- From: Karen Coyle <kcoyle@kcoyle.net>
- Date: Tue, 15 Oct 2013 11:00:07 -0700
- To: Dan Scott <denials@gmail.com>
- CC: "public-schemabibex@w3.org" <public-schemabibex@w3.org>
Dan, your argument makes sense, however in actual libraries inventory is done with barcodes - that is, that is how libraries count what they have. And inventory # has to be 1:1 with things owned. So maybe the issue is that we don't want to use the term "inventory identifier" for call numbers because it will confuse those who use the barcode to do their inventory. This means that we are still lacking a term for the call number/shelf number. Part of the complication is that the shelf number has a locating function, but the location is relative, not fixed. Another part of the complication is that it's not just a location, it's an indication of the subject matter. I think getting the idea of location into the name or the definition would be helpful. Lacking that, bringing out the classification aspect might speak to potential users. kc On 10/15/13 10:34 AM, Dan Scott wrote: > On Tue, Oct 15, 2013 at 1:14 PM, Karen Coyle <kcoyle@kcoyle.net> wrote: >> Thanks, Richard. Personally, I would switch inventoryIdentifier and >> serialNumber -- the barcode on the book is the inventory identifier. Serial >> number works just as well for either, so it could be the call number. > > I still think that's the wrong way around. This is not "serial number > as in ISSN", but "serial number as in uniquely identifies a single > item". barcode is a much, much better fit for schema.org/serialNumber > in my opinion, as while we have established that some libraries use > the same call number for multiple copies of a given item, I don't > think there are any libraries that use the same barcode more than > once. > > http://en.wikipedia.org/wiki/Serial_number says "A serial number (also > manufacturer's serial number or MSN) is a unique code assigned for > identification of a single unit. Although usually called a number, it > may include letters, though ending with digits. Typically serial > numbers of a production run are incremented by one, or another fixed > difference, from one unit to the next." That last bit also sounds an > awful lot like how barcodes are typically generated, and not at all > how call numbers are assigned (accession numbers, sure, but that's a > different beast). > > Richard, do you have a proposed definition for > schema.org/inventoryIdentifier? I'm keen on finding out how it differs > substantially from schema.org/sku. http://en.wikipedia.org/wiki/Sku > uses the definition: "a unique identifier for each distinct product > and service that can be purchased in business"; that's pretty close to > what I would think of as an inventory identifier. If we're going to > argue for the addition of a new property, it's going to need to be > convincingly different! > >> Other than that, I think this is good to go, but we never got a definitive >> answer about de-commercializing the definitions, did we? However, we also >> got only positive responses, as I recall. > > Yes, there seems to be a limited attention span on public-vocabs and I > think most of that attention recently has been gobbled up by SKOS and > to a lesser extent the accessibility proposal... but like you I don't > recall any opposition to the notion. I wouldn't be surprised if > schema.org 1.0d was released and the changes were just there! > -- Karen Coyle kcoyle@kcoyle.net http://kcoyle.net m: 1-510-435-8234 skype: kcoylenet
Received on Tuesday, 15 October 2013 18:00:37 UTC