Process for resolving the new issues from ballot re-circulation

- Tony proposed to the Working Group that the Group work through the last
set of comments received (from both positive and negative balloters),
redraft to those reviewer's comments, and re-circulate to allow those
negative balloters to recast their vote. However, the negative
balloters haven't had a chance to review yet, so we must first solicit
these comments from these balloters.

We also have comments from one positive balloter - Rohit Kapur submitted
comments to an affirmative vote. Also Don Organ submitted several
comments to the current draft, and attended this meeting, so these
sets of issues were reviewed.

Rohit's comments

Peter and Greg had a conference call with Rohit to review his comments
- at least the technical ones. Greg distributed his summary of this
discussion to this group.

- First issue is a request to generate an index. Tony proposed to not
go to this effort, as electronic copies (pdf) of the spec are searchable
and more effective than most indexes. WG agreed.

- Rohit is unclear about the term "family of patterns". Tony proposed
to clarify this with Rohit rather than doc changes.

- One editorial item is a typo and will be fixed.

- 3 technical issues were raised about scanchain description
capabilities. Greg summarized that these issues were discussed with
Rohit and he accepted the status of no change to these sections,
although if other issues were raised about this then he requests
that a complete review of functionality be performed. There have
been no other comments about this section.

- Rohit's last issue is about a collision of an example used in both
p1450.1 and p1450.6, and requests a change to the example sufficient
to differentiate the two. Tony proposed 3 changes to the annex
containing this example: 1. change title of annex to reference
'allowinterleave' , 2. change structure of example from 1
concatenated chain to 2 parallel chains, and 3. identify that this
solution is only one of several options, and add a reference here to
.6.

- One of Rohit's other technical issues was regarding complex scan
cells, and correlation between .1 and p1500 naming. Rohit identified
that he will follow-up with Jason Doege here to see if there is
something to be done here.

Don's comments

Don's first issue was with the concept of "type-safe" or
"type-consistent" expressions, where the expression contains SI unit
references that are of similar type (like "seconds" and "volts"). The
foundation of the current behavior (in dot-0) is that there is no
requirement stated for consistent unit application in an
expression. During the discussion Don identified that the dot-1 spec should
not advocate "type-unsafe" expressions either. The Working Group
agreed that examples should be presented as type-consistent. Tony
indicated concern about words that may be present around this issue,
and Greg identified that whatever words present should be minimized in
the spec, otherwise the spec ends up advocating a particular
perspective that is not necessary at the language level.

Tony asked whether reals are 'typed' here, and the response was that
reals have a type of their own.

Don's second issue is with the presentation of booleans. His concern
is that the current presentation is someplace midstream between
whether booleans are a hard-type (consisting of defined entities
"True" and "False"...) or whether they are just an interpretation on
a real value (and therefore not "typed" at all). After some discussion
this was tabled, with Greg's sense that the current document is
placing more notions of "type" on these constructs than appropriate
because the intent - at this time - is not to have this as a distinct
type.

Don's third issue at this time is with the definition of integer
expressions (and all expressions). For example, table 5 identifies
that an integer expression cannot contain SI units, yet it's easy to
create integer expressions that do (esp. when the units cancel). After
some discussion on this, it was agreed that a group would get together
to work the right-hand columns of table 5 in particular, and possibly
add some definitions around how expressions consist of operations of
expressions in order to allow appropriate constructs.

other issues

Tony identified that there will be a series of meetings occurring in
Austin, TX, Jan 24-26, to discussion STIL/STC collaborative efforts
and how to advance these efforts in the industry. All STIL people -
and others - are invited to attend these meetings.

Tony also identified that there has been some positive proposals for
creating an analog standard PAR, with a person at ST interested in
chairing this work. Greg recommended putting a call out on the STIL
reflector to identify interested parties, and form a preliminary group
to try to establish a PAR scope for the effort. Jose identified his
company would have very strong interest, and potential constructs
ready to offer, to such an activity.

Meeting ended at 11:35