Re: Harvesting from Roy's paper

Jonathan Rees wrote:
> On Mon, Mar 7, 2011 at 10:40 AM, Nathan <nathan@webr3.org> wrote:
>> Jonathan Rees wrote:
>>> On Mon, Mar 7, 2011 at 9:31 AM, Nathan <nathan@webr3.org> wrote:
>>>> Jonathan Rees wrote:
>>>>> Blog post including this material:
>>>>>
>>>>>
>>>>> http://odontomachus.wordpress.com/2011/03/07/are-you-confused-yet-about-the-word-representation/
>>>> :) the last one made me chuckle a bit!
>>>>
>>>> definitions I that "sit right" with me:
>>> It's not a question of "sit right" generally, but specifically for
>>> some articulated purpose. I have no trouble at all with an art
>>> historian saying that a painting is a representation of a saint.
>>> That's just a different sense of the word. The "hateful" thing is when
>>> one person takes a statement made assuming one sense, then
>>> reinterprets the statement with another sense for propaganda reasons.
>>> For example, AWWW might be taken as using "representation" in Roy's
>>> sense when actually it's using it in Tim's sense (although I admit you
>>> have to read between the lines to infer that - really it may be
>>> gingerly using it  in *neither* sense since the argument wasn't
>>> resolved at the time of publication).
>>>
>>> Both Roy and Tim would have done better to coin new words. I'm
>>> experimenting with 'specialization' for Tim's sense, although even
>>> that would be more a term of art than an adaptation of a common-sense
>>> meaning.
>> Can you just clarify, regardless of what word is suggested/adopted, what the
>> meaning of it would be? in-line with TimBL's and Niklaus Wirth's definition?
>> to be used in relation to "information resource" and httpRange-14?
> 
> I thought I answered this already... but let me try again...

you did, but this is nice and terse and clear :)

> If the relation is R, then its operational meaning would be: If you
> see a statement IR P O, then you can conclude a statement CM P O,
> whenever CM R IR and P+O is in an approved set; and if you know (or
> are willing to believe) CM P O for all CM with CM R IR, then you can
> conclude that IR P O, again assuming P+O is in the approved set.
> 
> I am very happy to haggle over what's in the approved set; a candidate
> list is in the latest version of the ir-axioms note.
> 
> Sorry if this seems unsatisfying, but in four years it's the only
> thing I've come up with that supports metadata 'curation' of the kind
> Tim and I want to encourage. There are many ways to model this
> ontologically - IRs are generic individuals, of which CMs are
> specializations, and so on - but every such attempt seems to be
> inadequate and we always end up with Ruttenberg's Scylla.
> 
> Jonathan
> 
>

Received on Monday, 7 March 2011 17:29:45 UTC