Richard L. Hamilton wrote:
A third metric (arguably as silly) is the percentage of bytes in files in the
whole
Solaris distro derives entirely from open-sourced source files (counting shared
libraries separately from what links to them).
Perhaps you just need to:
* think about what metric is most acceptable to the audience for this info
(whoever that
will be, but keeping in mind that it will probably have to look reasonable in
public
one day too)
* be sure to describe what metric(s) you used along with the raw numbers. That
way,
at least there would be truth in advertising, and less excuse for someone
disputing the
figure (assuming it was accurate according to your chosen metric)
I'd oppose such a metric. The underlying bits (Nevada) should be as
small as possible, so a metric doesn't look good for just Nevada. Even
if you loop in all things open source, JDS, Nevada, etc, it still
requires explanation.
The problem with supplying one number (a percentage in this case) is
thats all they want and all they need. No explanation will be read, by
and large. We'd just be setting ourselves up to look bad. When that
percentage hits 100% I'll feel diffrent, but even "Solaris is 98% Open
Source!" looks bad when your looking for something to harp on.
benr.
_______________________________________________
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org
