------------------------------------------------------------------------ Date: Tue, 9 Mar 2004 16:40:05 -0500 (EST) From: Rebecca S. Guenther To: Andy Powell Cc: dih1@cornell.edu Subject: IMPORTANT: Roles ------------------------------------------------------------------------ Here are my comments on your markup and where I do and don't agree. We've had some previous email on the question of the narrowness of semantics in dc:contributor. Your answer concerning the value of core vs. some of these more specific roles that are outside core is well taken. But I still argue that what is a contributor is in the mind of the beholder; i.e. it depends upon what sort of resource you are describing. If in the legal field, the plaintiff may be a contributor if it is a transcript of the proceedings. I could argue that for almost everything on the list. Also, I didn't think you were consistent in your assessment (not that it was an easy job!). So here are those that I don't entirely agree with: Applicant: if the resource were an application for something, this would certainly be a contributor. Attributed name: "Use to relate an author, artist, etc. to a work for which there is or once was substantial authority for designating that person as author, creator, etc. of the work." This says it is possibly a contributer, although it is not certain, so should be a refinement of contributor. Auctioneer: what if the resource is a list of items to be auctioned; the auctioneer would have contributed. Bibliographic antecedent: "Use for the author responsible for a work upon which the work represented by the catalog record is based. This may be appropriate for adaptations, sequels, continuations, indexes, etc." The only reason I could see for you not considering this as contributor is the 1:1 principle, but I don't think you can control how people are describing these works. We already know that there is lots of controversy about that. Certainly the "bibliographic antecedent" played a role in the creation of the resource. Client: "a person or organization for whom another person of organization is acting" This is what is used e.g. for the owner of a house that an architect designed. I would certainly think the client would have input in that design so would consider a contributor. Complainant: "Use for the party who applies to the courts for redress, usually in an equity proceeding." I won't list all of the legal terms that are in this document, but for any of them, if the resource is the proceeding itself, I would consider them contributors to the content of the resource. You have excluded all of the legal terms (and there are lots of them). Contractor: Often the contractor does all the intellectual work. Copyright claimant: if the resource is a copyright registration, then this and copyright holder would be contributors. Dedicator: Use for the author of a dedication, which may be a formal statement or in epistolary or verse form. If some famous person wrote the dedication, I would bet you would want to consider him a contributor-- it is part of the intellectual content, although perhaps a small part. Degree grantor: wouldn't that institution have provided much that contributed to the intellectual content? Can you imagine all these theses and dissertation projects not being able to call the institution a contributor?? Distributor: should this be a subProperty of publisher? (or maybe we're not making these associations) Expert: "Use for a person in charge of the description and appraisal of the value of goods, particularly rare items, works of art, etc. " What if an art catalog of some sort? Under funder: The part highlighted as being incomplete I can remove-- it was part of a reference that should not be there. Monitor: "Use for a person or organization that supervises compliance with the contract and is responsible for the report and controls its distribution. Sometimes referred to as the grantee, or controlling agency." This you indicated was refinement of contributor, but why this and not others that don't have a direct contribution? Opponent: "Use for the person or corporate body responsible for opposing a thesis or dissertation." You have this one as contributor, but not, e.g. degree grantor. Other: this is our extension mechanism. I can understand why you wouldn't want it to be included. This does bring up to me how we will maintain this document if we start removing terms. But I guess that's our problem. Patent applicant and patent holder: same as copyright; what if the resource is a description of the patent? Patron: why this and not client? Seem to be similar roles. Publisher: should this be a subProperty of Contributor, since creator is? Witness: you've included this one as contributor, but none of the other legal terms (which I would argue should all be included). This exercise brings up the question: Are we creating a "core" list of roles? Previous discussion said no, we don't want to do that. Application profiles can do that. or Are we just saying that the term does or does not refine contributor? But then you have to think broadly what the possible resource could be which is being described. This exercise again says to me that maybe it isn't best to decide for others what they want to consider as contributors, since it depends on what they're describing. So where do we go from here? If you want to comment on those above, I'll just accept it and update the document. My argument for some not mentioned above was the need for access to the resource from the name whose role is the term from this list. You answered that by saying use the term as one from the marc namespace. I'll accept that, although I still question whether we should be doing this. It also means that for any new roles that we propose adding to our MARC list, we will have to bring it to the UB to see if it refines contributor, rather than just automatically adding? This will make maintenance a bit of a headache. Sorry for the length of this message and being such a pain. Copied to Diane so that she can comment. If we can come to some conclusion right away I can revise on Wed.-- leaving Thurs. night (I'll be in the office for most of Thurs.). Rebecca On Mon, 23 Feb 2004, Andy Powell wrote: > On Thu, 19 Feb 2004, Rebecca S. Guenther wrote: > > > If you do the first, will you also attempt to evaluate whether the given > > term is a refinement of dc:contributor? That seems to me what will take > > the most time. That approach is fine with me if that's what you meant. I > > would then see if I agree. In the meantime I might look at the list and > > independently try to assess which should be contributor and which not. > > My first attempt is at > > http://www.ukoln.ac.uk/metadata/dcmi/marcrel/codes.xml > > Note that it addition to removing subproperty of dc:contributor in quite a > few cases, I have also done the following > > - marcrel:cre --euqivalentproperty--> dc:creator > - marcrel:aut --euqivalentproperty--> dc:creator > - marcrel:pbl --euqivalentproperty--> dc:publisher > - marcrel:dpc --subpropertyof-------> dc:subject > > And I've corrected the URL for dc:contributor throughout! :-) > > I've checked this file still validates. > > Andy. > ------------------------------------------------------------------------ Date: Wed, 10 Mar 2004 21:28:46 -0000 From: Pete Johnston Subject: Re: IMPORTANT: Roles (fwd) To: DC-USAGE@JISCMAIL.AC.UK ------------------------------------------------------------------------ Rebecca, > Here are my comments on your markup and where I do and don't > agree. We've had some previous email on the question of the > narrowness of semantics in dc:contributor. Your answer > concerning the value of core vs. some of these more specific > roles that are outside core is well taken. But I still argue > that what is a contributor is in the mind of the beholder; > i.e. it depends upon what sort of resource you are > describing. If in the legal field, the plaintiff may be a > contributor if it is a transcript of the proceedings. I could > argue that for almost everything on the list. Also, I didn't > think you were consistent in your assessment (not that it was > an easy job!). So here are those that I don't entirely agree with: [snip] > This exercise again says to me that maybe it isn't best to > decide for others what they want to consider as contributors, > since it depends on what they're describing. Just to clarify.... by declaring a property example:role a sub-property of dc:contributor, we are saying that: If you state _x example:role _y then (in all cases) I can infer that _x dc:contributor _y so we should only make the sub-property assertion where that rule holds in all cases. Now, if example:role is _not_ declared as a sub-property of dc:contributor I can't draw the conclusion above. _However_, it doesn't stop you saying (where appropriate) that both _x example:role _y and _x dc:contributor _y i.e. it doesn't mean an agent who is a plaintiff can't also be a contributor; it just means I can't infer that all plaintiffs are contributors. Which is what we want, I think? (I think maybe it's a bit confusing to think of these properties as "roles": they describe relationships between a resource and an entity/agent, just as the dc:contributor property does.)