I have seen most standards that have floundered and not gained
acceptance have been due to lack of active vendor participation. The
real advantage of an open standard (as opposed to openly available) is
that competitive needs can be resolved - thus leading to a better
standard. The standard process itself takes awhile - but that is
because you are forced to produce unambiguous documentation of syntax
and semantics. Most of the standards I have been involved with have
had the best senior level people from the companies involved. Yes
there are sometimes technical arguments that drag out - but I would
rather have this then a competitors application engineers ignoring my
requests for features.
-------------------------------------------------------------------------------
Kevin R. Grotjohn LSI Logic Corporation
Staff Design Engineer 5050 Hopyard Blvd, Suite 301
DEVelopment Optimization (DEVO) Pleasanton, CA 94588
Product-Development Technologies Tel : (925) 730-8811
krag@lsil.com Fax : (925) 730-8700
-------------------------------------------------------------------------------
> From owner-dcwg@eda.org Wed Dec 16 14:20 PST 1998
> X-Authentication-Warning: server.eda.org: majordom set sender to owner-dcwg@eda.org using -f
> From: dwp@aloft.micro.lucent.com
> Date: Wed, 16 Dec 1998 16:56:09 -0500 (EST)
> To: dcwg@eda.org
> Subject: Re: DC-WG: Proprietary ConstraintLanguages
>
>
> To all:
>
> I have some general questions concerning using a proprietary
> design constraint language and would like to understand a few things.
> I have not been an active participant, but I have been observing.
>
> 1. An open standard provides opportunity and competition in
> developing the best possible documentation. Where is the
> incentive for making the best possible documentation.
> For example:
> I have several VHDL books, some of them good and some of them
> not so good, regardless, I have a significant amount of choice
> in books, media format etc.
>
> 2. A proprietary language may not lend itself to innovation.
> If it becomes a standard, why would suppliers have to
> respond to customer demand since customers will be locked
> in?
>
> 3. Any synthesis, layout tool etc. should be able to
> associate design information/constraints with as many
> constructs in the source code as possible.
> e.g. a tool should be able to uniquely assign constraints
> to processes, statements, blocks etc. as defined by
> configurations, generate statements and hierarchy.
> However, if the constraint language does not support
> some constructs, would the customers be able to get the constraint
> language changed if they are already buying a tool that
> covers the shortfall?
>
> 4. There are many different open standards that define
> electronic systems (ethernet, MPGEG, JPEG, etc.), so
> why does anyone think that a good and evolving standard
> has to be proprietary?
>
>
> 5. Lastly, should consumers take this committee seriously
> or are the members of this committee taking advantage
> of complacent customers, and would like to thwart a
> free-market standard?
>
>
> David W. Potter
>