RE: [vhdl-200x] Announcement

Hi Kevin,

Some facts relevant to your comments:

1.  1076.6 was developed under individual membership, not entity
membership WG.  Thus, you have provided an example of how an EDA vendor
was not held accountable under individual membership organization.

2.  It is my understanding that level 1 was defined to officially
incorporate the de facto standard (Synopsys's DC).  The WG then went on
to define a broader subset that could be synthesized.  If I'm wrong on
this, I'm sure someone will correct me.

3.  Synopsys refused to allow their employees to participate in 1076.6
due to a fear that such participation would endanger patents they have
in synthesis.  Unfortunately, this fear exists whether individual or
entity membership based.  However, one could make the claim that entity
membership would provide the corporation with more direct oversight and
control of what their employees are doing in the WG.  Whether that would
be enough to offset the patent protection concerns, I do not know.

Finally, I must question whether there is sufficient market relevance
for a broader synthesizable subset.  If there is, the VHDL users need to
get busy putting pressure on the synthesis providers.  Perhaps the
individual membership organization of 1076.6 failed to provide an
effective means for them to do so?  Clearly, it was not effective in
that regard if such demand does exist.

BTW, corporate /= EDA vendor.  Accellera requires using corporations to
back their standards efforts.

-Steve Bailey
 
> Stephen Bailey wrote:
> >1.  Entity (corporate) membership ensured that what is 
> important to end 
> >users got done.  This is what has been
> 
> >termed "market relevance."
> 
> 
> I am all in favor of corporate involvement in standards 
> development but the concern expressed by Mr. Lavelle can be 
> demonstrated by what happened with the original 1076.6 "Level 
> 1" synthesis subset. It was no coincidence that many of the 
> good, useful, and synthesizable features of VHDL that were 
> left out of the subset definition were those same features 
> that Synopsys VHDL Compiler (of that vintage) didn't support. 
> The Synopsys tool supported only those features that had 
> direct Verilog cognates with a few of the "must have" VHDL 
> features tacked on to appease the masses (configurations? 
> what are those?). It is quite disturbing that a standards 
> body would let itself be pushed around by a single corporate 
> entity just so that company can promote themselves as 
> compliant with a standard.
> 
> 
> What is the point of market relevance if the standards in 
> question are dominated by companies with a reputation for 
> sluggish response to the market? In the case of Synopsys, 
> their CEO has gone on record stating that they are backing 
> SystemVerilog in favor of (to the detriment of?) VHDL. I 
> believe that the apprehension Mr. Lavelle expressed is based 
> on a distrust of companies that may have a hidden agenda that 
> contravenes the needs of the VHDL user community. That being 
> said, I have no reason to believe that Acellera has been so 
> deliquent in this regard but there is a legitimate reason to 
> question corporate control of standards development.
> 
> 
> Kevin Thibedeau -- Eagerly awaiting a vendor that really 
> supports VHDL-93 in 2005
> 
> ________________________________
> 
> From: owner-vhdl-200x@eda.org on behalf of Bailey, Stephen
> Sent: Wed 7/6/2005 9:38 AM
> To: vhdl-200x@eda.org
> Subject: RE: [vhdl-200x] Announcement
> 
> 
> All,
> 
> 
> I would add the following to Gabe's and Evan's comments:
> 
> 
> No one at DASC, CS or IEEE consciously made the decision to 
> turn over standards development to Accellera.  What happened 
> is that Accellera saw a need for a better way to develop EDA 
> standards than what was occurring in DASC.  So, Accellera 
> created a better way.  I see 3 key improvements that Accellera made:
> 
> 
> 1.  Entity (corporate) membership ensured that what is 
> important to end users got done.  This is what has been 
> termed "market relevance."
> 
> 
> 2.  Finances.  Because companies are involved and not 
> individuals, and as a check on market relevance, if it is 
> important to the users, money will be found to make the 
> standard happen.  And then when it happens, more money is 
> found to promote it.  (Make it successful.)
> 
> 
> 3.  Efficiency.  The entire process is more efficient.  Part 
> of this is due to the fact that #1 and #2 ensure that only 
> those things that are really important are done and that when 
> they are done, funds are available to make it happen.  But, 
> it is also the part that many on this forum do not like:  
> Making standards is somewhat analogous to making sausage.  
> You may not personally like the political give-and-take 
> involved, but it is necessary.  Primarily, within Accellera 
> end users have the power to make EDA vendors do what is best 
> for the industry.  (Or to put it another way, an EDA vendor 
> cannot be successful promoting a position on a standard 
> without significant end user support.)  Inside DASC, the 
> individual membership organization makes it difficult as the 
> members only represent themselves and cannot be assumed to be 
> promoting their company's interests.  In Accellera, the 
> positions of members are public to all.  End user companies 
> can hold EDA vendors accountable.
> 
> 
> Finally, the individuals who are representing their companies 
> in Accellera have done a more efficient job in operating the 
> organization.  Sure, the DASC can improve here.  But, it is 
> because of the market relevance (importance to the member 
> companies) that there is more visibility of what happens in 
> Accellera.  Therefore, the structure of Accellera helps to 
> ensure greater efficiency.
> 
> 
> Now that Accellera has raised the bar, can the IEEE raise it 
> further and provide a better environment for incubating 
> standards development?  Until something changes, do not 
> expect the current trend to change.
> 
> 
> -Steve Bailey
> 
> 
> ________________________________
> 
> 	From: owner-vhdl-200x@eda.org 
> [mailto:owner-vhdl-200x@eda.org] On Behalf Of gmoretti@comcast.net
> 	Sent: Wednesday, July 06, 2005 5:01 AM
> 	To: vhdl-200x@eda.org
> 	Subject: Re: [vhdl-200x] Announcement
> 
> 
> 
> 
> 	
> 
> 	I would like to amend Evan's remarks.  The IEEE did not 
> give away the EDA standards to Accellera.  The Computer 
> Society did, and Accellera earned it.  The Computer Society 
> dismal management of the DASC has in fact been one of the 
> premier factors in the creation of the EDA Council within the 
> IEEE in a belated attempt to plug the dike after the valley 
> has been flooded.  Has any one ever considered how 
> interesting it is that DAC is sponsored by the Circuits and 
> Systems society and not by the Computer Society?  If the IEEE 
> society that owns DASC is so disinterested in EDA, is it not 
> logical that the IEEE would want to foster the develoment of 
> standards using hat is today the most effective forum for 
> such activity?  And, by the way, by not sponsoring DAC the 
> Computer Society is loosing the opportunity for significant 
> revenues that would have more than been sufficient to finance 
> DASC.  Perhaps the EDA Council will find a way to regain th e 
> leadership in the standards development  field, but in the 
> mean time tempus fugit and someone has to get the work done!
> 	Gabe
> 
> 		-------------- Original message --------------
> 
> 	
> 
> 		> Bailey, Stephen wrote:
> 
> 		> > All,
> 
> 		> >
> 
> 		> > At its Board of Directors meeting at DAC, 
> Accellera approved the
> 
> 		> > formation of a working group for VHDL. The 
> IEEE has provided Accellera
> 
> 		> > permission to create derivative works based 
> on 1076 as long as the
> 
> 		> > derivative works are submitted back to the 
> IEEE for standardization.
> 
> 		> > First, I cannot overstate why I believe 
> this development is good for VHDL.
> 
> 		>
> 
> 		> Well, I couldn't let this pass without 
> comment. This may or may not turn
> 
> 		> out to be 'good for VHDL', but it certainly 
> marks the end of the road
> 
> 		> for DASC. When 1647 goes over to Accellera 
> DASC will be finished, period.
> 
> 		>
> 
> 		> What I find extraordinary is that, over the 
> past couple of years, the
> 
> 		> IEEE has effectivel y given away the EDA 
> standards space to Accellera,
> 
> 		> without seeing fit to even inform, let alone 
> consult, the DASC members.
> 
> 		>
> 
> 		> I, for one, would like to know precisely 
> which individual at the IEEE
> 
> 		> has been responsible for this process, and 
> what mandate they had to
> 
> 		> carry this through.
> 
> 		>
> 
> 		> Evan Lavelle
> 
> 
> 
> ************************************
> This e-mail and any files transmitted with it are proprietary 
> and intended solely for the use of the individual or entity 
> to whom they are addressed. If you have received this e-mail 
> in error please notify the sender. Please note that any views 
> or opinions presented in this e-mail are solely those of the 
> author and do not necessarily represent those of ITT 
> Industries, Inc. The recipient should check this e-mail and 
> any attachments for the presence of viruses. ITT Industries 
> accepts no liability for any damage caused by any virus 
> transmitted by this e-mail.
> ************************************
>