ALF IEEE kickoff meeting
Feb. 20 2001 @ LSI Logic, Milpitas

Attendees:
	Simon Favre	Monterey Design Systems
	Greg DuFour	Mentor Graphics
	Jay Abraham	Silicon Metrics
	Pierre Girouard	LogicVision
	Liz Chambers	Tera Systems
	Cynthia Parrish	Aristo
	Wolfgang Roethig NEC
	Tim Ehrler	Philips
	Alex Zamfirescu	Alternative System Concepts, IEEE DASC member
	Nancy Nettleton	SUN Microsystems
	Joe Daniels	Accellera
	Antonero Carvalho LSI Logic
	Vinay Srinivas	Sequence Design
	Li Pen Yuan	Avant! Corporation
	Arun Balakrishnan NEC
	Steffen Rochel	Simplex

Next meeting
------------
Tue March 27 at NEC Santa Clara, from April onwards
we try to co-locate with OLA meeting for convenience.

Discussion on ALF presentation (see powerpoint file)
------------------------------
Alex: must define SCOPE and PURPOSE
must clarify whether it defines language, format, datamodel, API

Nancy: Strategy and tactics
Architecture and technology

Ballot: > 50 voters amongst community: companies, academia
70% majority required, negative votes must be addressed
several rounds of ballot may be needed.

Alex: Are there competitive proprietary formats?
Wolfgang: Synopsys .lib and Cadence TLF, both companies
were invited to this meeting

Nancy: Why could a PAR not be approved?
Alex: If we fail to include pertinent information

ALF-OLA releationship
---------------------
- Do they have some data in common? (answer: yes)
- Are the versions synchronized? (answer: no)

Only existing IEEE standards can be mentioned in PAR as
"related standards". Mentioning a "related standard" imposes
to define a system for interoperability as normative part
of the standard. In practice, this would mean to standardize
the equivalent of OLAWorx. Also, only the approved 1481 can
be mentioned, not OLA.
Since ALF standardizes only on syntax and semantics of the
language, it is appropriate not to declare 1481 a related
standard in the PAR. On the other hand, the text of the ALF
specification may refer to 1481, wherever the ALF semantics are
equivalent and well-explained in 1481 (e.g. DELAY, EARLY, LATE).
Precedence: SDF and Verilog. Both standards are independent,
but SDF refers to Verilog in the specification text.

The subject will be discussed again in the OLA meeting the next day.
The recommendation is to keep the ALF/OLA relationship informal
and have enough individuals on both committees to ensure that
the contents of the specification are compatible with each other.

Alex: upcoming standard ".net" (see ECMA) may make other
procedural interfaces unnecessary

A.I. Discuss ALF/OLA relationship in OLA meeting - Dan Moritz
DONE (see OLA meeting minutes)

A.I. Generate the final power point presentation for DASC meeting,
based on this meeting's conclusions - Wolfgang
DONE - see powerpoint file

A.I. find pointer to more information on EMCA and its possible
implications on ALF and OLA - Alex