chat-transcript_20100909c_edited.txt // PeterYim: . . Welcome to the Joint OOR-Ontolog-NCBO-CC-IAOA-OASIS "OpenOntologyRepository_IPR Policy and Issues" Panel Discussion (session-1) - Thu 9-Sep-2010 * Topic: "OOR-IPR session 1: An Exposition on Relevant IPR Regimes" * Chair: Mr. PeterYim (Co-convener, Ontolog & OOR; Secretary, IAOA) * Keynote Speaker: Dr. GeorgeStrawn (Director of the (US) National Coordination Office for Networking and Information Technology R&D) - "IP and IT" * Panelists: o Mr. JamesBryceClark, JD (General Counsel, OASIS) - "Some Notes on OOR IPR" o Mr. JohnWilbanks (VP of Science, Creative Commons) - "licensing and ontologies: research from creative commons" o Mr. BrucePerens (original author of the "Open Source Definition") - "OOR Open Source Software Licensing" Please refer to dial-in and other details (include slides download links) on the session page at: http://ontolog.cim3.net/cgi-bin/wiki.pl?ConferenceCall_2010_09_09 . JamieClark: Pinging, just to establish that I'm in the chat room. I also have visual into the VNC desktop. Personally, as a speaker, I plan to ask Peter to advance my slides, and not to bother with manipulating the remote session at my end. Cheers all PeterYim: Hi JamieClark ... I actually do get to advance the slides for everyone! anonymous morphed into SteveRay anonymous1 morphed into JohnWilbanks anonymous2 morphed into BrucePerens BrucePerens: Testing. anonymous1 morphed into JoelBender anonymous1 morphed into CameronRoss JamieClark: JamieClark's here (on voice) anonymous2 morphed into YuriyMilov anonymous morphed into BobbinTeegarden anonymous1 morphed into MikeBennett PeterYim: Dr. GeorgeStrawn speaking ... JamieClark: Thanks, George: excellent remarks and we appreciate the US government's support for developing this concept. RaviSharma: Dr. George Strawn - Many thought provoking and positive areas covering S&T, Knowledge and Information and ontology / OOR carving its space in the open / global IPR space. These valuabe inputs will nurture the purpose of this series and spearhead the focus of deliberations. many thanks. anonymous morphed into DougFoxvog PeterYim: Mr. JamesBryceClark, JD speaking (started 11:02am PDT) ... JamieClark: JBC slide 1-2: Topic introduction. JamieClark: JBC slide 3: Definition of "open" largely is a defined community's policy chat about its requirements. Consider: even a variety among standards projects: OASIS, IETF, W3C, ISO, ITU ... maybe microformats.org? ... JamieClark: JBC slide 4: Some say that some restrictions are *virtuous*. BrucePerens: You could do that with trademark rather than copyright and it would still work. MikeBennett: OASIS "Keeps Copyright" in their standards - doesn't (C) exist automatically? Presume means "Asserts Copyright"? BrucePerens: Mike, this means that OASIS doesn't abandon its copyright and dedicate the document to the public domain. MikeBennett: Thanks Bruce. JamieClark: JBC slide 5: License form matters too. Key issue for real user: "safe to ignore" JamieClark: JBC slide 6: If you can't get at material, its theoretical legal availability is sort of irrelevant. ... Reliance on open data sources = sunk costs. JamieClark: JBC slide 7: As for legal uncertainty, let's plan for the worst. Note, a LATER session in this OOR-OPR series will discuss what best polices to use. JamieClark: JBC slide 8: How one stable host organization (OASIS) handles the policy issues. JamieClark: JBC slide 9: Look at the *workflow* of rights and contributions as well. It's not a static snapshot. JamieClark: JBC slide 10: Using a federated system introduces some additional issues. TerryLongstreth: JBC Slide 9: I'm unsure of how fine a distinction can be drawn between 'read' and 'use' permission. Is there a restriction on what I, ('public') can do with materials others submitted with open access? JamieClark: JBC slide 11: Using a rating / opinions system introduces some additional issues. JamieClark: JBC slide 12: Thanks. Call anytime with questions. JamieClark: --- PeterYim: Mr. JohnWilbanks speaking (started 11:27am PDT) ... JohnWilbanks: see - http://creativecommons.org/about/licenses/ RaviSharma: @JamieClark - lifecycle related - is it a workable proposition that even if some organization makes some comercial value add on what is largely open source that it automatically turns to open status in a shorter period say 3-5 years for all such proprietary add-ons to become public. BrucePerens: RaviSharma, is this question regarding standards IP submissions or something else? RaviSharma: Bruce - I was thinking of OOR primarily, but really it would also apply to Knowledge-bases. Standards are facilitators for connecting, validating and communication of the knowledge pieces and bases. MikeBennett: @Ravi that does raise an interesting question, whether there are the same IPR considerations for class-level ontologies versus instances. BrucePerens: RaviSharma - For reasons that have just been discussed I am far from sure you can make any requirements stick upon the creators of commercial derivative works. JamieClark: Agree with Bruce. Don't rely heavily on terms that probably don't stick; and we know from experience that some don't. JamieClark: Reading the Dental Association quotes. Assertions of creativity in classification sound a LOT like the 'Feist' and 'West v. Mead Data' assertions that phone directory listings and caselaw page numbers were creative ... Saying this not as a copyright lawyer in 2010, but because John is giving me some flashbacks from 1987. (As a frisky young judicial law clerk, I was involved in the West vs. Mead case.) RaviSharma: Mike - if class-level is open, others can use it as pattern for creating instances but such creator might change status to non-open but it would become open based on original open-use license declaration. BrucePerens: There is substantial doubt that a reference to an open ontology caused the referencing ontology to be a derivative work of the open one. JohnWilbanks: Bruce - agreed. But copying ontologies is an essential part of ontology practice, and making changes to the local copies is as well... MikeBennett: @Ravi agreed. That would be a significant use case for having an ontology in the first place, hence the importance of distinguishing T-Box v A-Box in thinking about these IPR issues I think. CameronRoss: John - For clarity, is the CC BY license amiable to using an ontology within a commercial application? JohnWilbanks: Cameron - yes, CC BY is completely compatible with commercial use PeterYim: Mr. BrucePerens speaking (started 11:47) ... PatrickCassidy: @John Wilbanks: concerning the question of whether an ontology is software: at least the ontologies that are produced in a format that has associated reasoners (such as OWL or FOL format) should be considered as a form of software, since they are in fact instructions about how to perform inferences from data. In fact, I think of logic-based ontologies as little more than software in a specific declarative format, easier to understand and easier to modify than traditional code. JohnWilbanks: Patrick - this is a fundamental question. RaviSharma: Bruce and JamieClark - can we not declare on open source material that any use based on this material is only non-open for x years? JohnWilbanks: my instinct is that the courts will be utterly confused by the ontology question JamieClark: John (a W3C alumni) mentions the W3C patent policy as a possible prototype. For the most part, the rules and workflow of the W3C, OASIS and IETF policies on patent handling are pretty similar. So I suspect the models would all lead to a similar design. didn't quote W3C's instance in this case, as has a deeply-negotiated rule (it's section 7) that royalties are permitted, but must be studied and made the sole personal decision of an individual. (Sort of like the IANA under Jon Postel in 1996, versus ICANN today.) Here's the link to the W3C policy: http://www.w3.org/Consortium/Patent-Policy-20040205/#sec-PAG-conclude JohnWilbanks: To JamieClark's point, yes - the technical standards world has worked out patent policies reasonably well, and OASIS and IETF are both excellent models as well JohnWilbanks: I used the W3C one because I know it better, no other reason! RaviSharma: Bruce - was that Katz case? IT related? BrucePerens: Jacobsen v. Katzer. Look in Wikipedia MikeBennett: @Patrick but many use an ontology as a technology-neutral, formal business definition of subject matter. It so happens you can then do those things with it if it's in OWL or similar. JamieClark: @JohnWilbanks, I think our tribe figured out, a while ago, that the delta between the established practices is quite small, except in marketing brochures :D JohnWilbanks: JamieClark, heh. True. I can send you some lovely corporate marketing documents for closed ontologies. CameronRoss: John: Would CC BY be considered "Gift" licensing? JohnWilbanks: Cameron, I believe so - given that Bruce classifies BSD in there, and that CC BY is an analog to BSD. PatrickCassidy: @Mike - if an ontology is fairly simple and doesn't include restrictions or other structures that enable inference more complicated than simple taxonomic subsumption, it does in fact contain instructions for computation that can be automatically executed. This is the function of software, regardless of whether there are lots of documentation that is also useful independent of the instructions. MikeBennett: @Patrick yes but I don't think you can necessarily determine intention from inspection. PatrickCassidy: @MIke-oops! if its simple, it *isn't* necessarily a program, but if is more complicated, it is hard to distinguish from software. PeterYim: @Bruce - slide#8 ... can you give a quick rundown on how people would circumvent GPL please? BrucePerens: It's a long talk. Can we do it another time? JamieClark: @Ravi - The idea of a sunset rule on restrictions is a darn interesting one. If compromise is needed. But there's a strong opendata movement these days; as a negotiating matter in the service of openness, as a personal opinion, I'd rather start from "open", as the default, and save ideas like "open soon" as middle-ground options only when needed. TimBL in the UK, OSTP in the US, and others are creating some great social pressure to open up data right now. (If you think about userfriendly simple rights, on an icon-based, no-lawyer-need basis like CC, how would we happily represent "not open now, but it will be later"? Feh.) JohnWilbanks: @Ravi, there is some sunset stuff in the US National Institute of Health's open access repository policy, but I tend to agree with @JamieClark - let's go for open first. Default rules of open, with opt outs clearly defined and accessible, are a good starting point. PeterYim: @JohnWilbanks - is there a (cc) license (for content) that is similar/analogous to LPGL (for software)? JohnWilbanks: @Peter, we only provide a single share-alike license JohnWilbanks: although we do provide layperson readable and metadata versions of GPL, LGPL, and BSD licenses PeterYim: thanks, John JohnWilbanks: @BrucePerens, the core right is the right to make non-commercial copies RaviSharma: presenters: Ontologies are not patentable - is this defendable position as connected ontologies might be open or common relationships are open and thus not patentable? Similar argument for proprietary ontologies? JohnWilbanks: (all CC licenses carry this right, not just the right to read) JohnWilbanks: thanks @BrucePerens for noting my comment JamieClark: Bruce asserts: part of proper host/repository/community management may be to *permit* forking, for virtuous quality & contribution reasons. JamieClark: @Ravi, the simple statement of problem is that "Eventually we will win this baseless lawsuit" is not equal to "This will be easy because we have the right to do it." JamieClark: SCO JohnWilbanks: apologies all, i have to jump off the call for a meeting JohnWilbanks: it has been an honor and pleasure, and thanks to @PeterYim JamieClark: Thanks John! great stuff JohnWilbanks: please buzz me at wilbanks@creativecommons.org if you want to talk more, and i'll be on the next call as well JohnWilbanks: @JamieClark - let's talk more! JohnWilbanks: (also, honored to be on a panel with @BrucePerens and George!) JohnWilbanks: bye all PeterYim: Bruce's slide#15 - @Wilbanks and @Perens - John, while your recommended against a reciprocal license, does Brunce's case of the MySQL dual licensing arrangement worth considering for (some portions of) OOR? ... the reason I am asking (and in line with GeorgeStrawn's "Open access is good, as is profit seeking") is that we need to make innovation sustainable! CameronRoss: @Peter It would be unlikely that I, for one, would release contributions under a dual license regime. JamieClark: BTW, here's an instance of FOSS folks (Apache) doing its legal diligence to make sure that the OASIS patent & copyright rules would permit OASIS standards (OpenDocument) to be used under the Apache licenses. http://www.softwarefreedom.org/resources/2006/OpenDocument.pdf (PDF link) This is an instance of the kind of "you better check with your lawyer" review that John and Bruce mentioned. If only all lawyers wrote as clearly as Eben Moglen ... JamieClark: --- PeterYim: Q&A and general discussion starts now ... (12:20pm PDT) ... GeorgeStrawn: "Are ontologies copyrightable or patentable now?" - panelists please discuss Bruce / Jamie: @George - the jury is still out, and these are questions that probably won't have a clear answer for a while yet RaviSharma: All- please see same question earlier where I feel these should not be easy or easily possible to patent. There is however a compelling counterargument for lifesaving pharmaceuticals that may use bioinformatics? CameronRoss: Would releasing an ontology under a Open Source or CC license, enforcible or not, constitute prior art for future patent claims? BrucePerens: @CameronRoss, any publication can constitute prior art. DougFoxvog: @CameronRoss: A public release of material should certainly (imho as a non-lawyer) constitute prior art. PeterYim: @anyone ... are we past the point of no return on the US Patent Reform to go from "first-to-invent" to a "first-to-file" regime? BrucePerens: Peter, no but our main enemy on this entire argument is pharma. RaviSharma: All- I am reminded of "standing on Shoulders of giants" and desire for knowledge in open regime. Can the community support of this type of assertion help keep ontologies open sourced? CameronRoss: So, could we not employ an Open Source or CC licensing strategy to protect the OOR against future patents. Of course there's still the issue of existing patents. DougFoxvog: One ontology can be expressed in multiple semantic languages. Copyright couldn't cover the same expression in both OWL and CycL, say. CameronRoss: @DougFoxvog - So can one circumvent copyright using automated translation? DougFoxvog: @Cameron: good question. The information can not be copyrighted, although a specific expression of it in a specific language may be. MikeBennett: @CameronRoss copyright depends on being able to demonstrate something was copied rather than prior art (as patents are) so transformation if detectable would fall foul of copyright. CameronRoss: @MikeBennett - That's what I thought. Thanks. TerryLongstreth: So, we need an implementation neutral form of expression for the intellectual content of an ontology? BrucePerens: USC 17.1.102.b says: In no case does copyright protection for an original work of authorship extend to any idea, procedure, process, system, method of operation, concept, principle, or discovery, regardless of the form in which it is described, explained, illustrated, or embodied in such work. In ADA v. Delta Dental (at http://lw.bna.com/lw/19971014/964140.htm) the appeals court judge found that there is sufficient creative expression in a taxonomy to make it copyrightable. A quote from http://www.invispress.com/law/copyright/dental.html is educational: The ADA couldn't stop dentists from using the ADA Code in their forms and records, or stop Delta from distributing forms that invited dentists to use the ADA's Code. However, the Court found that it was a violation of the ADA's copyright for Delta to distribute a copy of the code itself. The degree to which this applies to ontology, and indeed the value of the case as precedent, is debatable. There is a discussion at http://www.benedict.com/Digital/Software/Ontology.aspx that concludes that ontologies that refer to other ontologies are not derivative works of those other ontologies under copyright law. Ontologies might be the subject of patent rather than copyright to the extent that they can be restricted effectively, and patenting one presents its own problems. JamieClark: Thanks for an excellent session, it's an honor to work with Peter, George, Bruce, John & this group. JoelBender: Thank you! MikeBennett: Thanks all! Most helpful session PeterYim: thank you all ... great session! MikeDean: Thanks for a great session PeterYim: -- session ended: 12:47pm PDT -- //