Jon’s Radio Comments

December 13, 2006

Data formats for digital democracy: XML vs CSV

Filed under: Uncategorized — jonsradiocomments @ 2:53 pm

The original item is here.



  1. I was just searching on this migration the other day, always looking for an easier way to visualize my delicious database. One day I’ll hit pay-dirt. Thanks.

    Comment by BillyG — December 13, 2006 @ 5:30 pm | Reply

  2. Hi Jon,

    We just happened to do a new deployment of Swivel early this morning. Not sure if you saw the page before or after that deployment. We had a bunch of stuff broken at launch, including our Swivel robot that generates graphs for us (I had a nasty bug in the crawler code I had written, yikes). If you go there now, do you see the same stuff you saw when you posted? At the moment, for your uploaded data Swivel has generated 1 primary trend graph, 8 scatter plots and 8 comparison trend graphs for each column. that’s about 150 or so graphs (plus each graph’s data table) for you and the other Swivelers to cruise.

    Thanks for the post about Swivel.

    Brian Mulloy
    CEO & Cofounder

    Comment by Brian Mulloy — December 13, 2006 @ 6:49 pm | Reply

  3. Hi Brian,

    Yes, there’s more there now. It would be interesting to hear more about the purpose and method of these elaborations. And also, since a month’s worth of data is not anything to base trends on, it’d be interesting to see, for some more robust data set:

    – Some examples of useful correlation within and across data sets

    – How people can make the useful ones bubble up from a background of junk correlations

    Comment by Jon Udell — December 13, 2006 @ 9:32 pm | Reply

  4. Hi Brian,

    Any chance of having the Swivel Data in one of the following forms:
    1. Valid XML
    2. Atom
    3. RSS 1.x or 2.0
    4. RDF/XML

    If you can get 1,2, or 3. I can get you RDF/XML by way of automatic generation etc..

    Comment by Kingsley Idehen — December 14, 2006 @ 5:04 pm | Reply

  5. This looks to be an easiest way

    Comment by Robin — December 15, 2006 @ 2:30 pm | Reply

  6. Hi Kingsley Idehen,

    Right now Swivel is csv only. We have an internal prototype of an API that supports input and output with valid XML. Look for us to harden that API and release it in the first half of 2007.

    Comment by Brian Mulloy — December 19, 2006 @ 5:42 pm | Reply

  7. Hi Jon,

    I just realized my long-winded response to your questions didn’t get posted last week. That might be better for all us 🙂

    Briefly, your point about “useful ones bubble up from a background of junk correlations” is one of the most important problems we have to solve at Swivel. 2 weeks ago we had only our friends and family using Swivel and very little data. Today we have 1400 data sets and 2300 registered users. So, now we can do a bunch of interesting tweaks to the rules in Swivel that help the interesting stuff come to the top.

    Brian Mulloy
    CEO & Cofounder

    Comment by Brian Mulloy — December 19, 2006 @ 5:50 pm | Reply

  8. for me its easier to use xml standart. But csv I think is usefull if you need to do integrated jobs with MS programs, etc.

    Comment by justinas — December 21, 2006 @ 2:37 pm | Reply

RSS feed for comments on this post. TrackBack URI

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

Blog at

%d bloggers like this: