License utilization metrics (was Re: [linux-elitists] RMS is at it again)

kmself@ix.netcom.com kmself@ix.netcom.com
Sat Dec 2 12:37:00 PST 2000


on Fri, Dec 01, 2000 at 11:01:32PM -0800, Heather (star@starshine.org) wrote:
> > on Fri, Dec 01, 2000 at 03:11:41PM -0800, Heather (star@starshine.org) wrote:

> > > As a curiosity: Anyone have a database somewhere of apps by license?
> > 
> > SourceForge does:
> > 
> >     http://sourceforge.net/softwaremap/trove_list.php?form_cat=13
> > 
> > ...drill down and you'll find that this is a software map browser by
> > license category.
> 
> Hmm.  I'll probably read it a bit during LISA.

> > However, my experience in crawling through /usr/doc/*/copyright a few
> > weeks back was that there is a strongly pronounced lack of consistancy
> > about how project copyrights are reported.  Anything from a
> > straightforward presentation to a detailed list of contact attempts with
> > original authors and/or transcripts of conversations or email.
> > 
> > It would be useful to tie this data to the popularity test though....
> 
> Interesting, how would we graph that?

The intent would be to come up with a license-use metric.  There are
several statistics which are interesting and meaningful, though of
varying diffuculty in measurement:

  - Use of licenses in software projects -- available through
    SourceForge or by ananlysis of Debian copyright files.

  - Installations of software by license -- combine above data with a
    popularity contest.  This is what I was pointing at in my comments
    above.  The metric would be something like "installed utilization":

        software[license] * installs

    ...or possibly a recency of installed utilization, where we look at
    software under licenses used within some time period, e.g.:  1 day,
    1 week, 1 month, or however the Debian popularity contest measures
    things.

  - Swapping of software -- an audit of software by sources -- what
    original projects have contributed to what derived works.  This
    points directly to the significance of code-sharing in GPL.
    Unfortunatly, a full measure would require significant code
    forensics.  A simpler metric might be to look at use of L/GPL'd
    libraries in various projects, though this misses much of the
    alledged direct benefits of a copyleft license.

-- 
Karsten M. Self <kmself@ix.netcom.com>     http://www.netcom.com/~kmself
 Evangelist, Zelerate, Inc.                      http://www.zelerate.org
  What part of "Gestalt" don't you understand?      There is no K5 cabal
   http://gestalt-system.sourceforge.net/        http://www.kuro5hin.org
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 232 bytes
Desc: not available
Url : http://allium.zgp.org/pipermail/linux-elitists/attachments/20001202/6c12d249/attachment.pgp 


More information about the linux-elitists mailing list