W3C home > Mailing lists > Public > public-ldp@w3.org > March 2013

Re: Section 4: LDPR/non-LDPR formal definitions

From: Erik Wilde <dret@berkeley.edu>
Date: Mon, 25 Mar 2013 13:30:55 -0700
Message-ID: <5150B3FF.4020501@berkeley.edu>
To: Henry Story <henry.story@bblfish.net>
CC: Kingsley Idehen <kidehen@openlinksw.com>, public-ldp@w3.org
hello henry.

On 2013-03-25 13:16 , Henry Story wrote:
>> On 3/25/13 3:50 PM, Erik Wilde wrote:
>>> absolutely nothing is wrong with that in my mind; it's actually the opposite: i think that's what we should be doing from the REST perspective. however, it seemed to me that whenever i suggested that it would be good to expose LDP semantics on the media type level, the majority opinion in the WG was that this is not what you normally do for RDF-based designs, and that instead we should be exposing generic RDF media types.
> Ok. I understand your reasoning now. But I don't agree that this is a good reason to do this. RDF was
> always meant to be used as linked data, just as HTML was always meant to have <a href=""> links be
> followed. As LDP gets deployed the overwhelming amount of published data will follow this pattern. Sites
> that don't do this are pretty useless, and so will just die out, or not being linked to, be invisible.

this would be a fascinating dinner conversation. text/plain was never 
meant to have links (and of course it doesn't), it simply has the 
potential. HTML cleverly used this potential. RDF also has no links, it 
simply has the potential. linked data cleverly uses this potential.

did plain text die? not at all, because it still is a wonderful way for 
humans to represent the structured textual data that they care about. it 
is just not a great web citizen, because you can link to it, but not 
link from it. so on the web, HTML is much more useful. but it actually 
would be interesting to know how many plain text files are *linked to* 
from the web. very very many, i would assume, because there is a lot of 
useful content out there in plain text (just think of the billions of 
plain text emails made available from linked mail archives). a lot of 
useful content is in plain text, and we can conveniently discover it 
using HTML. both are useful things for different purposes (pure data 
format vs. hypermedia format), and the same is true for RDF and linked data.

cheers,

dret.
Received on Monday, 25 March 2013 20:31:18 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:03:10 UTC