Paul O'Malley - gnu's not unix - schreef:
Hi,

binary large object

when you find this within source code but without corresponding source you have found a blob ..

very, very simple

if it contravenes the four freedoms happy scalpel use and removal

This leads me to a set of thoughts I have been having about the "removed code".

Perhaps there is room while doing this removal to inform the "copyright holder" for that block of code or submitter in the first case that it is on the way out .... your call.

To this end I would suggest that a "sane" standard letter is composed and then used by someone or a "function" which is several people, who then try to find authors or copyright holders to try to get the source.

If this is done in a really polite way if may actually help the cause because people would be informing the developers and development company that submitted the code that their hardware is actually useless to users of GNU/Linux systems.

This task would include people trying to find someone within the company or its successor in interest to go and release the code, or perhaps better yet to actually be active when it comes to kernel supply.

The approach is not so much saying "YOU ARE WRONG" as saying, did you realise that there was more to supplying code.

Then if the block of code is not provided at least an audit trail of who was contacted and what the motivation for contacting this particular party.

This work would require people to be discrete as you may not get what you want straight away, you might contact the wrong person but stumble up the right one later, thus you don't want to leave burning bridges behind you. (I.E. This is a political mission which may have real source code impact so should be done in a calm cool way.) It would most likely be very frustrating for those involved but if tackled in a polite way maybe upstream would benefit, eventually benefiting the whole community.

Just because I had the idea does not mean that I want to be active with it. If I was doing it I would look at things which have been added last to the "tainted" kernel.

If so after doing some preparation perhaps this should be placed under the guardianship of the: Workgroup for fully free GNU/Linux distributions <gnu-linux-li...@nongnu.org>

This sort of "license compliance lab" has been on my think-about-this-some-more list for a long time. I was actually talking to Marco Chieppa about something like this yesterday on #gnewsense-dev, in relation to bug 345 [1]. There are patches that we get for free from Debian, but if we'd make our own package we'd have to remember freedom problems.

The free distro compliance team would work (in cooperation with the FSF compliance team) on the issues we encounter in practice. Linux is a special case because it's so big and we remove a lot from it. So I thought it would be nice to have an 'inside man' on it to nudge contributors in the right direction (going from proper copyright statements to saying no to blobs). Maybe Alexandre Oliva already does this to some extent. For packages that have an 'inconvenient' license (e.g. Artistic) the copyright holder (who sometimes doesn't that much) could be asked to change it.

As I said, I haven't thought this through yet, this is just a braindump. If anyone wants to take this on I suggest either starting out with a new project for the freedom verification team [2] or (as Paul said) contacting the GNU/Linux Libre mailing list. I suspect this to be a rather challenging task.

[1] http://bugs.gnewsense.org/Bugs/00345
[2] http://wiki.gnewsense.org/VerificationTeam/VerificationTeam


_______________________________________________
gNewSense-users mailing list
gNewSense-users@nongnu.org
http://lists.nongnu.org/mailman/listinfo/gnewsense-users

Reply via email to