivan Frade, Mikkel Kamstrup, Alex Gabriel, Markus Korn, and Siegfried Gevatter are working on the engine. They are upgrading the format for the engine's database so that it can support more efficient queries, redoing the external API that clients can use through D-Bus to make queries on the engine, integrating support for Tracker as a metadata storage, and improving the contextual relevancy engine. This last bit is the magic that can tell you, "these are the documents/web sites/etc. that you used together while working on a certain project".

    Tracker as a metadata storage

    Zeitgeist would like to show you various useful things: the tags that you have assigned to a file (School, Work, Pr0n, Travel), the sources for email attachments or for files that you downloaded from web sites, etc. All of that is just metadata that needs to be stored somewhere, and in some well-defined format.

    Zeitgeist used to have a home-grown metadata repository, but it will be using Tracker from now on. Tracker is a metadata storage for RDF triplets. An RDF triplet is a subject/verb/object chunk that looks like "my-thesis.odt/has-tag/School", or "lolcat.jpg/was-an-attachment-from/$email_id". Those are just examples; the actual form of those triplets is formalized in the Nepomuk ontology.

    Gnome Activity Journal

    The Gnome Activity Journal is a prototype implementation of the journal idea that I presented during GUADEC last year.

    And they are closing the school right now, and we have to leave, so I'll tell you more about all of this tomorrow. Ta ta!

Syndicated 2009-11-09 12:02:00 from Federico Mena-Quintero - Activity Log

Latest blog entries     Older blog entries

New Advogato Features

New HTML Parser: The long-awaited libxml2 based HTML parser code is live. It needs further work but already handles most markup better than the original parser.

Keep up with the latest Advogato features by reading the Advogato status blog.

If you're a C programmer with some spare time, take a look at the mod_virgule project page and help us with one of the tasks on the ToDo list!