Subject:
Re: LDP Licence at http://tldp.org/COPYRIGHT.html, and changing in-place
From:
Rick Moen ####@####.####
Date:
26 Dec 2003 13:22:50 -0000
Message-Id: <20031226132249.GP1145@linuxmafia.com>
Quoting David Lawyer ####@####.####
[Licence text that Greg Ferguson allegedly has in some unspecified place:]
> It's www.ibiblio.org/ldp_old but it's not of much use.
Oddly, there is _nothing at all_ at that URL. Wayback Machine finds
nothing, either. More usefully, you later also mentioned:
> http://www.tldp.org/copyright.html
> http://www.tldp.org/LDP-COPYRIGHT.html
That provides one additional document, from among three URLs you've
provided. Thank you. (I also checked Wayback Machine.) I've
accordingly added it to my http://linuxmafia.com/kb/Licensing_and_Law
index.
Two things to note about this additional licence:
o Unlike the other two, I know of no reason to think it was specified
by any LDP document.
o Its name (Linux Documentation Project Copying License) differs from
that of the other two.
I've revised the descriptions at the above-referenced index to better
describe the three licences, including specifying their canonical URLs.
> True. But if the author failed to make a copy of his/her license and
> contents of COPYRIGHT.html changed, no one knows what the license was.
I'm unclear on what your point is.
My point is that informing authors of the situation, instead of papering
over the problem, has the benefits I mentioned previously. The author
may look at one of the variants thus far found, recognise it, say
"That's the one I used!", and change his document's URL to disambiguate
it. Alternatively, he may say "Hmm, I guess I don't remember what I
picked, but I'll select something _now_, to fix this situation of which
I've been made aware." He would have the benefit, at that point, of
LDP's advice about what is deprecated and what is recommended.
> OK. I would normally agree with you. But the situation is that we
> can't find any versions at all that might be useful. I'm sorry that I
> didn't realize this earlier. I had to look over some old emails in
> order to recall the situation better. If you note in the "LDP Policy
> Appendices" which are part of v2.0, it says that the license is at
> LICENSE.html. But no one used that as a link in a HOWTO so no one
> apparently used v2.0.
The fact that that licence's appendix includes an incorrect URL[1] is
completely irrelevant to the question. I have already provided,
repeatedly, an example of a HOWTO that claims to be under the LDP
licence at http://www.tldp.org/COPYRIGHT.html. (Technically, it said
one of the predecessor forms of the URL, when I assumed maintenance.)
> However, since the references in the HOWTOs refer to the LDP Licence
> and don't mention the "Copying License" and also use the words
> suggested for reference in the LDPL, they all seem to be links to
> earlier numbered versions on the LDPL, all of which are lost.
The words used to describe the licence are likewise irrelevant to the
question, for the same reason: The point is that LDP documents (using
mine as an example) claim to be under the licence at
http://www.tldp.org/COPYRIGHT.html , which has unfortunately been edited
in place.
> Thus if we put v2.0 and the Copying License there, we would be putting
> there licenses that no one used.
1. You have cited no reason to believe this to be the case. Explain,
please. Specifically:
2. If you wished to settle that factual question, the logical way to do
it would be to inventory all LDP documents, and find out where they
claim their licence terms can be read. All of them. However, it
strikes me as a great deal easier to just post the ones believed to
have been located _at URLs referenced by authors_.
http://www.tldp.org/COPYRIGHT.html (and equivalents) was/is one
such URL _known_ to have been referenced by authors. Do you have
any reason to believe either of the other two URLs you cited has
been thus invoked by authors?
3. You are ignoring reasons already cited why my recommendation would
apply regardless.
> Except that the LDPL v2.0 is somewhat close to version 1.x which was
> apparently used.
Where may I view the latter licence?
David, I'm getting a sense of deja vu: There seems to have been a
pattern of you making assertions that cannot in any obvious way be
examined. I ask questions about those; most of the time, they go
unanswered.
I'm basically about done. I've done significant work in assembling and
classifying two licences used by LDP authors. Now, you have (after much
wasted time) produced a third that might have been used, but you have
failed to clarify whether there's any reason to think so. Frankly,
assembling and classifying the texts was fairly quick and easy. By
contrast, repeatedly going over the same ground with you has been a
chore.
> So v2.0 is the closest we can come to the actual license.
I cannot determine the semantic content of this sentence.
> Well, I think that we could add a note to the start of v2.0 that it is
> not exactly the same license as the author originall used. Do you
> agree?
I'm sorry, but the question cannot be answered as posed. Please be
about two orders of magnitude more specific. Frankly, it is likely that
the question is moot, as it rests on that stuff about the incorrect URL
in LDPL v2.0's appendix.
> No, both you and I have utterly failed to find the versions used.
1. I think you must be confused. I have located, archived, and
classified two iterations of a licence posted at a URL known to be
referenced in LDP documents.
2. You purport to have some reason to think those specific licences
were never selected by any LDP authors, but I don't understand
why you think so.
3. Moreover, I am puzzled that you persist in ignoring my already
cited reasons why it's worthwhile to proceed as I suggested
regardless of that question.
> > > It would be confusing to have more than one license there.
> >
> > Correction: Your impression of a current _lack_ of confusion is an
> > illusion. That already-present confusion is merely papered over by LDP's
> > unfortunate practice of editing-in-place the historic licence text.
>
> True but, this was 6 years ago.
I'm sorry, but that is incorrect. Please re-examine what I stated about
the Linux User Group HOWTO, by way of example. That is a current HOWTO,
and it is adversely affected by the situation described. I think it
likely that many others are, likewise.
> Late in 1999, Guylhem updated the Copying License at
> http://www.tldp.org/COPYRIGHT.html by removing the statement that it
> applies to all LDP documents.
FYI, that clause was a no-op: Application of licences to creative works
lies solely within the purview of the applicable copyright owners.
> This license is for the "Getting Started" guide. Now if a HOWTO were to
> link to this (and a couple of them did --don't think they still do) do
> these provisions that apply to the "Getting Started" Guide also apply to
> their HOWTO?
Look, David, this isn't brain-surgery: If a HOWTO copyright owner
states that a particular set of permissions applies to an instance of
his creative work, then that is the case.
> It seems that this was a major mistake in the Copying
> License. It's never been corrected.
You are obsessing over meaningless trivia. That clause is likewise a
no-op, for the exact same reasons.
> > Absolutely no.
> I argue yes
Argue until you're blue in the face: You are contradicting the law.
> This changes if the copyright owner uses a license that permits
> relicensing, as LDPL does if one modifies the doc.
This is mistaken. Please read the copyright statute for your country,
as previously mentioned.
[1] Interestingly, the HTML encoding of this URL citation in the
appendix was botched, rendering it absent completely -- except to
someone combing through HTML source. Likely, that explains why the
error was never spotted and fixed.
--
Cheers, "There is hardly anything in the world that some man cannot make a
Rick Moen little worse and sell a little cheaper, and the people who consider
####@####.#### price only are this man's lawful prey." - J. Ruskin (attr.)
