On 9/16/2012 6:56 AM, Lawrence Conroy wrote:
Hi Scott, folks,
with due deference to Joe Touch & Bill Manning, whenever I have created/requested publication of an I-D, it never occurred to me that I was actually withdrawing the rights I had signed up to after six months (i.e., insisting on removal). That seems a novel reading of the boilerplate/2026.
There were times people have had to refresh a draft before it disappeared (with precious few changes), just to keep them published.
That always struck me as bonkers, and am roundly happy that/if expired drafts don't evaporate with the season.
I can't be alone in this -- these docs may not be current, but having a trail of drafts NOT take up gigabytes on [some] people's hard drives is good.
As for Scott's comments/efforts to have an expired-draft directory -- amen to that, for both reasons.
(I hadn't realised that Steve had just about done this; good chap, who's sorely missed).
It is VERY useful to be able to search through drafts to see how we got here, AND to see things that were explored and abandoned.
It's also useful to be able to check when things were published relative to applications; having that information potentially go away was/is a nuisance.
IMHO:
If authors insist, OK -- the expired draft is toast (although it being replaced by a published note that it has been deleted would be good).
Otherwise, why would/should the IESG decide to remove a document?
... which is a long way of saying: +1
all the best,
Lawrence
On 14 Sep 2012, at 16:21, Bradner, Scott wrote:
I don't think that the Note Well note has much to do with what Joe started talking about
we have had this discussion before
quite a few years ago (pre tools) I suggested moving "expired" IDs to an "expired IDs" directory
rather than removing them from the IETF public repository as well as posting all the old
expired IDs in the same directory (changing only the filename to prepend "expired-")
seemed like a good idea to me & to many other people, for the reasons people find the tools ID archive
useful & the IESG at the time said to move ahead
Steve Coya started the (long) process f pulling the old IDs from backup tapes
as he was finishing that process a new IESG was seated and the new IESG was not as in
favor of the idea and wanted a fuller mailing list discussion (there had been a short
discussion when I proposed the idea)
there were a few people (Joe was one) that felt that IDs were published under the rights implied
in rfc 2026, which said that IDs expired, and, thus, the IETF did to have the right to not remove them
(there fact that other repositories existed did not change the IETF's rights in their opinion - in at least
one case someone (I think Bill Manning) sent letters to some of those archives asking that their
ID be removed)
with support from the then IETF lawyer, I suggested that there be a way for a ID author to
request that their ID be removed from the expired IDs directory but that idea did not carry the
day and the expired IDs directory idea died
then, at some later point, the tools function showed up - I do not think I was still on the
IESG at that point so I do not know if the IESG discussed the question
I think this is a very useful service (for history of how the technology evolved
and for prior art searches in patent cases) and think that pretending that anything
published on the Internet ever quite goes away is not realistic.
Scott
The issue isnt whether the publications are available in an archive
taking up space but whether the IP licensing rights to the content of
that draft are persistent through a revision process or not. i.e.
whether that right evaporates with the new publications then anyone
implementing the protocol must also upgrade as well to stay in proper IP
licensing state or it persists.
Further if an IETF Draft is expired and there are support issues (from a
contractual basis between parties who sold code implemented on those
standards to a relying party) what happens to that contract?
One of the things this group has done is to ignore the practices of the
real-world and its time this was addressed. Code implemented from these
publication standards is tied to the flow and protocol practices defined
therein and its a real issue.
Todd
On Sep 14, 2012, at 1:35 AM, Joe Touch <touch@xxxxxxx> wrote:
Note well, as you noted well, does not go back to the beginning of all IDs.
-----
No virus found in this message.
Checked by AVG - www.avg.com
Version: 2012.0.2221 / Virus Database: 2437/5270 - Release Date: 09/15/12
--
//Confidential Mailing - Please destroy this if you are not the intended recipient.