Hoi, I don't think that n-ary Relations are not that evil as meant to be. If there is the Factbox it could display a Sign that marks incomming Relations. If you click on this sign, you will go to the page where the Statement was written. Well, this means that you may need to edit multiple pages instead of one, but I think that this is not really bad, but uncomfortablly.
A real problem could be security, becaue a spammer could make statements about a page that is already locked in another page, which would mean that tha Spam is shown on any page, nverless if its locked or not. This would be a bad thing. Anyway, I think it should be possible to make n-ary Relations about Subsections of articles. Ie the statement [[Atom#Mass :: is type of :: Mass (Physics)]] or short [[#Mass :: is type of :: Mass (Physics)]] might get allowed as long as it is written within the Article [[Atom]]. In this case it might not be needed to write a Header or <div> Element called so, the Identifier should be able to coose virtually. I think this technique would solve the problems that the n-ary guys are trying to solve, while keeping 1-ary security. As far as I remember, Markus already wrote that he will keep that in mind for a future release, but not for SMW v1.0. Sincerly, MovGP0 Ittay Dror wrote: > i read your summary of n-ary relations from Apr. 15th. It seems that the key > difference is that in this discussion people are asking not for a way to have > complex relations, but for a way to set a relation by means of explicitly > stating the subject, rather than having semantic media wiki infer it from the > current page. > > > ittay > > > Yaron Koren wrote on 05/07/07 03:48: > > > Well, let me rephrase what I said before, because it was misleading: > > the system already allows conflicts. You can say, in a page for Idaho, > > that it has a certain population, and then, later on in the same page, > > that it has a different population. The system can't check against > > that because there's no way to know that that's a conflict: sometimes, > > you want to have more than one value for a single attribute or > > relation. "Has state nickname", for instance, will sometimes be called > > more than once for a state. > > > > So, in the current system, conflicts are allowed to occur; but at > > least they're easy to spot and fix because you know exactly where to > > find each one. > > > > You could make the argument that my whole problem with arbitrary > > semantic triples is just about incorrect data; once the data is fixed > > and the conflicts are resolved, no problem. But actually it's worse > > than that. Take the example that started this whole thread, of a > > character named "Weems" in a book, who doesn't have his own page. Now, > > in this book he's a "mercenary". But what if the book has a sequel, > > also featuring Weems, but in this one he's been promoted and he's now > > a "wizard"? The second book's page has a triple stating that he's a > > wizard, so now you have two contradictory pieces of information in the > > wiki, even though both are accurate in the context in which they > > appear. > > > > n-ary relations would solve this problem, by placing all information > > about Weems in the context of whatever book he was part of. So the > > information about his occupation wouldn't be an attribute of "Weems", > > it would be an attribute of "the character who appears in > > such-and-such book, who has the name Weems" (to paraphrase). > > > > -Yaron > > > > > > On 5/6/07, Ittay Dror <[EMAIL PROTECTED]> wrote: > >> > >> > >> > >> can't you discover conflicts during the save of the page. reject the save > >> if > >> a conflicting triplet exists somewhere else? > >> > >> Yaron Koren wrote on 05/06/07 15:36: > >> > >> > >> Well, let me start by saying that I still disagree with the idea of > >> place-anywhere semantic triples, for the reason I expressed earlier: > >> you can easily have information conflicts. I couldn't follow > >> everything that was being said before, but I think most of the > >> discussion was about being able to *display* and *edit* the > >> information from a lot of different pages, not *store* it in different > >> pages. > >> > >> That said, the example you give now is much more reasonable than the > >> example you had before (where data about a U.S. state was placed in > >> the page of someone who lives there). In fact, it's a perfectly > >> reasonable example: sometimes giving sub-data their own pages is too > >> much work. The solution for this is not arbitrary triples but n-ary > >> relations, which define a sub-component, itself composed of triples, > >> and then a relation between the main page and the sub-component. > >> > >> n-ary relations aren't yet supported by Semantic MediaWiki, but they > >> could be and I really think they should be; see the email thread from > >> April 15 to see my summary of the discussion, and thoughts on the > >> issue. > >> > >> I'm curious to hear what you, or anyone else, has to say about it. > >> > >> -Y > >> > >> > >> On 5/4/07, Kelly Jones <[EMAIL PROTECTED]> wrote: > >> > >> > >> There are a lot of semantic wikis out there, but I really like > >> Mediawiki and want a semantic wiki that has all of its features. > >> > >> I've tried Semantic Mediawiki, but dislike its one major limitation: > >> on page X, you can only create triples starting with X. > >> > >> I'm sure this works well when importing semantic data into a wiki, but > >> doesn't work so well when you're creating a wiki from scratch. > >> > >> Example: for each book in a series, I'm creating a book page, and > >> listing characters in the book: > >> > >> * [[Galder Weatherwax]], male human [[wizard]] > >> * [[Buggy Swires]], male [[gnome]] > >> * [[Lackjaw]], dwarf jeweler > >> * [[Weems]], male human [[mercenary]] born in [[Morpork]] > >> > >> I'd like to semantically annotate this data, example: > >> > >> Weems appears_in (bookname) > >> Weems birthplace Morpork > >> Weems race human > >> Weems gender male > >> Weems occupation mercenary > >> > >> Semantic Mediawiki requires me to create a separate page for each > >> character, which quickly becomes a depth-first nightmare. > >> > >> Is there a Mediawiki extension that will let me do this? I know there > >> are arguments against this sort of thing (eg, data consistency), but I > >> think the benefit outweighs the risk. > >> > >> I'm tempted to use Mediawiki hooks to write something myself, but > >> wanted to check for existing extensions first. > >> > >> -- > >> We're just a Bunch Of Regular Guys, a collective group that's trying > >> to understand and assimilate technology. We feel that resistance to > >> new ideas and technology is unwise and ultimately futile. > >> > >> ------------------------------------------------------------------------- > >> This SF.net email is sponsored by DB2 Express > >> Download DB2 Express C - the FREE version of DB2 express and take > >> control of your XML. No limits. Just data. Click to get it now. > >> http://sourceforge.net/powerbar/db2/ > >> _______________________________________________ > >> Semediawiki-user mailing list > >> Semediawiki-user@lists.sourceforge.net > >> https://lists.sourceforge.net/lists/listinfo/semediawiki-user > >> > >> > >> ------------------------------------------------------------------------- > >> This SF.net email is sponsored by DB2 Express > >> Download DB2 Express C - the FREE version of DB2 express and take > >> control of your XML. No limits. Just data. Click to get it now. > >> http://sourceforge.net/powerbar/db2/ > >> _______________________________________________ > >> Semediawiki-user mailing list > >> Semediawiki-user@lists.sourceforge.net > >> https://lists.sourceforge.net/lists/listinfo/semediawiki-user > >> > >> > >> > >> > >> -- > >> Ittay Dror > >> Chief Architect, > >> R&D, Qlusters Inc. > >> Web: qlusters.com > >> Email: [EMAIL PROTECTED] > >> Phone: +972-3-6081994 > >> > >> openQRM - Data Center Provisioning > >> ------------------------------------------------------------------------- > >> This SF.net email is sponsored by DB2 Express > >> Download DB2 Express C - the FREE version of DB2 express and take > >> control of your XML. No limits. Just data. Click to get it now. > >> http://sourceforge.net/powerbar/db2/ > >> _______________________________________________ > >> Semediawiki-user mailing list > >> Semediawiki-user@lists.sourceforge.net > >> https://lists.sourceforge.net/lists/listinfo/semediawiki-user > >> > >> > > > > -- > *Ittay Dror* > Chief Architect, > R&D, Qlusters Inc. > Web: qlusters.com <http://www.qlusters.com/> > Email: [EMAIL PROTECTED] <mailto:[EMAIL PROTECTED]> > Phone: +972-3-6081994 > > openQRM <www.openqrm.org> - Data Center Provisioning > > ------------------------------------------------------------------------ > > ------------------------------------------------------------------------- > This SF.net email is sponsored by DB2 Express > Download DB2 Express C - the FREE version of DB2 express and take > control of your XML. No limits. Just data. Click to get it now. > http://sourceforge.net/powerbar/db2/ > ------------------------------------------------------------------------ > > _______________________________________________ > Semediawiki-user mailing list > Semediawiki-user@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/semediawiki-user > ------------------------------------------------------------------------- This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ _______________________________________________ Semediawiki-user mailing list Semediawiki-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/semediawiki-user
