- From: Karl Dubost <karl@w3.org>
- Date: Wed, 18 Apr 2007 12:23:57 +0900
- To: HTML WG <public-html@w3.org>
Le 16 avr. 2007 à 08:16, Henrik Dvergsdal a écrit :
> I share your concerns about the size of the specification and the
> corresponding error rate - especially since the language (according
> to the current proposal) will be defined solely by means of
> english prose - no formal grammar, schema or anything.
Maybe because there are multiple views and readers of this
specification.
I understand the repeated call of Ian Hickson that in the given
circumstances it is easier for him to work on a single document.
But labeling part of the document for specific readership could
improve the quality of the specification as a whole.
When we decide to have a frozen version of the specification some of
the things that we could do.
* reading the specification with class of products in mind.
One browser developer will not have the same reading than a HTML
library developer or an authoring tool.
-> this requires that some people with their own domain of
competences review the spec.
Example: I develop this wysiwyg authoring tool, I will review
with only this perspective.
* reading the specification with specific users in mind.
Denis Boudreau started a very good list of specification user
profiles.
-> this requires that some people with a specific profile review
the spec.
Example: I am a HTML code monkey (reference to wired), I will
review with this perspective
I am teaching HTML to students…
If we prepare well in advance with the list of class of products and
user profiles and designated volunteers for reviews. It would be very
valuable for the specification as a whole.
--
Karl Dubost - http://www.w3.org/People/karl/
W3C Conformance Manager, QA Activity Lead
QA Weblog - http://www.w3.org/QA/
*** Be Strict To Be Cool ***
Received on Wednesday, 18 April 2007 03:25:37 UTC