--On Thursday, 27 June, 2002 14:24 +0100 Lloyd Wood
<l.wood@eim.surrey.ac.uk> wrote:
> Given that a large number of drafts, including even
> draft-bradner-submission-rights-00.txt
> currently end in a boilerplate saying copyright "(year)"
> or an out-of-date year because the boilerplate has been cut
> and pasted from a previous draft, it would be impossible to
> rely on the information that you propose. Draft authors aren't
> generally awfully hot on proofing such details.
ok, so? (see below)
> The IETF has mailing list records, where draft submissions are
> announced and ideas are recorded. Isn't that sufficient?
I don't think so. If one is looking for a reliable source of
information, there is typically less of it in a title, and may
be less in an abstract, than the material you point to. Nor are
those records indexed in any useful way. But neither, I think,
is the point.
I'm not a lawyer, I don't play one on television, and I really
hoped to be able to make this suggestion without starting the
"amateur lawyer" loop again. I haven't said a word about what
can be trusted -- from either a legal or a technical/ "security"
standpoint -- and really don't intend to.
I made the suggestion because I thought it might be helpful to
offer authors of something that someone else might consider an
invention or component of one an opportunity to record and
maintain a backtrace. No requirement. If a given author can't
be bothered, or can't be bothered to do it accurately, no
problem.
If I'm a potential patent applicant, and I've been working on
something, and I'm acting in good faith (that does happen), and
I run across an I-D that seems to overlap what I'm doing, I'd be
happy to have a hint that it has a publication history. Whether
that publication history is relevant or authoritative is
something I'll just have to figure out, regardless of what is in
the backchain. And, as I just told someone in response to a
private note, I'd think, personally, that I'd be much more
likely to skim whatever is in front of me and then try to
initiate a discussion with the author. I would not try to
derive meaning from the specific dates beyond the observation
that the author considered them relevant.
Now, whether this would be of any use at all in cases where good
faith is questionable, I'll leave to the [real] lawyers. I'd
think it would be a bit harder to claim first invention or first
publication, with no reference to prior work, if there is an RFC
or active I-D out there that points to an earlier publication
that predates my claim of when I thought of the idea. I'd also
think it would be harder to make that claim with documents (like
RFCs and many I-Ds) that are picked up by indexing systems than
it would be on the basis of a mailing list announcement of an
I-D, when the mailing list announces hundreds or thousands of
I-Ds a year. But that is just common sense, not legal advice
or estimates of what lawyers might tell you.
> and if you're going to use ISO 8601 date format, specify it's
> that.
I tend to use that date format when I'm expressing all of
month-day-year and when I don't want to explicitly label it as a
date. Personal habit... Because I think this should be
voluntary and informal, I think specifying the format of dates
or other details would be a waste of time. I would strongly
oppose any such rules for I-Ds. It is conceivable that the RFC
Editor might have rules that they would like to apply for dates
in RFCs, but that is an entirely different issues.
john