Development of Creative Commons licenses

["Jonathon Blake" <jonathon.blake AT gmail.com>]

Nancy wrote:

> we do not have more time and money to do much else at the moment, as we are 
> currently operating on a shoestring due to funding cuts in government funding 
> ...

June is probably the wrong time of the academic year.  When the
academic year next starts (or for summer quarter), get somebody in the
CSc Dept  (or Linguistics, English, Sociology, Anthropology, or
Grammatology departments)  to do a "special project" for academic
credit that:
i) Scans the web looking for material under the different "open
source" licences;
ii) Identifies the creator of said material, and sends an email to
them, requesting permission to use their material;
iii) Responds to messages sent by the person from whom permission was 
requested;
iv) Downloads the material for which permission has been granted, and
hands it off to the bot that does whatever linguistic analysis your
project does.
v) Scan usenet, requesting permission from copyright owners to include
the data from their usenet messages;
vi) Scan blogspace, requesting permission to include material from
blogs, then downloading said material;
vii) Scan IRC, requesting permission from users, and grabbing the material.

There are several PERL modules that either do, or can be modified to
do i, ii, iv, v, vi, and vii;
A procmail script can handle iii.

All that really needs to be done, is for somebody to glue the PERL
bits together, and put it on a (probably dedicated) Linux box. Maybe
have four or three PERL programs.  One for usenet, one for blogspace,
one for the web, and an IRCBot.

[ All of this software is free --- unless your organization accepted
the Microsoft licence that prohibits any other platform from being
used by your organization. If your organization did accept that
Microsoft licence, then the CC NC licence shouldn't even be on your
radar as a "problem".]

> We do not verify that contributors ... are native American speakers--the 
> point of the material on the web site is for self-screening.

Hmm.  the aim is "American English", but the project does not verify
that contributors are "American English" users.  Seems like data
contamination would be an issue.

> the demographic information they provide when they contribute gives us the 
> info we need to determine it, if necessary.

Including the request in the initial question/permission request that
the bot sends out would be trivial.  One potential issue is how much
data is requested.   A short list of questions, with clear yes/no
answers, after the initial request for permission, would probably have
the most positive response.

xan

jonathon
--
Ethical conduct is a vice.
Corrupt conduct is a virtue.

Motto of Nacarima.