Re: AWWSW Telecon Tuesday 2011-08-30 ?

On Wed, Sep 28, 2011 at 11:06 AM, Henry S. Thompson <ht@inf.ed.ac.uk> wrote:
> Jonathan Rees writes:
>
>> And I haven't gotten much feedback on
>> http://www.w3.org/2001/tag/awwsw/ir/latest/ , which to me pretty
>> much answers to my satisfaction the question that created this group
>> (at the HCLS/TAG meeting way back when): what is this 'information
>> resource' deal and what does it have to do with the price of eggs?
>
> Note that some recent comments from DBooth on this list have referred
> to an earlier version of this document [1], but I'm explicitly
> addressing the version dated 2011-06-25 [2].
>
> I know some participants on this list don't like it much, but I really
> think you need to reference FRBR in section 2.  I don't mean that you
> have to _define_ information entity using the FRBR vocabulary, but you
> should point to FRBR as a much more complete working through of the
> relationship between various points on a specific/generic hierarchy
> for several well-known and widely/naively understood entity classes
> which overlap with jar:InformationEntity ( ==
> frbr:ProductOfIntellectualOrArtisticEndeavour ?*)

Hmm... I wrote http://odontomachus.wordpress.com/2011/02/13/frbr-and-the-web/
although I'm not sure I believe it any more (in particular there are real things
corresponding to Expressions and Works - but they are events or
processes, not things)...

I like FRBR, but my take is that the connection to web-things isn't
strong. Physical objects (Items) don't figure in web discussions -
they are patterns of magnetization or charge on disks.
wa:Representations are sort of like Manifestations, except that they
can be "published" (or originated) by multiple "printing presses"
(servers, mirrors, caches, etc.), so not sure it's a good match. And
obviously there are Manifestations that are not wa:Representations.
There may exist GenericResources having the property that their
wa:Representations are exactly the Manifestations (or potential
Manifestations ?? such a thing doesn't exist in FRBR) of some
Expression, and there may be other GenericResources having the
property that their wa:Representations are Manifestations of
Expressions of some particular Work; but I am not aware of a single
URI, anywhere on the web, having such properties.

Even where there is content negotiation by language (multiple
Expressions), the URI doesn't "name" a Work, because the domain owner
acts as a throttle and can decide which wa:Representations to
authorize - nobody has that power over a Work or Expression. The
wa:Representations you would get would only be a *subset* of the
Manifestations of the Expressions of the Work - if you made your own
translation, you'd get a new Expression (according to a librarian),
but its Manifestations would not be correct HTTP responses for the
domain owner's URI unless explicitly authorized.

FRBR could easily work in hypothetical new arrangements, such as a URN
namespace or particular http: URIs for which the domain owner
committed to an appropriate Work or Expression. But because of the
authorization arrangement what we see on the Web almost never maps
nicely onto FRBR.

If you think this discussion helps, rather than detracts, I could
include it, but as the results are all negative I would think it's
distracting. Maybe as an appendix?

> I also think in this regard you need to think harder about being a
> _bit_ more explicit/thorough in defining *information entity*, whether
> specific or generic.  Yes, the poem example and the tapir analogy
> point to something that feels familiar, but it leaves the door open to
> a slippery slope (!) at the bottom of which lies 3986-representation,
> doesn't it?  Does it offer a clear answer to e.g. the question of
> whether these two pages are specialisations of the same generic
> information entity:
>
>  http://www.ltg.ed.ac.uk/~ht/felisCatus/cattus.html
>  http://www.ltg.ed.ac.uk/~ht/felisCatus/chat.html
>
> given the results of fetching
>
>  http://www.ltg.ed.ac.uk/~ht/felisCatus/
>
> ?
>
> I don't think so.

A page (or wa:Representation) can specialize *many* generic resources,
as Tim describes in his 1996 note. And in my theory, information
entities are 1-1 with arbitrary sets of wa:Representations, so given
any collection of wa:Representations you can form the IE that has
those wa:Representations and no others.

Ontologically, as Alan keeps reminding me, the idea is rather
incoherent, so that's why the given semantics is axiomatic, not
ontological. The only way to do something more intuitive and
principled, in my opinion, would be to exclude most, if not all,
actual situations one finds on the web. That would not be helpful.

> ht
>
> * I.e. frbr:Work U frbr:Expression U frbr:Manifestation U frbr:Item
>
> Which raises an interesting question -- are e.g. the Mona Lisa, the
> David, even the Eiffel Tower, information entities?  It's hard to see
> how http://smarthistory.org/assets/images/images/leo_mona_face.jpg is
> _not_ a specialisation of the Mona Lisa if you accept that
> http://www.edwardtufte.com/tufte/graphics/poster_OrigMinard.gif is a
> specialisation of Minard's graphic of Napolean's retreat from Moscow,
> but that painting certainly has e.g. mass, which would disqualify it
> on TimBL's definition as I understand it. . .  Hmm, so does
> the original of Minard's graphic.  But not all drawings have paper
> originals -- presumably you would definitely want to acknowledge that
> there's an information entity somewhere in the picture (:-) in
> connection with
> http://www.w3.org/2001/tag/awwsw/ir/latest/generic.png . . .

Interesting, but not sure the question needs to be answered.  I'd
start with has mass => not information entity.

> Sorry, can't resist -- I suppose you might want to argue that there is
> an information entity which is a _photograph_ of the Mona Lisa, or a
> _scan_ of Minard's graphic, and that it is specialisations of _those_
> which the above URIs allow us to retrieve.  This gets us back to
> massless entities.  However I don't think FRBR demands a difference of
> Work in cases like this, or even of Expression.  A photocopy of my
> copy of Martin Chuzzlewit is, I guess, a Manifestation, and a
> photograph or a scan (modulo intent, I suppose) could be considered
> the same. . .

In my reading Items have mass, none of the others do. A piece of paper
would be an Item, something whose identity is preserved by copying
would be a Manifestation... The upper three levels might have
duration, as they are [tied to] processes (Work =
inspiration/creation, Expression = perspiration/realization,
Manifestion = publication/reproduction). But I think this is off
topic....

> [1] http://www.w3.org/2001/tag/awwsw/ir/20110517/
> [2] http://www.w3.org/2001/tag/awwsw/ir/20110625/
> --
>       Henry S. Thompson, School of Informatics, University of Edinburgh
>      10 Crichton Street, Edinburgh EH8 9AB, SCOTLAND -- (44) 131 650-4440
>                Fax: (44) 131 651-1426, e-mail: ht@inf.ed.ac.uk
>                       URL: http://www.ltg.ed.ac.uk/~ht/
>  [mail from me _always_ has a .sig like this -- mail without it is forged spam]
>

Received on Wednesday, 28 September 2011 15:42:37 UTC