 |
 |
I am at a loss to understand the analogy of swiss army knife versus a ratchetting screwdriver as presented in Ricky's message. The most important reason to consider the proposed structural changes in DC is *precisely* to make the Dublin Core more useful for the most people in the most disciplines across various sectors... public/private/commercial/governmental. This can be done without compromising simple applications. It can be done in an extensible way that allows additional descriptive precision in a particular domain without compromising the general model. Will every application be interoperable in every detail with every other one? No. It is unlikely that a commercial rights metadata system will be able to use every detail of a library name authority record, or that a library application will be able to extract every bit of detail from an Interested Party name file. It *is* likely that a user could successfully search for a name in either variety of file... exactly the sort of semantic interoperability that we have been working towards from the very beginning of the Dublin Core. There are not two camps in the Dublin Core community. There is a continuum of needs and goals, as has been clear at least as far back as the Warwick meeting, and which surfaced in the Minimalist/Structuralist debate in Canberra. Ironically, the chances of accomodating the needs of the entire spectrum of applications is finally coming into reach. stu, who has always thought that it really should be called a Swiss Army Corkscrew. > -----Original Message----- > From: Ricky Erway [SMTP:[log in to unmask]] > Sent: Monday, December 28, 1998 8:15 PM > To: [log in to unmask] > Subject: Re: Camp 1 / Camp 2 and 1:1 confusion > > Stu said: > >If Camp 1 is about those who would restrict their deployment of > >metadata to embedded metadata, and will never extract such data into > >a seperate database, and expect no one else to harvest such metadata > >as surrogates, then I think it will be a small camp with limited > >prospects for interoperability. > > > >I do not believe that is what Ricky was describing but I hope she > >will respond with a clarification. > > I've been napping, but am back on board. Here's how I defined the two > camps: > > camp1) those who wish to use Dublin Core either for improved > discovery of web resources or as common access points across > varieties of web and non-web resources > > camp2) those who want to use Dublin Core as the basis for the > development of sophisticated systems for very specific descriptive > and functional requirements within a given domain > > among camp1 issues, I listed: > --collection/item and physical/digital distinctions > --undesirability of the 1:1 requirement > --ramifications of converted rather than created DC records > --means for searching across distinct resources > --camp1 interoperability across dissimilar camp2 implementations > > I think camp1 consists of BOTH those who would embed metadata in web > pages to enhance discovery AND those who would use it as "a commonly > understood set of descriptors that helps to unify other data content > standards" and that "is sufficiently flexible to represent resources > (and relationships among resources) that are both digital and exist > in traditional formats as well". > > And I believe the 1:1 "requirement" is moot in any camp. > > Ricky > > Camp2 carries a Right Angle Ratcheting Screwdriver whereas camp1 > carries a Swiss Army Knife, The Right Angle Ratcheting Screwdriver is > designed for a specific purpose -- one where the screwdriver on a > Swiss Army Knife would be worthless. But the Right Angle Ratcheting > Screwdriver is pretty worthless for filing your fingernails or > opening a bottle of wine... > > To: [log in to unmask] %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%