Re: [Statcvs-users] mailing list archives + user testing

[Richard C. <ri...@cy...>]

Hi Aaron,

Am 12.04.2005 um 00:15 schrieb Aaron Zinman:

> As I mentioned previously, I'm working on a visualization that shares 
> some similar goals to statcvs, but with the one big addition that I 
> also include communication into my visualization.

That's great -- there's much value in getting all the information 
related to a project (mailing lists, issue tracking, CVS, file releases 
etc.) nicely integrated.


> I'm getting to the point where I want to start pulling in a mailing 
> list archive, and since statcvs has few developers and I would 
> consider successful (I'd be interested to hear your thoughts), I am 
> requesting a single-file archive of the mailing list.  I used to have 
> one (which I accidently deleted), but now it seems that source forge 
> has a web-only interface?  Furthermore, I don't have access to the 
> developer mailing list which is where my interest really lies.

Originally there have been three mailing lists: statcvs-users, 
statcvs-devel and statcvs-announce. All but the first have been retired 
some time ago. This means they no longer show up on the website, but 
are still there in the system.

I don't have any full archives either, but you might be able to ask the 
SourceForge support staff for an archive of the lists in mbox format. 
Ask here:
http://sourceforge.net/tracker/?func=add&group_id=1&atid=200001


> Since I'm attempting to publish on this in the near future (infovis 05 
> due april 20th...soon!), I'll be happy to send a link to my paper PDF

I'd be very interested in that.

> and program (I don't know about source yet.. I have to request from 
> MIT first to GPL it).  Even better, I'd love to hear some feedback on 
> it before the 20th that I might be able to put in my informal 
> evaluation section.

Will happily comment.

> One of the biggest differences of my project in looking at code is 
> tracking where people's code was taken out, and subtracting that from 
> a persons running total.

Yeah, that's a fine approach too. The StatCVS approach measures 
everybody's overall contribution over time, yours measures whose code 
actually makes up the current code base. Both are different things, 
both are interesting and useful. The perfect tool would show both and 
clearly and intuitively explain what both numbers mean.

> Also, I think my visualization looks much nicer than the statcvs :)

I reserve judgement on that until I've seen it ;-)

Cheers,
Richard




> I'd appreciate any feedback/help,
>
> --Aaron
>
>
> Aaron Zinman            az...@me...
> MIT Media Lab           E15-390
> Sociable Media Group    617.452.5606