On Tuesday 15 November 2005 03.08, Anthony Towns wrote: [...] > And beyond that, there really are a lot of good ideas stuck in the > -private archives that it'd be nice to be able to refer to properly. This seems to be the only reason to me - the other stated reasons can be paraphrased as 'because govts do it with their secret documents' and 'because everything in Debian should be public'. I find the argument about Debian promising that content to stay private quite convincing, and in any case beating the kind of reasoning that I paraphrased. So, to me, the 'interesting ideas' part is what remains. Since I claim that those really interesting discussions are a minority of the content (even if [VAC]'s are dismissed before the count), I wonder if a opt-in solution, instead of a systematic declassification with opt-out for some content, would not be more appropriate. But having a default declassification date for future content still seems a good idea to me. So there are two sort-of proposals here. So: - the DPL appoints a d-p declassification team. - the team implements the automatic declassification of new content as described below. - for old content, the team will publish selected content of d-private on request. The guidelines below about not-to-be-published content apply, and all authors must consent explicitly. - the current published policy is modified to say that postings to -private may be published after five years unless the author explicitly states that he doesn't wish this. Authors can make their wish known either in the posting or at any later point by notifying the d-p declassification team. - this is not the case for [VAC] messages - this is not the case for sensitive personal information - this is not the case for sensitive corporate information This needs to be put into better language, but I wanted to float this as an informal idea first. I'm not sure - if there should be a 'DPL override' or a 'd-pdt override' for postings made by deceased or totally MIA former DDs. - if 'sensitive information' needs to be specified precisely. Obviously we could hold a 2 year long discussion of what, exactly, is sensitive information, but OTOH I think that in 99% of the cases it should be clear what kind of information is sensitive. And having a list (health, financial, ...) has a high risk of not mentioning some kind of sensitive information that may come up in the future. After all, it's not some dumb automatic parser doing the declassification. (And btw, why three years? Why five years? I don't really care, and if somebody has a reason for a certain number of years...) cheers -- vbi -- Protect your privacy - encrypt your email: http://fortytwo.ch/gpg/intro
Attachment:
pgpNl0lxg28kr.pgp
Description: PGP signature