W3C home > Mailing lists > Public > semantic-web@w3.org > February 2011

Re: URI in a URI

From: Nathan <nathan@webr3.org>
Date: Fri, 04 Feb 2011 16:28:34 +0000
Message-ID: <4D4C2932.6050808@webr3.org>
To: Hugh Glaser <hg@ecs.soton.ac.uk>
CC: "semantic-web@w3.org" <semantic-web@w3.org>
Hey Hugh - glad to see that (hopefully) my mail came across right, tbh 
I'm totally fed up with all the in fighting in the communities, and 
between the communities around the web in general - or perhaps better 
termed "posturing" (that includes me), and so I'm trying to reposition 
to get rid of all my own opinions and just point out the benefits and 
drawbacks of each thing (should I understand them well enough that is!), 
and where possible get people who do understand the conflicts and issues 
talking to come up with solutions.

There are brilliant people who all have great knowledge and 
understanding about different bits of the web, data and sem web, we need 
to leverage that to get the designs for each use case out there and 
understood.

Heck most of what we do needs trade offs to actually work, not picking 
something we feel is "the right thing" - what you said is right for your 
use case, not right for every use case, and we're better to focus on the 
valuable info you can bring here than on whether it applies to every 
case (for example one wouldn't publish ontologies in that fashion).

Anyway, I'm rambling - hope the mail finds you well,

Nathan

Hugh Glaser wrote:
> On 4 Feb 2011, at 15:41, Nathan wrote:
> 
>>> It is an important issue for the emerging services accessing the semantic web.
>>> Yes in general using a # is a bad thing.
>> It should be noted that the above is a personal opinion by Hugh, is certainly not consensus, and the truth of the matter is that for the purpose of naming, the lexical form of a URI is irrelevant.
>>
>>
> Ah, thanks Nathan.
> Very helpful message in general, and in particular I realise that this bit of mine was perhaps misleading.
> 
> I guess I did not mean "bad thing" :-)
> What I meant was that they can cause problems in these sort of situations, so I try to avoid them where possible.
> In particular, where there is a # in the name of a service, or a # in an argument to a service, then a bunch of issues can arise that make the service harder to invoke.
> And then you go on the explain them - thanks!
> 
> And for the record, I don't have a problem with # URIs for Linked Data - there are very good reasons why they may sometimes be the best solution.
> And yes, I am wary about terming things RESTful APIs without very careful thought - that is why I chose an example, Watson, that has nothing to do with me :-)
> 
> As you point out, one of the great things about doing this is that it is familiar to people who deal with forms; this means that these services are more likely to be accessible to non-SemWeb people.
> 
> Best
> 
Received on Friday, 4 February 2011 16:29:46 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 21:45:41 GMT