No activity for 18 months; reclassifying as a wish instead of urgent.

The cumulative csv files have a lot less info than I'd like

DATE, "TOTAL", TOTAL, SOURCES

The second is useless.  Want several hundred additional columns -- one for each
license URI.  And come to think of it, additional ones for each license class
(by, by-nc, zero, sampling+, etc.), each version of the six main ones (1.0, 2.0,
2.1, 2.5, 3.0), and each jurisdiciton for each of the six main ones (blank, cz,
uk, etc.)  The "additional ones" should come before the hundreds for each
license URI.

And an additional request -- missing dates should be interpolated, perhaps
noting "interoperated" as the source.

(I'm requesting feedback and suggested changes from Mike.

If he has none, my next step is to document this on the CC wiki and call it
resolved.)

Yo Mike,

I believe http://labs.creativecommons.org/~paulproteus/stats/ml-minimum-estimate
is exactly what you want.

The columns are:

ISO_DATE, LICENSE_URI, COUNT, SOURCES

Where SOURCES is either "Linkback", or "Linkback,Flickr" for the days that are
FLickr-enhanced.

Method described in http://code.creativecommons.org/issues/issue277

Would like to have estimates (at least based on Yahoo + content sites) going
back as far as we have data for either, preferably in one big csv file.  don't
really care whether dates or license urls go in the columns.