RE: FW: [vhdl-200x] 1076 & Entity balloting

From: Bailey, Stephen <SBailey@model.com>
Date: Wed Jun 23 2004 - 15:44:34 PDT

Thanks Evan. It is important to provide facts and specifics to back up any argument being made. You shouldn't assume that everyone knows what you are referring to.

Yes, there was definitely a tug-of-war going on within Accellera in this context. I have not participated in FVTC directly. So, the observations below are not from an "insider." But, the results are the results regardless of the specifics of how it happened.

1. PSL 1.01 was approved by the Accellera Board.
2. PSL 1.1 was approved by the Accellera Board.
3. PSL 1.1 != SystemVerilog 3.1a Assertions.
4. PSL 1.01 and 1.1 != ForSpec
5. PSL 1.01 and 1.1 != OVA
6. In fact, SVA 3.1a != ForSpec and SVA 3.1a != OVA.
7. PSL 1.1 is closer semantically to SystemVerilog 3.1a assertions than what PSL 1.01 was. But, it is my understanding that the formal semantics specification of PSL was preserved and that SVA now benefits from that work.

There were many people that thought that a single assertion/property language would be best and sufficient. The alignment process that the Accellera Board "dictated" was directed to get SVA and PSL closer to the ideal of a single language. For various reasons (some technical and some related to syntactic sugaring), one language was not the result. However, a common semantic base was significant progress towards the goal. (Think of the benefits in lower cost of training/cross-training, lower implementation cost to tool vendors and that each language has equivalent capabilities in the intersection of functionality that both groups found important.)

So, if organization entity membership is the evil that this example was meant to demonstrate, why:

1. Isn't PSL simply SVA? (Or vice versa)
2. Why did the Accellera board approve PSL 1.01 and 1.1?
3. Why didn't the Accellera board dissolve the FVTC or instruct the SV working group to narrow its scope to not include assertions?
4. Why has Accellera donated both SV and PSL to the IEEE? It could have easily squashed one or both.

If there's any lessons to be learned by the Accellera/FVTC experience it is that all organizations with a vested interest need to participate and have their voices heard. If 4 companies failed in one domain and succeeded in the other, their success was only possible because the organizations responsible for their failure in the one did not participate in the other.

To further the legislative analogy a bit further: If making legislation is like making sausage, then the same applies to standards. There will always be give and take and some of it won't be pretty to see. (The "strains" you refer to.) Companies will have agendas that they will be promoting. They should only be successful if they can get a majority (or, in the IEEE, 75% of 75%) to agree.

If you think individual membership eliminates this problem, you are very naïve. As I have said before, it is currently easier for an organization to dominate the outcome of a WG by purchasing the $40 membership of enough of their employees. And, individual membership makes it easier to camouflage what is happening as the assumption is that individuals are making their own decisions and not just carrying water for their company.

Worse, individual membership can result in individuals having agendas with no market relevance or negative market relevance. Well meaning engineers can be notoriously stubborn in holding out for the holy grail of technical solutions no matter the cost. The interplay between individual participation and organizational approval/oversight provides a check against these issues. On the other side of things, if organizations overplay their approval/oversight role, then that will drive off the individual participants (even some who are working for an organization will quit or request re-assignment) which will make it difficult or impossible to complete the technical work.

In 1076, it wasn't that long ago that we experienced an effort to dominate a WG vote. One company attempted to defeat the advisory vote of using PSL as the basis for VHDL's property and assertion capabilities. If these votes had not been ruled out-of-order/ineligible for process reasons, then that vote would have failed.

In conclusion, your example does not prove that organizational entity membership is inherently inferior to individual membership. In fact, the results from Accellera prove that organizational entity membership can result in coordination of standards and different standards that overlap in scope. It also shows that while a single entity (or even 4 like-minded entities) can influence, they cannot dictate the outcome.

On the other hand, I have provided a concrete example of how individual membership was used in an attempt by an organization to control a WG. Although that attempt failed, if the company involved fixed the technical reasons for the disqualification of their votes in time for future votes, it will become a much more difficult problem to fix. I'd much rather that companies have a single vote and be required to persuade others.

-Steve Bailey

> -----Original Message-----
> Bailey, Stephen wrote:
>
> > Evan,
> >
>
> > There is a need for explanation on your comment WRT FVTC and
> > Accellera. There are currently more tools that support PSL than
> > support SystemVerilog Assertions. So, what do you mean by "the
> > entities simply ignored the work and did what they wanted anyway."
>
> The technical work that the FVTC did resulted in PSL 1.01,
> which was based on Sugar 2. When they'd essentially finished
> this work, the Accellera board informed them that it was
> unsuitable for SystemVerilog, and that they would have to
> converge it with OVA, with no technical justification
> whatsoever. The result is PSL 1.1. For the detail, see the
> third item in
>
> <http://verificationguild.com/modules.php?name=Forums&file=vie
> wtopic&t=301&highlight=ova&sid=517e939baa98c59eb4e88f804121e469>.
>
> This is a perfect example of the strains that would exist in
> a group which was entity-controlled, but in which the
> technical work was done on an individual basis. 'Individual
> experts' do not have the same agenda as EDA vendors.
>
> > Also, if you believe that the FVTC and Accellera failed
> with PSL, then
> > how is it that it failed and why is that failure related to
> > Accellera's organizational entity membership?
>
> I believe that the FVTC did a very good job. The result was,
> however, incompatible with the political goals of Accellera,
> following the OVA donation to Accellera. Anybody who
> disagrees with this should study the link above carefully and
> provide a detailed rebuttal.
>
> A WG which is entity-based will inevitably be driven by a
> political agenda. This may or may not be a good thing, but I
> personally don't believe that it's the role of the IEEE to go
> down this road. If we can't get the funding to carry on on
> the current basis, then I don't believe that the work is worth doing.
>
> Evan Lavelle
>
Received on Wed Jun 23 15:44:40 2004

This archive was generated by hypermail 2.1.8 : Wed Jun 23 2004 - 15:44:53 PDT