On Monday 22 November 2010 22:53:36 Nick Barnes wrote:
> At 2010-11-15 11:09:15+0000, Adam McGreggor writes:
> > On Sun, Nov 14, 2010 at 06:19:15PM +0000, Nick Barnes wrote:
> > > At Ravenbrook we have done a lot of social network drawing and
> > > analysis code under contract to a client.  The drawing code all
> > > belongs to us, and we are looking for other uses for it.  We are
> > > considering providing a Google-Charts-alike service, eventually.
> > 
> > If the code's in a repo (and nicely licensed), I imagine others might
> > branch+build on that, if they have a need.
> 
> Unfortunately for this plan the code is not open source (also it's in
> Common Lisp, so not many people have the chops to branch+build on it).
> 
> > (And I'd imagine, that, to 'do things with' the Who's Lobbying
> > database (data), there should be the "attribution, share-alike license
> > for databases" -- per
> > Message-Id: <[email protected]>)
> 
> We would absolutely not intend to impose any unpleasant license on the
> resulting drawings, even if we could.
> 

I've been experimenting with NetworkX and the data in my "rough who's 
lobbying" scraperwiki parser. The biggest issue I can see is that there is a 
*lot* of data - a big challenge to keep the graphs legible.

Interesting result: weighting the nodes according to a rough guess at the 
relative importance of departments, and the edges based on how many lobbies 
took part in each meeting (so a one-to-one is best), the Department of Work 
and Pensions is where it's at. them and the Cabinet Office. 

But I was trying to measure the *lobbies* not the departments. I have some 
other ideas for layout strategies, notably using a version of CAIDA's strategy 
for visualising the Internet routing table.
-- 
The only thing worse than e-mail disclaimers...is people who send e-mail to 
lists complaining about them

Attachment: signature.asc
Description: This is a digitally signed message part.

_______________________________________________
Mailing list [email protected]
Archive, settings, or unsubscribe:
https://secure.mysociety.org/admin/lists/mailman/listinfo/developers-public

Reply via email to