W3C home > Mailing lists > Public > public-sweo-ig@w3.org > March 2008

Re: Comments on http://www.w3.org/TR/2007/WD-cooluris-20071217/

From: Leo Sauermann <leo.sauermann@dfki.de>
Date: Sun, 16 Mar 2008 15:04:30 +0100
Message-ID: <47DD28EE.2050401@dfki.de>
To: Andy Powell <andy.powell@eduserv.org.uk>
CC: public-sweo-ig@w3.org

Dear Andy,
thanks for the feedback

we closed our round of reviews for the 20071217 version of "cool uris" 
and incorporated feedback.
The feedback and resolved issues are here:

After another telco with TAG, we will publish a last-call working draft.

The current version of the editors draft is here:

below the answers to your feedback, until the next public working draft 
no immediate reaction is needed from your side.

It was Andy Powell who said at the right time 21.12.2007 18:37 the 
following words:
 > This document looks pretty good and I agree with the general thrust of
 > what it is trying to do. However, I think there are problems with its
 > use of terminology. Some comments follow...
 > Abstract
 > I wonder if the abstract could be improved a little. For example, the
 > first sentence doesn't capture the fact that RDF can be used to describe
 > resources that are digital, physical or conceptual. The paragraph also
 > underplays the fundamental importance of URIs. How about

Good ideas,

 > The Resource Description Framework RDF allows the users to describe
 > digital resources (e.g. Web documents), physical resources (e.g. people
 > and things) and conceptual resources (e.g. colors and topics) in a
 > computer-processable way.

This is much for a sentence. The digital/physical/conceptual
discrimination is - although correct and useful - too academic
in my eyes.
I replaced it to this:
The /Resource Description Framework/ RDF allows the users to describe both
Web documents and concepts from the real
world—people, organisations,
topics, things—in a computer-processable way.

 > Publishing such descriptions on the Web
 > creates the Semantic Web. URIs (Uniform Resource Identifiers) are very
 > important, providing both the core of the framework itself and the link
 > between RDF descriptions and the Web. This document presents guidelines
 > for their effective use. It discusses two strategies, called 303 URIs
 > and hash URIs. It gives pointers to several Web sites that use these
 > solutions, and briefly discusses why several other proposals have
 > problems.

changed to your version.

 > 1. Introduction
 > Editorial note: "website" should be "Web site"?

changed to "Web site", as used elsewhere

 > 2. URIs for Web Documents
 > This section appears to introduce 'Web document' as a symonym for
 > 'information resource'? I prefer this in some ways. However, I presume
 > that the use of 'Web document' was rejected by previous W3C WG
 > discussion? Why re-introduce it here?

This is under discussion, some members of the TAG also do not like
"non-information resource"
for "cool uris" our decision is now documented here:

 > I have no idea what the 'traditional Web' is!? :-) I am aware of people
 > using http URIs to identify non-information resources (e.g. the DCMI
 > uses http URIs to identify its conceptual metadata terms) for well over
 > 5 years - probably nearer 10 years.

You are aware of it, but we want to reach a broader target audience,
instead of using the flashy "web 2.0" we would say "traditional"
in contrast to "semantic" :-)))
I do not change the text, as I interpret this comment as a matter of style,
not written-in-stone factual definitions of the term "traditional".

 > In the final sentence of this section, the tense moves from past to
 > present.

done, changed to
"The notion of resource /identity/ was not so important on the 
traditional Web,
a URL simply identified whatever we see when we type it into a browser."

 > 2.1 HTTP and Content Negotiation
 > Somewhere in this paragraph there is a change from using
 > 'representation' to using 'version'. This is potentially confusing. I
 > think that 'representation' should be used throughout.

good idea, done.

 > 3. URIs for Real-World Objects
 > I think this section needs to be clearer about when it is making
 > statements that are true for all time and when it is making statements
 > of current best practice. For example, is the statement "The standard
 > Web transfer protocol, HTTP, should be used." true for all time or an
 > indication of current best practice? It seems to me that the
 > requirement around ambiguity will still apply in 100 years time (say)
 > whereas the requirement around use of HTTP may well not.

The document's scope is deliberately about "cool" uris, being hosted on 
HTTP servers.

If, in 100 years, HTTP is not there anymore, the document is not cool 
No W3C document makes statements for all times,
as a scientist, I can understand your argumentation,
but will not follow it given the target audience.

Adding "cool" instead of "good" in this sentence should clarify the goals:
"This is so important that we make it our number one requirement for 
/cool/ URIs"

 > Under '2. Don't be ambiguous' the text uses both 'document' and
 > 'Web-retrievable document' for what has previously been referred to as a
 > 'Web document'. This mix of terminology is confusing.

good point, streamlined to "Web document"

 > The figure at the end of this section is also potentially confusing.
 > Firstly it uses both 'URI' and 'URL' without explaining why - I'd prefer
 > to see URI used throughout, i.e. "Resource URI", "RDF Document URI" and
 > "HTML Document URI".

point, ok, done.

 > Secondly, the language around this figure has
 > moved from 'version' to 'description'. An HTML document is not
 > necessarily a 'description' of a non-information resource - it is a
 > 'representation' of it. Again, 'representation' should be used
 > throughout this section.

replaced some of the "describing" by "representing", but not all
due to stye (its still english, and a little change in terms
makes it more pleasent to read).

 > 3.1 Distinguishing ...
 > It seems to me that introducing 'web documents' and 'real-world,
 > non-document objects' as new terminology is ultimately more confusing
 > than simply using 'information resources' and 'non-information
 > resources'.

others think not, and some comments label "non-information resources"
a term that should never have seen the light of this earth, which
I humourously can connect to.

Also, you above say yourself that you like/prefer "web document"

again, see

 > The latter aren't perfect (far from it!), but the
 > introduction of the new terms doesn't seem to help, especially since the
 > new terms subsequently get used inconsistently - sometimes 'real world'
 > is used, sometimes ' non-document' is used, etc.

our inconsequent use is due to the general confusion about these terms.
In the final version, one term will be conistently used.

 > Again, the words 'describes' and 'describing' are used in this section
 > where 'represents' and 'representing' would be better.

ah, creepy "describe", be gone.... (but not in SPARQL...)
Sometimes, we keep it deliberately, for example in
"For example the person Alice is described in a web document, Alice's 
(this sentence may still change depending on the actual used term for
web document)

 > 4.4 Cool URIs
 > It feels slightly odd that 'Cool URIs' is used in the title of this
 > working draft, but the section explicitly on this topic is quite small
 > and buried. It feels to me a bit like the main title of the working
 > draft is wrong - perhaps "The effective use of URIs for the Semantic
 > Web" would be better?

Yes, that may be better for a W3C recommendation, but this document
sums up TAG decision http-range-14 and current best practice.
Also, the title is a pun on TimBl's nice "Cool URIs don't change"
and a deliberate decision. It also catches the addressed audience,
web developers.

 > ---
 > Summary: Overall, I think this draft suffers from inconsistent use of
 > terminology throughout, particularly in the area of what it means to be
 > 'on the Web', a 'Web document', an 'information resource', a
 > 'non-information resource', in the 'real world', and so on. Agreed
 > terms used need to be used consistently throughout and, preferably
 > defined separately in a glossary.

Due to limited time and space, we will not do a glossary,
your analysis of the incosistency helps us.

 > Hope this helps,

YES! thank you very much.

kind regards
Leo Sauermann

DI Leo Sauermann       http://www.dfki.de/~sauermann 

Deutsches Forschungszentrum fuer 
Kuenstliche Intelligenz DFKI GmbH
Trippstadter Strasse 122
P.O. Box 2080           Fon:   +49 631 20575-116
D-67663 Kaiserslautern  Fax:   +49 631 20575-102
Germany                 Mail:  leo.sauermann@dfki.de

Prof.Dr.Dr.h.c.mult. Wolfgang Wahlster (Vorsitzender)
Dr. Walter Olthoff
Vorsitzender des Aufsichtsrats:
Prof. Dr. h.c. Hans A. Aukes
Amtsgericht Kaiserslautern, HRB 2313
Received on Sunday, 16 March 2008 14:05:21 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Sunday, 16 March 2008 14:05:22 GMT