From: Don Organ
Sent: Friday, October 18, 2002 11:03
AM
To: STIL. 4 (E-mail)
Subject: stds-1450.4: Minutes from
our STIL.4 teleconference call on 10/18/2002
Dave
Dowding
Ernie
Wahl
Tom
Micek
Jim
Oreilly
Don
Organ
Larry
Moran
Jim
Mosley
Dave to send out
summary from the meeting at ITC.
Verbal summary:
Synopsys sponsered a STIL summary on Monday afternoon. Went really well and was
well attended. Tony Taylor presented a timeline for the various working groups.
Greg Maston reviewed some of the difficulties of STIL adoption. Various other
presentations.
Issue #10 from the
Flow Control issues list (of 5/14/2002)
10: Where is the Pattern Exec called from?
(Calling it from low in the hierarchy - such as from within a TestMethod, allows
more control and flexibility. However, calling it from high in the hierarchy
would allow an entire sub-flow be grouped under a single Pattern Exec - which
could be desireable when creating speed-binning flow - where 2 subflows differ
only in the Category/Selector. Or do we consider breaking it
apart?)
Ernie - a question regarding variables and
categories.
Jim - could allow
the Category to be identified within a FlowNode and optionally also from within
a PatternExec - the lowest level (most specific) "wins". We'd need to precisely
define scoping rules for the Category.
Ernie - concerns
about the automated maintenance of such a system.
Don - let's redefine
the question: Shall we consider breaking open STIL's PatternExec - and maybe
using its components (such as Categories) separately at different levels of the
Flow Model?
There was consensus
on this point.
Ernie - I have
some concerns about how we define our use of Categories - I want to be able to
do histograms without creating Categories for each step in the histogram. Also,
would like to have spec variables values that are not constant - but are
dependent on other specs.
Larry - use
enumeration for small number of values, use an algorithm for a large number of
values - don't use Categories for a histogram.
Don would like to
contact individual working members directly while working on the draft document.
There were no objections to him doing so, provided summaries are
distributed.
Jim - it would be
useful to develop a collection of test-program fragments to illustrate different
test requirements and techniques.
Dave will send out
the summary of the ITC meeting, along with a start on a glossary and a start on
a issues document. We'll start out our next meeting by reviewing
that.
Dave will collect
phone #s and emails - as a directory.
-DVO-
---------------------------------------------------------------------------------
Don
Organ
Inovys Corporation
---------------------------------------------------------------------------------