Re: Accellera and DASC

Date: Mon, 14 Jun 2004 18:27:15 -0400
To: stds-dasc@server.eda.org, <stds-dasc@server.eda.org>
From: "Gregory D. Peterson" <gdp@utk.edu>
Subject: Re: Accellera and DASC

Evan et al,

I've been participating with the DASC and other standards organizations for
nearly a decade now, representing various interests over time. Here's my
two cents worth:

- The fact that Accellera (and OVI before it) funded LRM development for
standards seems reasonable to me. Simply put, an individual or small team
must do the heavy lifting for developing an LRM. A committee will take
forever to do this and even an individual will take a long time if someone
isn't
paying for it (this has been borne out numerous times). If there are
interests that
are motivated enough for a standard to help pay the freight for its
development,
then that is part of the market forces that we all deal with. These
efforts have
not circumvented the DASC process, but they have made some of the
implementation
details for more efficient.

- Everyone that participates in standards developments comes to the mix with
their various biases and prejudices. In a prior position, I managed some
Air Force
contracts that helped develop LRMs for VHDL, VHDL-AMS, and SLDL/Rosetta. The
language committees were still open, but our funding for the standards
followed
a decision that the standards helped with the quality, affordability, and
sustainability
of military electronic systems. My participation in the committees was to
represent
my employer's (USAF) needs as I saw them. Was that somehow nefarious? I
don't think so. Through my interactions with the DASC, I've gotten to know
folks
like Gabe,
Paul, Victor, Peter, Dennis, and numerous others. I have certainly not
always agreed
with their opinions, but I have never had reason to question their
integrity. I expect
that each of them will represent the best interests of the industry as they
see it (and,
of course, this will almost always align with their corporate interests as
well).

I was chair of the VHDL International Users Forum (VIUF) when Accellera was
formed,
and then chaired the AUF. As the chair, I sat in on Accellera board
meetings. What I
witnessed included companies wrestling over what technologies would become
standards
and which would win in the market. There are lots of competing
technologies and
strongly held (often diverging) opinions. The process may not always be
pretty, but I
never felt that it was corrupt.

Note that now, as an academic, I don't have any real financial interest in
which particular
standard wins (although cool new technologies do make for interesting
research). The
same is true of Gabe, who now participates in the industry as a member of
the press.
I have never seen anything other than honorable actions from Gabe, and I
truly think he
acts for what he thinks is best for the industry.

One can debate the merits of developing standards with individuals versus
corporations
voting. It may be intellectually stimulating and good for venting stress,
but it's a waste
of time. The IEEE added the CAG to respond to demands for various
industries. If the
demand for such an avenue for approving standards works, it will
flourish. If not, it will
fade away. Time will tell, but I suspect the CAG will be popular for new
standards.
(Once again, as an academic, I will cheerfully discuss this over a beer
with you.)

The System Verilog debate then focuses on whether it was proper or
intelligent to use
CAG as opposed to DASC. Each member company of Accllera certainly had the
right to
choose the best path for SV as they saw it. Nothing I've heard makes me
doubt the integrity
of the process and that each company voted according to its best
interests. If other
companies disagree, then perhaps they will join Accellera or CAG to have
their voices
heard. The question of whether this decision will result in standards
divergence will have
to be addressed by the committees and IEEE oversight. The question of
whether the
Accellera decision is stupid or not I will leave to all to debate as they
see fit.

Bottom line: be careful about making accusations. Nothing says we will
all agree (ever :)
but differences in opinions do not necessarily imply improprieties. I
guess that was more
like a nickel's worth...

Greg

At 02:46 PM 6/14/2004, Gabe Moretti wrote:
>Evan,
>this information is actually incorrect. Stu Sutherland non-voting
>position on the Board of Directors of Accellera is the result of Stu
>being the chair of the Accellera Designers Forum. The Forum was my
>idea, but I have never had any active participation after the initial
>organization of the Forum. Since I had the idea and organized the
>beast, my name was associated with it. Although Stu served as Chair
>the past twelve months, he has now resigned the position and sent out a
>public email announcing his decision. As you well know it is difficult
>to keep a web page up-to-date, especially when doing so depends
>strictly on volunteers. I am sure you do know that, although the
>eda.org page receives some financial support from Accellera, it is run
>completely independent from Accellera and that its system
>administrators are not connected with Accellera in any way. Gabe
>----- Original Message -----
>From: "Evan Lavelle" <anti.spam1@dsl.pipex.com>
>To: <stds-dasc@eda.org>
>Sent: Monday, June 14, 2004 1:58 PM
>Subject: Re: Accellera and DASC
>
>
> > Gabe Moretti wrote:
> >
> > > I was one of the three persons that worked on and executed the
> > > merger of
>VI
> > > and OVI that created Accellera, because at the time I was the
> > > Chairman
>of
> > > VI. Since I no longer work for a corporate member of Accellera, I
> > > have
>no
> > > formal connection with that organization and have not had it for
> > > the
>last
> > > four years.
> >
> > Gabe -
> >
> > Thanks for your reply.
> >
> > eda.org still lists you as the chair of the Accellera Designer's
> > Forum. Is this information correct and, if so, perhaps you could
> > kindly tell us if this is an independent organisation, or if it
> > connected with Accellera? In addition, would you mind telling us
> > whether or not you receive any financial remuneration for your
> > advisory work for the Accellera board? If, alternatively, you
> > believe that there is no conflict of interest in your activities,
> > then I would be happy for you to simply state this.
> >
> > Evan Lavelle
> >

Dr. Gregory D. Peterson
Assistant Professor
Electrical and Computer Engineering
University of Tennessee
Knoxville, TN 37996-2100
(865)-974-6352 (voice)
(865)-974-5483 (fax)
gdp@utk.edu
Received on Mon Jun 14 23:15:31 2004