Re: [vhdl-200x] Stephen Bailey's concerns on Group Organization and PAR

From: Jim Lewis <Jim@synthworks.com>
Date: Tue Dec 21 2010 - 09:21:20 PST

Hi All,
I am writing this to address some of Stephen Bailey's
concerns.

> [Stephen]
> I find it impossible to believe that there is no
> clue of what the scope of this effort will actually
> entail.
The amount of detail that is in the PAR is exactly the same
as the PAR for VHDL-2008 (which is based off of one from when
Stephen was chair). The VHDL-2008 par is here:
   http://www.eda-twiki.org/vasg/p1076_2008_approved_par.pdf

Due to your concern, I checked other PARs and have found
them to have a similar level of detail. For example the
latest (2010) SystemVerilog PAR is here:
   http://www.eda-twiki.org/twiki/pub/P1800/P1800Workinggroup/P1800_PAR_2010.pdf

The 2010 P1800 SystemVerilog purpose states:
   This standard develops the 1800 SystemVerilog language
   in order to meet the increasing usage of the language in
   specification, design and verification of
   hardware. This revision corrects errors and clarifies
   aspects of the language definition in the 1800-2009
   standard. This revision also provides enhanced features
   that ease design, improve verification, and enhance cross
   language interactions.

> [Stephen]
> SG's have 6 months. They should be able to determine the relative scope
> of the project in that time frame. Asking for PAR approval after 2
> meetings that focus only on the type of organization of the WG is
> inappropriate.

The IEEE-SA by-laws actually say under some conditions the PAR can be
approved in one meeting. Instead we took two.

> [Stephen]
> Why would anyone want to be
> involved? Can you think of a better way to ensure maximum
> dissatisfaction with the result than to leave an understanding of the
> scope unstated? The email discussion since I distributed my negative
> vote on PAR approval indicates how wide the expectations are.
The following is similar to a quote I heard (was it Judy Gorman or
Mick Jagger?):
   In the standards process you don't necessarily get
   what you want, but if you participate you usually get
   something you can live with.

If we follow the Accellera process, there are lots of checks
and balances in it. This is how I see the process working
(note, these are just my thoughts, the WG will have to
approve any process that we decide to use):
   1) Requirements team
      Who: All
      Purpose: Develop and prioritize requirements.
      Action:
        High priority requirements get forwarded to the
        proposals team. Low priority requirements don't
        get WG authorization to go forward.
   2) Proposals team
      Who: People who feel comfortable writing or reviewing
            one or more proposals
      Purpose: Write proposals for implementation of requirements
      Actions:
        Champions volunteer to work on requirement.
        If an enhancement does not get a champion,
        it does not go any further.
        All proposals are reviewed in detail approved by
        the proposals team before they go to the next step.
   3) Proposal vs Requirements Review
      Who: All (Requirements + Proposals team)
      Purpose: Make sure the proposals implement the requirements
                from the viewpoint of the requirements team.
      Actions:
        Vote on proposals to make sure they address
        the requirements.
   4) Write LRM changes
      Who: LRM mechanics
      Action:
        Write LRM language change specifications for the proposals.
   5) LRM change vs. Proposal Review
      Who: All (LRM + Proposals + ?Requirements?)
      Purpose: Identify any differences in the proposal and the
                LRM change (if any).
      Action:
        Vote to approve or disapprove any differences in
        LRM change from proposals.

Peace,
Jim

-- 
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Jim Lewis
Director of Training             mailto:Jim@SynthWorks.com
SynthWorks Design Inc.           http://www.SynthWorks.com
1-503-590-4787
Expert VHDL Training for Hardware Design and Verification
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.
Received on Tue Dec 21 14:13:08 2010

This archive was generated by hypermail 2.1.8 : Tue Dec 21 2010 - 14:13:59 PST