W3C home > Mailing lists > Public > www-qa-wg@w3.org > April 2005

Re: [Draft-SpecGL] Umbrella specifications

From: Dominique HazaŽl-Massieux <dom@w3.org>
Date: Wed, 27 Apr 2005 18:14:43 +0200
To: Karl Dubost <karl@w3.org>
Cc: www-qa-wg@w3.org
Message-Id: <1114618483.5653.68.camel@stratustier>
Le mercredi 27 avril 2005 ŗ 11:52 -0400, Karl Dubost a ťcrit :
> 4.2. Umbrella specifications
> A specification is a document that prescribes technical requirements to 
> be fulfilled by a product, process or service.
> In the last ten years, the technologies developed at W3C have evolved a 
> lot. W3C Working Groups have moved from monolithic specifications to 
> set of specifications to define the requirements of a technology. Often 
> this move was necessary to ease the editing process or/and to handle 
> effectively the functional division of the technology.

I would avoid talking about W3C history at this stage; the point made
makes sense outside of the simple W3C scope; I suggest instead:
There is a tension between defining a technology by setting as many
requirements as possible inside one document and setting a few
requirements in many documents. The former allows to get a more cohesive
set of requirements, while the latter enables a more flexible

> A monolithic specification is a document which contains everything that 
> is necessary to implement the technology.

That's too broad a definition (e.g. I think what we intend by monolithic
specification could very well have normative references, which would
make the previous sentence false). What about:
A monolithic specification sets all the requirements created for a given
technology in a single document.

>  A set of specifications is a 
> set of documents which define one or a few requirements of the 
> technology. There are most of the time strong dependencies between the 
> documents.

I don't know that the term "set of specifications" is useful...

> The W3C Process document provides a framework for editors to help them 
> to publish their document and to enforce some quality practices (for 
> example, implementation phase during CR).

s/editors/Working Groups/
s/their document/their specifications/
s/to publish/publish/
s/to enforce/enforce/
s/CR/Candidate Recommendation/

>  Though it doesn't define per 
> se the notion of technology that would be covered by a coherent set of 
> specifications.

Suggested rewording:
But it doesn't define of what a technology consists, nor how a
technology relates to one or several specifications.

> Defined in one or several documents, specifications can import 
> requirements of other specifications with normative references. Some 
> specifications, denoted below as umbrella specifications, create their 
> own interfaces by simply grouping requirements of existing 
> specifications in a well-defined manner.

Since we don't use the term interfaces in our definition of
specification anymore, I don't think it should appear here. So:
-> "...create all the requirements of the technology they define by..."

> Figure 1: Umbrella specification
> 	[Here the figure]
> 	http://www.w3.org/TR/qaframe-spec/Umbrella-Specification.png
> On this figure, the technology is composed of two modules (defining 
> functional division of the technology), a profile (defining the 
> requirement of implementation for a specific device) and a primer 
> (introducing the technology and its basic concepts). An "umbrella 
> specification" document groups them together making it a logical, 
> usable and complete technology.

(I think the notion of umbrella specification should get more
discussion, e.g. incorporating what we discussed during our F2F in
Reading; but I think the current text with the proposed changes is fine
for our intermediate publication)

Dominique HazaŽl-Massieux - http://www.w3.org/People/Dom/

Received on Wednesday, 27 April 2005 16:14:47 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:14:34 UTC