RE: SUO: RE: Cyc Upper Ontology offer
As a small irony, it should be noted that the quote which Doug cited in
reply to Graham's message originally came from Pat Hayes (see notes attached
below). Not that this is anyone's fault, yet this illustrates the point that
it is easy (especially in newsgroups) to lose attributions to the original
source of statements, which is the topic and motivation of the copyright
clause under discussion for OpenCyc. Just so, statements that people tend to
agree about may quickly become commonplace and their origins lost in "common
sense".
I agree with Chris (and Graham and Pat and Jon and Mike U. and ...) that no
one questions the moral and legal obligation to give credit where credit is
due, and that we need to clearly understand whether there is a potential for
problems involved with giving credit for copyright to statements and
concepts that did not actually originate with Cyc.
This seems particularly important because of Doug's suggestion that the
OpenCyc copyright clause is intended to protect against what he calls
"semantic plagiarism". If the intent is to copyright or claim ownership to
semantics, rather than syntax, then some unexpected problems may develop.
For example, suppose Cyc happens to include some common axioms, say for set
theory, or geometry, or group theory, etc. If someone uses Cyc, or some
portion of Cyc, do they have to say that Cyc owns the copyright to these
axioms, even though these axioms and their semantics predated Cyc? Anyone
who happened to write a system that contained these axioms would produce
code for which in principle there might be a translation to/from the
representation of the axioms contained in Cyc.
Hopefully one way out of this quandary is to interpret the terms of the Cyc
legalese very literally. The sentence "The terms of this paragraph equally
apply to renamings and other logically equivalent reformulations of the
Material in any natural or formal language." could be interpreted as meaning
that the terms apply in one direction only, i.e. to renamings and
reformulations starting from the Cyc material, but that the terms do not
apply to independent developments of syntax or semantics that happen to be
equivalent to the Cyc material.
If this is the correct interpretation, then it would seem that this
interpretation should be clearly specified as an addendum in any agreement
between Cycorp and SUO.
Some further questions arise along these lines, e.g., does Cyc itself have a
mechanism for disclaiming copyright and giving attribution to concepts it
includes that originated elsewhere? Does Cyc have a mechanism for
disclaiming copyright (or acknowledging public domain) for concepts for
which there is no clear origin, yet which it is clear that everyone knows,
e.g. that humans normally have two eyes? Does Cyc have acknowledgments to
the works of Aristotle, or one Charles Sanders Peirce? (1839-1914) :-)
Cheers,
Phil Jackson
-------------------------------------------------------------------
"Imagination is more important than knowledge. Knowledge is
limited. Imagination encircles the world." - Einstein
-------------------------------------------------------------------
Standard Disclaimers. www.philjackson.prohosting.com
Chris Menzel wrote:
> Doug Lenat wrote:
> > The intent is not for Cycorp to claim credit for Aristotle's work,
> > but rather to avoid a situation where someone runs some simple
> > mechanical transformation of our axioms, producing a set of
> > equivalent ones for which they then take credit. (Normal
> > copyright protection does not protect against such "semantic"
> > plaigarism, since the character-by-character appearance of
> > the axioms would be different from the form in which we supplied
> > them, even if all the "adapter" did was alter the names of the
> > predicates and variables in some systematic manner.)
> Doug, I don't think that Graham's remark questioned the moral and legal
> obligation to give credit where credit is due, but rather concerned the
> problems and possible complications involved in determining *when* such
> credit is due, especially regarding the axiomatization of rather general
> logical and ontological principles.
> -----Original Message-----
> From: owner-standard-upper-ontology@ieee.org
> [mailto:owner-standard-upper-ontology@ieee.org]On Behalf Of Chris Menzel
> Sent: Friday, March 16, 2001 6:05 PM
> To: Doug Lenat
> Cc: 'hovy@isi.edu'; 'fritz@cyc.com'; IEEE Standard Upper Ontology List
> Subject: RE: SUO: RE: Cyc Upper Ontology offer
>
>
>
> On Fri, 16 Mar 2001, Doug Lenat wrote:
> > Graham,
> >
> > The intent is not for Cycorp to claim credit for Aristotle's work,
> > but rather to avoid a situation where someone runs some simple
> > mechanical transformation of our axioms, producing a set of
> > equivalent ones for which they then take credit. (Normal
> > copyright protection does not protect against such "semantic"
> > plaigarism, since the character-by-character appearance of
> > the axioms would be different from the form in which we supplied
> > them, even if all the "adapter" did was alter the names of the
> > predicates and variables in some systematic manner.)
> >
> > Since we aren't asking for royalties or license fees, I expect
> > there shouldn't be much objection to giving us the requested
> > citation/acknowledgement. I'm sure you would expect the same
> > in the reverse situation: i.e., you've written a corpus of 30,000
> > axioms and some fellow scientist asks you for a copy, and you
> > supply it, and they then adapt and built upon it for their
> > work (academic or commercial); then surely in that situation
> > it would be wrong for them not to credit your contribution.
>
> Doug, I don't think that Graham's remark questioned the moral and legal
> obligation to give credit where credit is due, but rather concerned the
> problems and possible complications involved in determining *when* such
> credit is due, especially regarding the axiomatization of rather general
> logical and ontological principles.
>
> -chris
>
> --
>
> Christopher Menzel # web: philebus.tamu.edu/~cmenzel
> Philosophy, Texas A&M University # net: chris.menzel@tamu.edu
> College Station, TX 77843-4237 # vox: (979) 845-8764
========================
>
> At 04:13 PM 3/16/01 +1100, Horn, Graham wrote:
> ><snip> Notice that the agreement requires that any use of any
> axioms which
> >are *logically equivalent* to any part of Cyc, even if
> translated into any
> >other natural or artificial language, must be credited to
> Cycorp. This means
> >that if one uses any part of the Cyc upper ontology and if they
> also assert
> >that, say, every relation is a thing, or that spatial containment is
> >transitive, then they must acknowledge that Cycorp owns the copyright to
> >that assertion, even if it is made in, say, EPISTLE, or even in
> English, and
> >even if they had said it before in other places; in fact, even
> if Aristotle
> >had said it before. </snip>
> >
> >Wow!
> > . Maybe not so marvellous after all. I presume this means
> >there is at least considerable negotiations in front of us before we can
> >even look at taking up Cyc.
> >
> > . I would suggest Cyc would get considerable advantage by
> >being adopted by the SUO. I would hope that would make it attractive for
> >them to retract this condition. I presume the clause would be totally
> >unacceptable to all of us, as well as the IEEE.
> >
> > . The offer did seem too good to be true. Maybe it was.
> >
> >
> >
> >Cheers Graham Horn
> >National Data Standards Unit
> >Australian Institute of Health and Welfare
> >================================================
> >Phone: 02.6244.1094
> >Fax: 02.6244.1199
> >E-mail: Graham.Horn@aihw.gov.au <mailto:graham.horn@aihw.gov.au>
========================
> -----Original Message-----
> From: owner-standard-upper-ontology@ieee.org
> [mailto:owner-standard-upper-ontology@ieee.org]On Behalf Of pat hayes
> Sent: Thursday, March 15, 2001 11:06 PM
> To: Horn, Graham
> Cc: standard-upper-ontology@ieee.org
> Subject: Re: SUO: RE: Cyc Upper Ontology offer
>
>
>
> >Dear all,
> > . Marvellous! Thanks indeed to all involved in
> >achieving this situation. I hope it clears the attorney, so we
> can use it.
> >
> > . Once that aspect is cleared up, how will this new turn of
> >events affect the work of Ian Niles, Matthew West, and Robert Kent?
>
> I imagine that its prime purpose was to render harmless to Cycorp the
> SUO and all similar efforts, insofar as they might be seen as in any
> way competitive. Notice that the agreement requires that any use of
> any axioms which are *logically equivalent* to any part of Cyc, even
> if translated into any other natural or artificial language, must be
> credited to Cycorp. This means that if one uses any part of the Cyc
> upper ontology and if they also assert that, say, every relation is a
> thing, or that spatial containment is transitive, then they must
> acknowledge that Cycorp owns the copyright to that assertion, even if
> it is made in, say, EPISTLE, or even in English, and even if they had
> said it before in other places; in fact, even if Aristotle had said
> it before. The hubris of claiming ownership of all statements which
> are logically equivalent to a fair amount of human common sense is
> quite astonishing.
>
> Pat Hayes
>
> ---------------------------------------------------------------------
> IHMC (850)434 8903 home
> 40 South Alcaniz St. (850)202 4416 office
> Pensacola, FL 32501 (850)202 4440 fax
> phayes@ai.uwf.edu
> http://www.coginst.uwf.edu/~phayes
>
>