Print

Minutes of Weekly Meeting, 2010-11-15

Meeting called to order: 10:38 AM EST

1. Roll Call

Brian Erickson
Ian McIntosh
Peter Horwood
Brad Van Treuren
Tim Pender
Michele Portolan
Adam Ley (joined 10:47)

Excused:
Carl Walker
Patrick Au
Heiko Ehrenberg

2. Review and approve previous minutes:

10/25/2010 minutes:

3. Review old action items

4. Discussion Topics

  1. Report and comments from ITC
    General Comments
    • [Tim] What was attendance like at ITC this year?
    • [Ian] Apparently, attendance was up around 20% on 2009. It might be interesting to see what happens to attendance next year as they've booked the Disneyland Hotel, California, for ITC 2011. I'm not so sure that's a good idea: I guess they got a deal based on some expectation that families of attendees might take the opportunity to take a holiday at the same time. That might work for mainland US, but not so much for overseas people.
    • [Brian] It will likely depend on how the industry performs over the next year. Many companies are still reluctant to send people to conferences.
    • [Ian] I'm sure that's the case. I guess it's good news that attendance was up this year, so we just have to hope that's indicating a recovery that will continue.
    • [Brad] Part of the motivation behind taking ITC to San José and Austin was to try to drum up business from the silicon companies, so that seems to be a move that's going against that strategy.
    • [Brian] The other thing I'd say is that board level test has really fallen off the agenda at ITC.
    • [Ian] Yeah, I guess I was a bit disappointed at the amount of IC-centric material being presented. I know IC DFT is important, and it may just be my perception, but I feel there are more people involved in designing and testing boards and systems than for ICs.
    • [Brad] I think 2005 was the last time there was a lot of board test representation at ITC. At that time Ben Bennetts was working with Bill Eklow to try to get a board test track, but there wasn't really enough material around for that to be viable.
    • [Tim] Maybe things like the Board Test Workshop is stealing people?
    • [Brian] But that's maybe only 25 people.
    • [Ian] That's not a lot.
    • [Adam] Usually, it's closer to 40, and that's typical even for ITC workshops which are probably the best attended. Only 3D Test did better and had around 80.
    • [Adam] I think the ITC committee are well disposed to accepting board test submissions. If we want to see more SJTAG papers at ITC then we need to sharpen our pencils and get the submissions in.
    • [Ian] And there isn't a lot of time to do that - Jan 21 is the deadline for submissions, and as a group we don't have anything on hand to use. I guess Adam is right, but there's a kind of vicious circle - if there's little board or system test material on offer then you don't attract much of that the following year, etc.

    Poster
    • [Ian] As a first time attendee, it's maybe difficult for me to judge how things went. It was busy, but I wasn't swamped. People were generally 'browsing' round the all posters, so there was steady flow of people. I believe the exhibition booths also got business on the back of people circulating round the posters.
    • [Ian] Some of the questions reinforced the opinions we'd formed from the survey regarding the lack of awareness in the wider community both of the developing standards in general and SJTAG in particular.
    • [Ian] We were allowed to leave the posters up all week, so I was able to man the poster over the next few lunchtimes too, which also got a little more attention. I left the leaflets out by the poster so people could help themselves when I wasn't around, and only one was left from a bundle of 20.
    • [Ian] There was one kind of trivial takeaway I had from this, if I can illustrate.
    • {Photograph of Poster shared}
    • [Ian] We pretty much filled the whole area allocated to us, so I was left wishing that I had left a gap in the bottom of the middle row so I could open my laptop without obscuring any of the poster. As it was, the laptop stayed pretty much shut all evening. I think we may have been one of the few posters to have a laptop on hand anyway.
    • [Brad] I'm interested in what sort of questions you were getting asked; was there interest in the primitives?
    • [Ian] Yes there was, but I was getting enquiries from a broad range of perspectives. Some folks started out by asking about the board architectures while others picked up on the primitives. I think some of the questions on the primitives were coming as a result of us only presenting a few of the slides so there were maybe some gaps, but mostly I think people were following the slides straight off. It think it was good to have the "board track" and the "primitives track" both on show as people could pick on either.One person had a view of JTAG that was completely production test-centered and initially couldn't see why he'd want JTAG at the system level at all.
    • [Ian] Bill (Eklow) and Rohit (Kapur) both felt that the joint standards poster worked well, so the same thing will be set up for next year. Rohit is also to be given his own table next year too.
    • [Tim] Have you noticed any additional website traffic since ITC?
    • [Ian] Yes. I haven't really checked in the past few days, but immediately after I got back I was checking at there had been a lot of traffic from the Convention Center itself. WiFi was available throughout, so people were obviously browsing from their laptops. There was also traffic from IPs in Austin that could possibly be one or more of the hotels. It's a pity that none of that interest seems to have resulted in any signups though.

    TTSG Meeting
    • [Ian] There are two related things to pass on. While we're not a formal WG under the TTSG we're not obliged to follow their rules, but we have generally chosen to do so, which means we probably need to look at updating our Policies and Procedures.
    • [Ian] The first thing is that Officers of the group should be reaffirmed in their positions annually. This doesn't meant that the group needs to go through a full election process, but that there should be a ballot to confirm that the group is satisfied with it's leadership.
    • [Ian] The second point is that if anyone feels unable to discuss concerns about their group's leadership within the group itself, then they can take the matter to Rohit (or whoever may subsequently take Rohit's role) in confidence and Rohit will investigate.
    • [Ian] It was commented that if a group feels they need to replace a chair that they'd need to have a replacement in mind, and sometimes that might not be easy.
    • [Ian] You can infer that this comes on the back of a problem perceived in one of the groups. The suggestions were mainly coming from Adam Cron; I'm not really aware of his background here or what his role in the TTSG is?
    • [Adam] I don't think Adam had any particular agenda here, I think he was just saying "We've seen this happen, so maybe this is way we could handle it." Adam is a past chair of TTSG and has been involved in a lot of the working groups.
    • [Ian] OK, that makes sense. Rohit seemed to pay a lot of attention to what he had to say and I just wondered if Adam was providing some TTTC or SA oversight.

    Fringe Meeting
    • [Ian] I was disappointed by the turnout, given the amount of interest that had been expressed through the week, so we ended up with a handful of our regulars plus Ken Parker.
    • [Ian] After I described the current status of SJTAG, Ken suggested that maybe SJTAG needed a "900lb gorilla" to take up SJTAG and show the need for it. Cisco was P1687's 900lb gorilla and at one time Lucent looked to be doing the same for SJTAG.
    • [Brad] I think there are cultural differences there. Lucent was heavily involved but Alcatel isn't so keen on getting involved.
    • [Ian] I had a similar kind of discussion over lunch with Michele: Here in the UK we see most companies aligning with the IET so IEEE activities don't get much of an airing, so I guess a predominantly French company like Alcatel may be dismissive of something they see as primarily American.
    • [Brad] I think it maybe goes deeper than that. A lot of companies don't understand the value that can be gained from standardization. Some don't even see the value in 1149.1 even though they use it.
    • [Ian] Another suggestion from the meeting was that not having a PAR might be putting some people off, and that SJTAG looks like group that isn't making progress. I think some people are holding off from participating until they see the formality of a PAR.
    • [Brad] I'm not sure I agree, for example 1581 has had a PAR for a long time and still sees those issues.
    • [Tim] Is there a requirement to complete the standard within a certain time from the start of the PAR?
    • [Ian] Sort of. A PAR is initially for a period of 4 years. If you have to request an extension then that will be for 2 additional years, and you can re-extend; 1581 has been running for almost 10 years.
    • {Post meeting note: During the TTSG Meeting, it was requested that the WG Chair's regular reports to Rohit should include a statement on the expiry date of the group's PAR. This was to ensure that expiry dates were being tracked.}
    • [Ian] We've been going for 5 years now, so I guess if we'd gone to a PAR in the first year or so, we'd be looking to extend now, so I guess it was right not too rush in too quickly.
    • [Brad] In my experience, the working groups that have been most successful are the ones that had done their homework before going to PAR. Those that were more fuzzy in their understanding have found things more difficult.
    • [Brad] Back when Ben was the chairman, we discussed when is the right time to go for a PAR; we want the status but need to know where we were going.
    • [Ian] It's a difficult call to make. On one hand there's a danger of over analysing, on the other there's the issue of not being properly prepared.
    • [Ian] During the meeting we mentioned the idea of a core standard and extensions, and it was suggested that having a roadmap for that might have an appeal to some of the present observers. They might want to get involved in a particular extension or concentrate on the core; the parts that seem most relevant to their own experiences. I guess this would lead back to the idea Brad presented early on, of a core group and several subgroups.
    • [Brad] Yes, that was what I envisioned but we didn't have a way to divide out into subgroups then.
    • [Ian] Nor the critical mass of membership to support that, and that's maybe still the case. There are a lot of things we identified through our past discussions that we labelled as something that looked like an extension or part of the core, but I'm kind of struggling to pull all that back together now.
    • [Brad] Well, we need a coherent story. That was part of the motivation for that presentation we were preparing, and didn't finish. In completing that, we should be able to go back over it and start carving up what will be in our core and what will be an extension.
    • [Ian] In recent weeks we had other things we needed to concentrate on, but I did intend coming back to that presentation in our meetings.
    • [Ian] There are some industry sectors not represented in our group, and there was some consideration of whether we should be trying to lure some of those in. In the past we tried, unsuccessfully, to get the EDA people involved, but there's also things like the automotive sector and maybe consumer electronics.
    • [Brad] I'm not convinced that consumer electronics will see any kind of ROI out of SJTAG. They're very cost sensitive items, high volume, and probably too small to see any benefit. Maybe down the road, with products with multiple cores there may be some future.
    • [Brad] With respect to automotive, we're not really in the distributed systems area the way they are.
    • [Ian] Maybe not, but support for distributed systems was a popular item that came back in our survey results. Of course, we don't know that the scope envisioned there is necessarily the same as automotive would expect. And distributed systems could well be one of our extensions.
    • [Brad] That's right.
    • [Ian] In a similar vein, another thought was whether it was worth trying to hook in a University somehow, get someone from academia involved, which might spawn interests through other channels.
    • [Brad] Due to the economy, I think many will be reluctant unless there's some industry funding tied in. My brother, an Associate Dean at Baylor, runs into this everyday, although that is in mechanical engineering. But I think it's the same everywhere; they're having the same issues in Tallinn and I think Michele's old university is the same.
    • [Michele] Yes.
    • [Ian] I suspect you're right, but it doesn't cost anything to ask, and I know my old course head founded a university spinoff, SeeByte, which has a product related to remote diagnostics, so it's not totally offbeat.
    • [Brad] You're right, it doesn't cost to ask. One thing I am surprised at is that people like NASA haven't shown an interest.
    • [Brian] Well if we can get to a PAR, then maybe we can market it a bit more.
    • [Ian] I agree, but it maybe a bit of a "chicken and egg" situation.

5. Schedule next meeting

Regular Meeting: November 22nd

Schedule for November 2010:
29th

6. Any other business

The motion on signal naming conventions from the meeting of May 3rd remains tabled.

7. Review new action items

None.

8. Adjourn

Brian moved to adjourn at 11:44 AM EST, seconded by Peter.

Respectfully submitted,
Ian McIntosh