RE: Work-instance and isDerivativeOf

I agree with Niklas that alignment with not only bibliographic/library
contexts but with Schema.org and semantic architecture-based concepts should
be an essential part of our efforts here.  Isn't the goal of all of this to
enable the library community to have a presence in the linked data web
environment?  In that case, let's take a lesson from the enterprise world
where granularity of terminology may be important, but not at the expense of
timing and product launch.  Thus, our due diligence here will be beneficial
in the end but should not hinder progress.

...Will enjoy when all of this is compiled in an ontology.

Tom

Tom Adamich, MLS
President
Visiting Librarian Service
P.O. Box 932 
New Philadelphia, OH 44663
330-364-4410
vls@tusco.net


-----Original Message-----
From: Niklas Lindström [mailto:lindstream@gmail.com] 
Sent: Thursday, March 21, 2013 9:28 PM
To: kcoyle@kcoyle.net
Cc: public-schemabibex@w3.org
Subject: Re: Work-instance and isDerivativeOf

Hi Karen,

Indeed, those are important points (both the difference in primary meanings
and that no word's going to be perfect). Also in Swedish (my native tongue),
"realiserar" has the primary meaning of "making real".
I furthermore agree that derivative doesn't mean the same as instance.

I think the most important part is to strive for a term which is common in
as many bibliographic/library contexts as possible, and likely to endure.
Given the desired(?) movement from WEMI categorization towards the simpler
work-instance notion of BibFrame (whose basic components certainly seem more
fit for schema.org), how likely does it seem that this notion of "instance"
will prevail (at least longer than "manifestation")? I am fairly new to this
context (professionally), so I cannot judge the stability of these terms and
concepts.

I can imagine that "instanceOf" is general enough to work. (The schema:model
property does seem to be quite related, albeit belonging specifically to the
http://schema.org/Product class (which reasonably intersects with the
Instance notion, but is a much more specific class, not the least
culturally).)

Should the property work on multiple levels – i.e. can you have instances of
instances? That might make it easier to maintain a "work/expression"
difference, if that is desired; and to relate "manifestations" to "items".
Also, it should reasonably not be applicable on the same level, right? That
is, where relations like derivation, adaptation or translation (and
commonEndeavor) applies.

An alternative I kind of like is "embodies", as in "to give a tangible,
bodily, or concrete form" (see e.g. [1], [2]).

But among the presented alternatives, I suppose that "instanceOf" has the
least variations in this context, at least connotationally.

Cheers,
Niklas

[1]: http://en.wiktionary.org/wiki/embody
[2]: http://www.thefreedictionary.com/embody


On Thu, Mar 21, 2013 at 10:11 PM, Karen Coyle <kcoyle@kcoyle.net> wrote:
> It occurred to me after the call that in some languages the concept of 
> "realization of" == frbr:Manifestation. It means to bring something 
> into existence, thus, make it "manifest." Because the verb to realize"  
> has a different primary meaning in English (similar to "come to an 
> understanding
> of") it rarely gets used this way, but I think the meaning is what we 
> are aiming at.
>
> Larousee online has:
>
>  - Faire passer à l'état de réalité concrète ce qui n'était que virtualité
:
> Des désirs difficiles à réaliser.
>
>  - Concrétiser quelque chose : Cette maison réalise son rêve.
>
>
> My Italian dictionary has:
>
>  - realizzare: portare a compimento; concretizzare
>
> Thus the concepts are: bring to completion; and make concrete or 
> physically real.
>
> This is one of those times when English is just soooooo frustrating. 
> (Some of you may have more of these times that I do, I realize :-)!).
>
> In English, "derive" usually means to change something, to "adapt" it, 
> not to make an instance of it. That seems to be the sense in the EDM
vocabulary.
>
> That said, since no word is going to be perfect... I think we're back 
> to Instance unless folks think that we can sneak in "realizationOf". 
> Or maybe just switch to French when it's convenient. ;-)
>
> kc
>
>
> On 3/21/13 11:16 AM, Antoine Isaac wrote:
>>
>> Dear all,
>>
>> Following the discussion we had today on the work-instance 
>> relationship [1] and the proposal to label it with something like 
>> "derivative", I copy below some definition bits for the property 
>> isDerivativeOf from the Europeana Data Model [2].
>> Not that I want to push it by all means. Maybe you'll want a 
>> different definition. As long as it works... But I don't see why I'd 
>> conceal it, if there's something to be re-used instead of this group 
>> working hard to re-invent his own words.
>>
>> I'm going to ask the people in the task force I've mentioned if they 
>> can send museum or archive examples using this property.
>>
>> Best,
>>
>> Antoine
>>
>> [1] http://www.w3.org/community/schemabibex/wiki/Work-Instance
>> [2] http://pro.europeana.eu/edm-documentation
>>
>> ==========
>>
>> Definition: "This property [...] relates a resource to another one, 
>> obtained by reworking, reducing, expanding, parts or the whole 
>> contents of the former, and possibly adding some minor parts. 
>> Versions have an even narrower meaning, in that it requires common 
>> identity between the related resources. Translations, summaries, 
>> abstractions etc. do not qualify as versions, but do qualify as
derivatives."
>>
>> Obligation & Occurrence: A resource may be a derivative of to 0 to 
>> many resources. Conversely, a resource may have 0 to many resources 
>> that are derivative of its.
>>
>> Example: The Italian translation of Moby Dick is a derivation of the 
>> original work.
>>
>> Rationale: This property enables associating resources that are one 
>> the derivation of the other. This is required since Europeana may 
>> collect descriptions about resources and their derivations. It also 
>> supports browsing of resources by derivation. Finally, it allows the 
>> integration of all properties used in content providers' descriptions 
>> that capture the notion of derivation in the sense outlined above, 
>> such as those capturing versioning, translations and abstractions. To 
>> this end, any such properties should be declared to be a (direct or 
>> indirect) sub-property of edm:isDerivativeOf.
>>
>>
>
> --
> Karen Coyle
> kcoyle@kcoyle.net http://kcoyle.net
> ph: 1-510-540-7596
> m: 1-510-435-8234
> skype: kcoylenet
>

Received on Friday, 22 March 2013 10:45:11 UTC