< User Survey 2009 Results 2009 Survey, Section 3>

Introduction


This survey was conducted at the end of 2009. It followed on from a previous, simpler survey that ran during the Summer of 2008: During the preparation of that earlier questionnaire, which was essentially sent out unsolicited to our contacts list, we realised that there were details that we wanted to elicit but felt unable to do, as it may represent too much of a time imposition on the recipients.

The 2009 survey was run on an "opt-in" basis, with the recipients having previously indicated willingness to take part.

SJTAG is not simply an engineering matter, so the survey covers apects of cost and management as well as technical questions. We understood that many of the questions may have fallen outside of the areas of responsibility for some of the respondents, or that some questions may broach on subjects or techniques that the respondent had no knowledge or experience of, so we asked that people only answered questions where they felt comfortable providing an answer. It is perhaps notable that despite this, the vast majority of people who took this survey elected to answer almost every question.

 


 

Section 1 - About you...


1.1 to 1.3

These questions requested personal details and, in the interests of privacy and confidentiality, the responses are not presented.

 


q1-4.png

1.4 - How would you describe your present role within your organisation?

  1. Product design/development
  2. Test design/development
  3. Manufacturing/Operations/Product Support
  4. Training/Education/Consultancy
  5. Management/Administration
  6. Tool Vendor (EDA/JTAG/DFT)
  7. Device Vendor
  8. Other (please state)

The largest group represented in our survey were those involved in test development: This doesn't really come as a surprise since these are the people most likely to have expressed interest in SJTAG. The only group we didn't get a response from was the device vendors. Although disappointing, it is probably not unexpected.

 


q1-5.png

1.5 - What is your relationship with the SJTAG initiative?

  1. Member of core group
  2. Member of extended group
  3. Would like to be a member of the core group
  4. Would like to be a member of the review group
  5. Would like to be a member of the extended group
  6. Not interested in joining a group

The largest proportion of responses was from people already affiliated with SJTAG in some capacity. However, with some 29% expressing an interest in becoming more closely involved then that is something that the group needs to be encouraging, and it is a pleasure for the group to see this high a percentage of interest.

 


Section 2 - Your prior knowledge of System-level JTAG...


q2-1.png

2.1 - Have you read the SJTAG White Paper?

  1. No
  2. Yes, I read the original White Paper (v0.4) a few years ago
  3. Yes, I read the White Paper on the SJTAG wiki

In reviewing this, we realized that we did not offer an option for anyone to indicate that they had both versions of the White Paper. We have to assume that anyone who had read both chose to indicate that they had read the later, wiki-based version, although there is some risk in that assumption.

It is worrying that many people have still only read the older v0.4 edition of the White Paper. While it still makes good points, our thinking, especially on the Use Cases for SJTAG, has developed considerably since that document was released, so those people will be missing a lot of new information and may be taking away outdated and increasingly irrelevant messages about SJTAG.

 


q2-2.png

2.2 - In your opinion, an SJTAG Test Manager is:

  1. An external system for generating and managing tests
  2. A hardware interface to the boundary-scan chains on the board
  3. Any combined hardware/software test control system
  4. None of the above

This question, was part of a "calibration test" for the survey: The term "SJTAG Test Manager" is defined in the White Paper, so the fact that a large percentage of people did not respond with the White Paper definition indicates that our nomenclature is not being absorbed, even by those who have read the White Paper.

 


q2-3.png

2.3 - In your opinion, an SJTAG Test Controller is:

  1. A hardware interface to the boundary-scan chains on the board
  2. A feature of a runtime-control Test Manager built into the UUT
  3. Any combined hardware/software test control system
  4. None of the above

Again, "SJTAG Test Controller" is a White Paper defined term, so the spread of answers indicates that our terminology is not getting across.

 


q2-4.png

2.4 - In your opinion, a JTAG Protocol Manager (JTAG-PM) is

  1. Handshake protocol between a Test Manager and a Test Controller
  2. Any combined hardware/software test control system
  3. A hardware interface to the boundary-scan chains on the board
  4. None of the above

Another White Paper defined term here, "JTAG Protocol Manager", and most people's opinion on what this is differs from our definition! Clearly, taking the responses in this section together shows that we either have a big task in education to undertake or that we need to revise our nomenclature to be more intuitive.

Another view, given the responses to Q2.1, is that the old White Paper simply isn't effective. This is something we suspected and seems to be confirmed by these results, and justifies our effort to produce a new White Paper on our wiki site. We strongly encourage everyone to review the new White Paper, even while it is still a work in progress.

 


q2-5.png

2.5 - In your opinion, SJTAG is:

  1. Predominantly a Software issue and is NOT a Hardware and Architectural one
  2. Predominantly a Hardware issue and is NOT a Software and Architectural one
  3. Predominantly an Architectural issue and is NOT a Hardware and Software one
  4. An even mix of Software, Hardware, and Architectural issues
  5. A even mix of Software and Hardware issues and less of an Architectural issue
  6. An even mix of Software and Architectural issues and less of a Hardware issue
  7. An even mix of Hardware and Architectural issues and less of a Software issue

No trick question here, and the spread of responses is interesting. Although options b), c) and e) attracted no responses, those that do appear represent a pretty broad spectrum of opinion; a) and g) are almost diametrically opposing opinions, for example.

With the modal response giving roughly equal weight to the each of Software, Hardware and Architecture, there is clearly no overwhelming opinion that SJTAG is primarily about Hardware, or primarily about Architecture or primarily about Software.

This is demonstrating the difficulty in defining the SJTAG activity and is why we haven't rushed into preparing a PAR before we have a clear understanding of the issues involved.

< User Survey 2009 Results 2009 Survey, Section 3>