Minutes of Weekly Meeting, 2011-03-14

Meeting called to order: 11:09 AM EDT

1. Roll Call

Ian McIntosh
Heiko Ehrenberg
Tim Pender
Brad Van Treuren
Brian Erickson
Richard Foster (joined 11:57)
Peter Horwood (joined 12:02)

Carl Walker
Patrick Au
Eric Cormack

2. Review and approve previous minutes:

03/07/2011 minutes:

  • Draft circulated on 03/07/2011.
  • No amendments noted.
  • Insufficient attendance for approval.

3. Review old action items

  • Adam proposed we cover the following at the next meeting:
    • Establish consensus on goals and constraints
    • What are we trying to achieve?
    • What restrictions are we faced with?
  • All: do we feel SJTAG is requiring a new test language to obtain the information needed for diagnostics or is STAPL/SVF sufficient? see also Gunnar's presentation, in particular the new information he'd be looking for in a test language
  • Ian/Brad: Condense gateway comments and queries into a concise set of questions. - Ongoing
  • All: Forward text file to Ian containing keywords from review of meeting minutes. - Ongoing.
  • Ian: Publish the Newsletter as soon as possible. - COMPLETE
  • Carl/Brad: Get annotated keyword worksheets to Ian by Wednesday Close of Business. - Ongoing
  • Ian: Revisit and rationalize the 'Q' markers in keyword list. - COMPLETE

4. Discussion Topics

  1. Identification of key "Take Away Points"
    - Progress on review of past minutes
    - Review of down-selected keywords
    • [Ian] I went over my list to reassess the 'Q' entires I had, and sent the list back out.
    • {Ian shared the keyword spreadsheet}
    • [Ian] You'll see that all the 'D' entries have gone as I've updated these, so it's just the 'Q's we need to look at.
    • Dangerous
    • [Ian] I was thinking of "dangerous conditions" for JTAG operations; "dangerous" by itself isn't very meaningful, but it might be useful in specific contexts.
    • [Brad] I think we need to see this in the context it was used.
    • {Queried in site search engine}
    • [Ian] The search results for "dangerous" don't really provide much value.
    • [Heiko] There are probably other keywords that would return the same results but with a better context.
    • [Ian] Yes, these contexts aren't helpful. This looks like a toss to me.
    • [Brad] Agreed.
    • {Toss}
    • Decision
    • [Ian] I was thinking of cases where a specific decision had been made.
    • {Queried in site search engine}
    • [Ian] The search results for "decision" don't really seem to identify any relevant key issues.
    • [Brad] The one that look relevant could be covered with "embedded". I'd be fine with striking "decision" from the list.
    • {Toss}
    • Delegation
    • {Queried in site search engine}
    • [Ian] Looking at the search results, I've got a feeling that this one is relevant. It covers the case where you might delegate some control to a lower level.
    • [Brad] Could it be covered by "model".
    • [Ian] I don't know that "model" would work; we don't always mention a model and if I were searching for this subject I'd find "delegation" a more appropriate search term.
    • [Brad] Looks like a "keep".
    • [Ian] Any objections?
    • {none voiced}
    • {Keep}
    • Distributed
    • [Brad] I think this is going to be one that is relevant to us.
    • {Queried in site search engine}
    • [Ian] "Distributed systems" is what I immediately thought of, but we also have "distributed repositories". On it's own, "distributed" probably doesn't mean much.
    • [Ian] My inclination would be to toss "distributed" and to keep "distributed system".
    • [Brad] But "distributed repository" will also relevant, especially for the embedded case. It's an aggregation, so that it looks like one repository for the benefit of the tooling.
    • [Ian] The argument being that "distributed" on its own is not descriptive enough.
    • {Exchange "Distributed system" and "Distributed repository" for "Distributed"}
    • EBTW
    • [Ian] It's more than EBTW I'm questioning here; there's BTW, ITC, etc., too. Is it meaningful to search references to these conferences?
    • {Queried in site search engine}
    • [Ian] "EBTW" only returns a few results, "BTW" returns quite a few more. These are mostly all references to papers and presentations.
    • [Heiko] I think it’s worth keeping those.
    • [Brad] It may short cut us to finding those links. I think we've been quite good at citing those papers properly in the minutes.
    • {Keep}
    • Economic
    • {Queried in site search engine}
    • [Ian] Neither of the references are helpful. Nor are any of the hits for "economics" or "economical".
    • [Brad] I'm inclined to say we remove it.
    • [Ian] OK. I'm a little surprized because 'economics of test' is a popular discussion subject these days.
    • [Brad] Value proposition would be more appropriate as the term we have used.
    • [Heiko] I agree that there are probably better and more specific keywords, such as "ROI" which is in one of those search results.
    • [Brad] Yes, Return on Investment.
    • {Toss}
    • EDA
    • [Ian] A large number of search engine hits for "EDA".
    • {Queried in site search engine}
    • [Ian] I had a view that EDA had some relevance in our discussions.
    • [Brad] Some of this was with regards to modeling.
    • [Ian] I wonder if there are better keywords to identify the relevant issues. "EDA" returns a lot of references to action items in our minutes, but not much else.
    • [Heiko] Or at least a phrase, such as "EDA support" or "EDA tool".
    • [Brad] I don't see a lot of significance of those search results for "EDA"; there's not a strong enough context to keep this.
    • {Toss}
    • Education
    • [Ian] We've referred to the role of the White Paper for education, but I wonder if it is too general to be a useful search term.
    • {Queried in site search engine}
    • [Ian] Would you use that in a summarizing minutes?
    • [Brad] I think it's unlikely that people would use this as a search keyword.
    • Embedded
    • {Queried in site search engine}
    • [Ian] The question is whether this should be used as a stand-alone keyword or rather as a set of key phrases, such as "embedded system" or "embedded test".
    • [Brad] Or even "embedded software".
    • [Ian] There are a lot of search results for "embedded".
    • [Brad] I'm seeing at least six different contexts here.
    • [Heiko] This is probably another instance where you would 'AND' together multiple search terms in order to get relevant search results.
    • [Ian] OK. There are common themes, but there's more than just one or two of them.
    • [Brad] I'm inclined to keep "embedded" as a keyword on its own.
    • {Keep}
    • Emerged
    • {Queried in site search engine}
    • [Ian] Not sure why I had that one as 'Q' in the first place.
    • [Ian] Looking at the line number (2021), this would have been in your block, Brad - Any thoughts on this?
    • [Brad] I'd say throw it out, I can't think of why I would have considered it as a potential "keep".
    • [Heiko] Maybe thinking of "emerging technologies" or "emerging standards"?
    • [Ian] Yes, I found it easy to imagine a relevant context as you look at the word, but find there is none once you do a search.
    • {Toss}
    • Excluded
    • {Queried in site search engine}
    • [Brad] That could possibly be a word that could help us refine what we are interested in.
    • [Ian] Neither of the search results we get for "excluded" seem very useful.
    • [Brad] No, not at all. I agree, seems to be another one we should toss.
    • {Toss}
    • Extend
    • {Queried in site search engine}
    • [Ian] "Extension" we have already marked as a "keep". Do we also need "extend" and/or "extended"?
    • [Heiko] I'd say if you summarize something, you'd probably use "extension" as a keyword instead of "extend".
    • [Ian] Yes, I can see that.
    • [Brad] I'm fine with removing "extend".
    • {Toss}
    • Failsafe
    • {Queried in site search engine}
    • [Ian] only one search result, and that relates to an I2C application.
    • {Richard joined; missed the US changing to DST}
    • [Brad] Yes, it can be removed. That was discussion of some of the diverse ways people use JTAG, not something we want to recommend.
    • {Toss}
    • Flows
    • {Queried in site search engine}
    • [Brad] "Flows" may be too general a search term.
    • [Ian] We only get two, diverse, search results.
    • [Brad] I don't see any significance with the search results we get: I'd be fine with striking "flows" from the list.
    • {Toss}
    • FMA
    • {Queried in site search engine}
    • [Brad] I thought this would be relevant even though there are only two hits.
    • [Ian] I think we tended to focus more on the "Root Cause Analysis" than on the "Failure Mode Analysis";
    • [Brad] I think it is still a useful keyword.
    • {Peter joined; missed the US changing to DST}
    • [Ian] Do we keep the acronym rather than the full term?
    • [Brad] That is a good question. People seem to mostly use "FMA".
    • [Ian] It's maybe more complicated because we tend to refer to FMEA (Failure Modes and Effects Analysis) or even FMECA (Failure Modes, Effects and Criticality Analysis).
    • [Heiko] Is there any reason we can't keep both the acronym FMA and the term Failure Mode Analysis?
    • [Brad] Considering that some people may not know exactly what FMA stands for it may be worthwhile keeping the expanded term. It may help the person doing the searching.
    • [Brian] That raises the question of whether we should then edit the minutes to match the search terms?
    • [Ian] Well, we're really searching the summaries, which don't need to use exactly the wording in the minutes.
    • [Brian] I thought we were planning on linking to the minutes?
    • [Ian] Yes, but we're sort of abstracted by the summaries: They are what we'll search, and they will reference the minutes they relate to.
    • [Brian] OK, yeah, that'll give us the dates so we can find the original texts.
    • [Ian] So we use the full terms instead of acronyms?
    • [Brad] That is what I propose.
    • {Exchange "Failure Mode Analysis" for "FMA"}
    • [Ian] This gets us through the list of keywords we had reviewed so far (not including the set of keywords assigned to Carl). So, we'll need to get working on additional keywords.
    • [Ian] I'll work on the block of the first 1000 keywords, since I had started looking at those anyway. If we get the block from Carl and I get through my additional block, that likely gives us an additional set of 'D's and 'Q's to review next time. {ACTION}

5. Key Takeaway for today's meeting

  • [Heiko] May need to use multiple search terms to find relevant summaries.
  • [Brad] Decision: Expand acronyms as keyword phrases.

6. Schedule next meeting

Next Meeting:
March 21st (11:00 AM EDT, 3:00 PM GMT)

Schedule for March 2011:

7. Any other business


8. Review new action items

  • Ian: Review first 1000 words on keyword list.

9. Adjourn

Heiko moved to adjourn at 12:16 PM EDT, seconded by Brian.

Thanks to Heiko for supplying his notes from this meeting.

Respectfully submitted,
Ian McIntosh