Re: Proposal to amend the httpRange-14 resolution

On Tue, Mar 27, 2012 at 5:42 PM, Tore Eriksson <tore.eriksson@gmail.com> wrote:
> On Wed, Mar 28, 2012 at 5:45 AM, Jonathan A Rees <rees@mumble.net> wrote:
>> On Tue, Mar 27, 2012 at 10:10 AM, Tore Eriksson <tore.eriksson@gmail.com> wrote:
>>> 2012/3/27 Jonathan A Rees <rees@mumble.net>:
>>>> 2012/3/26 Tore Eriksson <tore.eriksson@gmail.com>:
>>>>> Hi Tim,
>>>>>
>>>>> thank you for your detailed input. I'll add my comments inline.
>>>>>
>>>>> 2012/3/26 Tim Berners-Lee <timbl@w3.org>:
>>>>>>
>>>>>> On 2012-03 -26, at 01:31, トーレ エリクソン wrote:
>>>>>>>>> This proposal entails a partial reversion of the httpRange-14
>>>>>>>>> resolution. Specifically, it suggests that a representation retrieved
>>>>>>>>> from a HTTP URI will never* be equivalent to what the URI denotes (the
>>>>>>>>> resource), but will always be a description (of the state) of the
>>>>>>>>> resource, eliminating the risk of confusing a resource with its
>>>>>>>>> description.
>>>>>>
>>>>>> [...]
>>>>>>
>>>>>>
>>>>>>> However, if you don't own the URI, stating this seems to irresponsible.
>>>>>>> The owner might add a content-negotiated Swedish translation with a
>>>>>>> dc:title of "Hittad" and make your statement invalid.
>>>>>>
>>>>>>
>>>>>> That is hair-splitting -- yes, a generic IR URI may indeed by correspond to
>>>>>> a series of more specific versions in different languages
>>>>>> (See http://www.w3.org/DesignIssues/Generic and the associated ontology)
>>>>>> and one can argue whether people incorrectly actually use
>>>>>> one title to refer to the whole lot, but I think it is useful.
>>>>>
>>>>> I have no problem with adding the title to the generic resource,
>>>>> especially if you own the URI. My understanding of Jonathan's text was
>>>>> though that by looking at one representation titled "Trouvee", one
>>>>> could infer that all representations would have the same title.
>>>>
>>>> This is an incorrect reading of what I wrote. I was very careful in
>>>> what I said, and I did not say this.
>>>
>>> Sorry if I misrepresented your text. I'll explain why I thought it
>>> meant this below.
>>>
>>> You started with
>>>>>
>>> To say that any representation retrieved from "http://example/hen" has
>>> (or will have) "Trouvée" as its title, we can write (in Turtle
>>> [turtle])
>>>    [ir:onWebAt "http://example/hen"] dc:title "Trouvée".
>>> [this tells that] if they dereference that URI, they will get
>>> something with that dc:title [1]
>>>>>
>>>
>>> And then used the generic URI instead of the blank node.
>>>
>>>>>
>>> A common practice is to use an absolute URI as a name for a (generic)
>>> information entity that is on the Web at that URI.
>>> <http://example/hen> dc:title "Trouvée".
>>>>>
>>>
>>> Then you followed up with
>>>
>>>>>
>>> Whether we can expect in general that a dereferenceable URI will be
>>> understood as a name for a (generic) information entity on the Web at
>>> that URI is the essence of the heated httpRange-14 debate
>>>>>
>>>
>>> I assumed that this meant that when following httpRange-14 the RDF
>>> above is expected.
>>
>> It would only be expected if it were true, and it would only be true
>> if *any* representation, not just one of them, had that property.
>
> That is what I thought at first, by looking at the initial RDF. Then I read
>
> "This is a useful thing to say, since it is predictive: It tells
> someone that if they retrieve using URI, they will get something with
> that dc:title."
>
> This sounded a lot as if all the representations are constrained to
> have the same title.

The generic resource only has the property if all the representations
do, according to the theory. The constraint is mutual. You could have
some representations with the title and some without, but then the
generic resource wouldn't have the property.

> [snip]
>
>>> My train of thought was this: If a HTML document is retrieved by with
>>> a 200 GET, then under httpRange-14 this is an information resource,
>>> and also a generic information entity. Let's say that the HTML
>>> document received has the dc:title "Trouvee". Then the generic
>>> resource also has the same title (according to [1])
>>
>> That's the fallacy. The generic resource only has the property if, no
>> matter which representation is retrieved (across conneg variation
>> etc.), that representation has the property.
>
> Since this is impossible to check for the user, wouldn't it be
> difficult to have predictable meta data in practice?

Yes. Yet people do it all the time. Sometimes they do it because they
can control what representations are delivered; sometimes because they
have inside information; sometimes because of some kind of promise
made by someone who does control the representations; and sometimes
it's just a gamble, just like href= in HTML (think of that as
implicitly saying that any retrieved representation has the property
that it would be appropriate to display should the link be clicked -
it's very hard to know that, yet people gamble on it all the time).
The point here is that it's all about meaning, not truth.

At least this is the only way I can make sense of this crazy system.
The only alternative is to just to say it's all nonsense and walk away
from it.

Jonathan

>> I guess I didn't make this clear, but it's a difficult idea to
>> express. If you want to know what bit of philosophy inspired me to
>> express it this way, see _Truth, Meaning, Reality_ by Horwich.
>
> I'll make sure to put that on my e-book reader later.
>
>> I thought this would be clear in the overall context of the note. I
>> guess I have to look again to see how I failed to communicate.
>
> I hope I've been of some help.

Received on Wednesday, 28 March 2012 01:02:45 UTC