maddy (themadone) wrote,

Yuck. The weather has starting to look nice yesterday, but the afternoon it was awful. Can't see why the crazy folk were sitting outside the pub in the cold and potential wet.

Worked a bit on the LJ Tribe thingy when I got home last night. It should hopefully display some pretty mess of names now, unless you've not actually filled the form in first. Cos it's focusing the mess on "you", so if there aren't any relations stored then it won't know how to connect you to. I know it appears to assign people when you first go to it - but it's doing that based on whether or not they're on your list or not.

Next thing to work on, is some way of being able to display the entire dataset in one go. I suspect that this require a rethink and possibly writing my own code to organise the data, cos the current tool just can't cope with the large amounts of data involved. Although I've got some ideas on how to do it...

Currently the way it's working is that for each person it's got a line between them and all the friends. The lines are invisible and are weighted according to the values assigned on the form. That's an awfully large number of lines on top of a really large number of nodes.

Instead I'm thinking that maybe what I should do is work out a bunch of focal points - the centers of any tribes. Essentially working out the best cliques which should be a lot easier since people are assigning tribal status to folk. Then each name/node only links to the focal point that they're potentially a part of. Weighted in some manner that indicates how much a part of that focal point they are.

I think regardless of how I do it, I'm going to have to cache the image created cos it's probably still going to take a very long time to generate.

Tags: geekery, pub, update
  • Post a new comment


    default userpic

    Your reply will be screened

    Your IP address will be recorded 

    When you submit the form an invisible reCAPTCHA check will be performed.
    You must follow the Privacy Policy and Google Terms of use.