W3C home > Mailing lists > Public > www-tag@w3.org > October 2011

Re: naive question: why prefer absolute URIs to # URIs for linked data?

From: Jonathan Rees <jar@creativecommons.org>
Date: Fri, 21 Oct 2011 07:33:48 -0400
Message-ID: <CACHXnar01PASPC0kDoHp_Ea0rDYsecN3-HaoKg6iaTFxMk8TsA@mail.gmail.com>
To: Kingsley Idehen <kidehen@openlinksw.com>
Cc: www-tag@w3.org
On Thu, Oct 20, 2011 at 6:47 PM, Kingsley Idehen <kidehen@openlinksw.com> wrote:
> On 10/20/11 5:26 PM, Jonathan Rees wrote:
>>
>> On Thu, Oct 20, 2011 at 4:16 PM, Kingsley Idehen<kidehen@openlinksw.com>
>>  wrote:
>>>
>>> On 10/20/11 4:08 PM, Jonathan Rees wrote:
>>>>
>>>> On Thu, Oct 20, 2011 at 3:54 PM, David Booth<david@dbooth.org>    wrote:
>>>>>
>>>>> What document?  Pointer please?
>>>>
>>>> Sorry, http://www.w3.org/2001/tag/awwsw/issue57/latest/
>>>>
>>>>
>>> Jonathan,
>>>
>>> I note that you have an entry for information resource in the glossary
>>> but
>>> nothing for:
>>>
>>> 1. resource
>>
>> Other than in the collocation 'information resource', the word is used
>> only once, and there not in a technical sense. Did you understand it
>> to be in an undisclosed technical sense? I hate to give up a perfectly
>> useful ordinary English word (look it up if you want to know what I
>> meant), but I guess I should reword that sentence to avoid it, since
>> 'resource' has been defined technically in so many different ways,
>> with so many different connotations, that it is now nearly useless in
>> web-land.
>>
>>> 2. non information resource.
>>
>> If I tell you what a dog is, is it necessary then to also tell you
>> what a non-dog is?
>>
>> I do not use this term and take pains not to. I don't even want to
>> evoke it as a category. The subject has been covered many times on
>> this list, and this particular report does not need to take a stand.
>>
>> Since I believe that 'information resource' as a type distinction has
>> derailed this whole discussion, and if it's useful at all, is emergent
>> rather than fundamental, I do not want to say anything that suggests
>> that something exists that is not an information resource. I think
>> such a world model would be unreasonable, but it is not ruled out by
>> any consideration that is supposed to arise in this document. And (as
>> David Booth keeps pointing out) the existence of so-called NIRs is not
>> forced by webarch-like considerations, only by ontological
>> considerations, which no matter how compelling, are (as far as I've
>> been able to determine) out of scope.
>>
>> It would be nice to eliminate the 'information resource' category from
>> this document. It just seems to fan the flames. I believe it's
>> possible - I have done so in other pieces of writing with good results
>> - but I haven't attempted it in this particular document yet.
>
> That's another route to the essence of my post i.e., why not remove it since
> it sticks out in a really awkward way?
>
>>> My comments are more to do with completion and less to do with opinion --
>>> my
>>> views on either aren't a secret to anyone :-)
>>
>> Appreciated, but distractions can inflame.
>
> So I can safely conclude:
>
> 1. resource -- distraction.
I think it is; whether you agree is up to you, and I can't vouch for
your safety if you do.
> 2. non information resource -- distraction.
Similarly.
> 3. information resource -- a distraction that's on its way out?

I don't think any of us can predict what's on its way in or out.
Personally I'm not keen on the expression, however.

I'd like to banish the word "representation" as well, as it has been
similarly divisive, but it's a useful category, so I'd need to come up
with another word (right now I use "potential retrieval result").

But even if all the categories go away, there remains something much
more important, which is how retrieval bears on the properties of what
the URI refers to. When I say <http://example.net/doc> dc:title "The
Fish", the important thing, if you buy something like httpRange-14, is
that retrievals (2xx, in HTTP) using URI http://example.net/doc, even
if they differ in other ways, all have dc:title "The Fish". You don't
need to appeal to "information resource" as a category to think that's
important or sensible. I would just say the IR idea would emerge -
something that could play a role in situations like this might be
designated an "information resource". But there's no necessity to call
this out as a category distinction.

Even if there are other ways to think about retrieval vs. reference
that work better that this, I hope my point remains: the properties
are more important than the types.

Jonathan

> Kingsley
>
>>  The fewer the assumptions,
>> the easier it should be to swallow.
>>
>> Best
>> Jonathan
>>
>>> --
>>>
>>> Regards,
>>>
>>> Kingsley Idehen
>>> President&    CEO
>>> OpenLink Software
>>> Web: http://www.openlinksw.com
>>> Weblog: http://www.openlinksw.com/blog/~kidehen
>>> Twitter/Identi.ca: kidehen
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>
>
>
> --
>
> Regards,
>
> Kingsley Idehen
> President&  CEO
> OpenLink Software
> Web: http://www.openlinksw.com
> Weblog: http://www.openlinksw.com/blog/~kidehen
> Twitter/Identi.ca: kidehen
>
>
>
>
>
>
>
Received on Friday, 21 October 2011 11:34:27 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 22:56:40 UTC