Re: Paving Cowpaths

On 5/21/07, Philip & Le Khanh
<Philip-and-LeKhanh@royal-tunbridge-wells.org> wrote:
Daniel Glazman wrote:
>> And to be honest, this is not the problem here. A spec cannot
>> target two audiences that are so different, implementors and users
>> (here, users are document authors). And to be honest too, a spec
>> _should not_ target users. Tutorials and books are here to educate
>> users.
>
> I'm sorry, I don't agree.  /Initially/, the user will need a book
> (or a tutorial, but good well-written books such as Dave Raggett's
> HTML 4 or Bert Bos & Hakon Wium Lie's CSS are far better, IMHO,
> being written by people deeply involved in the design process and
> decision making), but once the user has assimilated all he can from
> the book, he /needs/ the specification to which to refer whenever he
> needs to be certain of the exact letter of the law.
>

There *are* cases of official primers being attached to W3C
Recommendations. They usually do not have the status of the
recommendations:

Examples:
GRDDL Primer - http://www.w3.org/TR/grddl-primer/
XML Schema Primer - http://www.w3.org/TR/xmlschema-0/
RDF Primer - http://www.w3.org/TR/REC-rdf-syntax/
WSDL Primer - http://www.w3.org/TR/wsdl20-primer/
More available by doing a Google search for {{ primer site:w3.org }}

The RDF Primer is certainly an instructive case. It provides a
high-level, functional overview of the technology and how to use it
(as an author of RDF documents), but it clearly states that it is a
non-normative document. It is the first part of the series in the
specification.

In the case of HTML, there may be a case to be made that since there
are a number of other supportive technologies which web authors ought
to be using, that providing a Primer would not be advisable, since
there is material in other areas (CSS, JavaScript etc.) which is
relevant to the subject. That said, there is no reason not to have a
Primer if it is felt that it is required.

As the next version of HTML is an evolutionary step, it does not seem
necessary, so long as differences between the HTML 4.x/XHTML 1.x
branch are clearly flagged up so that both those implementing the
specification (which, as we move towards a Semantic Web, where more
and more people are parsing web data, will be a lot more than just the
teams working on IE, FF and Safari etc.) and those authoring pages can
get just the differences. For those who need a high-level overview of
HTML with the intention of constructing a web page, the sort of Primer
provided in the RDF specification seems not to be worthwhile, since
there are a plethora of third-party resources who will fill that role
quite ably (other, lesser-popular/more-future-oriented technologies
may not have that).

-- 
Tom Morris
http://tom.opiumfield.com/blog/

Received on Monday, 21 May 2007 22:21:32 UTC