RE: [oiic-formation-discuss] Profiles. Attempting a definition.

[jose lorenzo <hozelda@yahoo.com>]

--- On Wed, 6/18/08, Hurley, Garry (L&I - OIT) <ghurley@state.pa.us> wrote:

> From: Hurley, Garry (L&I - OIT) <ghurley@state.pa.us>
> Subject: RE: [oiic-formation-discuss] Profiles. Attempting a definition.
> To: "Dave Pawson" <dave.pawson@gmail.com>, oiic-formation-discuss@lists.oasis-open.org
> Date: Wednesday, June 18, 2008, 8:56 AM
> Definition: "a profile is a set of policies,
> guidelines [[and constraints]] defined for a particular
> application."
> 
> 
> Unless I get any more feedback when USA wakes up I'll
> add that.
> 
> >
> > The second paragraph isn't really needed, I
> think.
> 
> +1.

Once finished, a profiles-based ODF would either imply or explicitly specify within it a more specific definition than given above.

An analogy would be with the term "plugin". It has a general meaning, but if you want to develop for or use (eg) Firefox, you realize that there is a much more precise working definition, whether or not it is explicitly stated on any Firefox documentation.

I actually like the wikipedia (engineering) definition: "In standardization, a profile consists of an agreed-upon subset and interpretation of a specification" except that I would consider "subset" to mean a "subset of the universe of all components of all existing profiles when combined". [Also, I may argue later that it's best not to require profiles to overlap.]

I think we can change any definition of profiles later, as we better flesh out the TC charter and TC working notes. For now, put anything that basically works. Later, the definition would in the minimum eventually be adapted to use similar terms to those already being used (or to be used) within ODF. [Eg, "application" and "policies" might be defined or substituted if a similar concept is already in use within the ODF spec.]

It's too early to know right now the *best* definition of what is still unfleshed vaporware in version 0.00.00.1.