W3C home > Mailing lists > Public > public-lod@w3.org > June 2010

Re: Inclusion of additional (non dereferencable) data?

From: Nathan <nathan@webr3.org>
Date: Fri, 11 Jun 2010 10:38:47 +0100
Message-ID: <4C120427.9040206@webr3.org>
To: Story Henry <henry.story@bblfish.net>
CC: Linked Data community <public-lod@w3.org>
Story Henry wrote:
> On 10 Jun 2010, at 17:24, Nathan wrote:
>> Here's a common example of what I'm referring to, suppose we have a (foaf) documenthttp://ex.org/bobsmith which includes the following triples:
>>  :me foaf:knows <http://example.org/joe_bloggs#me> .
>>  <http://example.org/joe_bloggs#me> a foaf:Person ;
>>    foaf:name "Joe Bloggs"@en .
> This is ok by me. Adding more information is useful, as
> mentioned as it helps reduce connections. If you had a 1000 
> in your foaf file without any information, your client would
> need to make 1000 calls to get the info.

In all honesty, I'd probably not use a client that showed me 1000 
connections at a time, paging is vital when displaying information to 
humans so as to prevent information overload - caveat that as soon as 
you include 'order by' clauses to the view then all that dereferencing 
comes back in to play..

> Pragmatically there is no need to make a big fuss about this.

Concur, no need for a big fuss, personally feel it was worth a quick bit 
of open discussion and consideration so we don't blindly fall in to 
certain patterns without first considering the future effects, namely 
the discussion we're having now.

> People add what they want. If what they do goes out of date, then
> people will notice this.
> Your task as a UI designer is to allow people to 
>  1. see where the information came from
>  2. remove information from graphs they don't trust
> Like sig.ma. But if you don't crawl the whole web, you will get
> a much better effect.

sig.ma and related are very different use cases to the common (in the 
future) use-case though, typical web applications will have no need to 
display 'all' the related information, just the information that is 
worth considering under the context in which the human is considering 
the data, so the amount of dereferencing that needs done will be 
somewhat smaller.

>   Resources that make claims about other people that keeps changing,
> will be a nussance and avoided. So there will be social pressure to
> get that right.

Yes, fully agree, and another angle to give consideration :)


Received on Friday, 11 June 2010 09:39:58 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 15:16:06 UTC