Patrick Logan <patrickdlo...@gmail.com> writes: > On Tuesday, May 28, 2013 2:51:51 AM UTC-7, Phillip Lord wrote: >> >> >> > >> > "Given a secret key and encrypted nonce for that key, assert the >> > unencrypted nonce." >> > >> > What I mean is that there is no way to express this in OWL alone. This >> > could be expressed in core.logic, in clojure, in java, in SPARQL + a >> custom >> > function, in SWRL + a custom function, in Prolog, etc. >> >> Perhaps I misunderstand the statements in the cryptography. I think he >> wanted to say "a person who has a secret key and the encrypted nonce for >> that key, also knows the unencrypted nonce". >> >> I think you could do that in OWL -- something like >> >> person (and (some knows secret_key) >> (some knows encrypted_nonce)) >> >> equivalent to... >> >> person (some knows unencrypted_nonce) >> >> Could be wrong -- I'm just a user of the logic, and a user of >> cryptography. >> > > You're correct about the desired goal. Looks like the actual decryption > does not need to be performed.
Yep, OWL is never going to achieve that. One of the things that I am sort of interested in with tawny is whether there is any value to the overlap of Clojure and OWL in the same syntax. It would be, for example, possible to annotate a Clojure function with the OWL; then, potentially, you could reason over the ontology, then use this to retrieve the executable function. Of course, this might be totally pointless or just daft. I haven't really thought about it enough to know whether there is any value in doing it. Probably not going to use cryptography as an domain! I did have some colleagues who were building a security ontology (and who do understand encryption!); I should look this up to see whether they had this sort of knowledge encoded already, and what we can reason. > > But the point stands that OWL does not define any inferences that will lead > to that desired OWL statement, given the original statements. > > A problem with saying these are equivalent in OWL is that "equivalent > class" is two-way. And so this implies that if you know the unencrypted > nonce then you also know the secret key and the encrypted nonce. Ah, okay, that's my lack of understanding of the domain. You can do this with a subclass relationship instead, which would make the implication unidirectional. >> > Yes, that's all I meant by the above, i.e. people unfamiliar with OWL >> > inferences would need to be aware of this and explicitly add statements >> > like "Alice and Bob are different individuals" or "the set of Male and >> > Female are disjoint", etc. >> >> Wrote a paper about male and female once; it's more complicated than you >> might think! >> > > Good point. One always has to make choices when choosing what to model... Indeed; pretty sure "male" and "female" will be more than enough for Alice and Bob. Phil -- -- You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegroups.com Note that posts from new members are moderated - please be patient with your first post. To unsubscribe from this group, send email to clojure+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/clojure?hl=en --- You received this message because you are subscribed to the Google Groups "Clojure" group. To unsubscribe from this group and stop receiving emails from it, send an email to clojure+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/groups/opt_out.
