Re: Test Leader in WGs

At 10:45 AM 11/26/2004 -0500, Karl Dubost wrote:


>Le 25 nov. 2004, à 12:31, Bjoern Hoehrmann a écrit :
>>It would make more sense to me to first determine all the work and then 
>>how to manage the work.
>
>Agreed. It's why I'm asking here what people think it should be.
>
>>In terms of what would make sense for the QA WG to deliver is a document 
>>listing all the issues around testing like dealing with test
>>contributions and tracking test issues, what that involves and what
>>could help to get that done.
>
>I'm talking about any kind of technologies. So let's say a LoveML is 
>created and a Test Suite is created at the same time than features.
>
>I can identify a few things:
>
>- license issues for receiving test (contribution)
>- license issues for publishing test
>- cvs repository for the test cases
>- packaging of the test suites
>- manual of the test suites and each individual test cases
>- Feature/Test assertions/Test Cases package
>- Interoperability report
>- tools to manage the test suites

This sounds like a condensed TOC of the guidelines in the old OpsGL 
[1].  Most of those topics have survived in more abbreviated and simplified 
form in QA Handbook [2].

[1] http://www.w3.org/TR/2003/CR-qaframe-ops-20030922/
[2] http://www.w3.org/TR/2004/NOTE-qa-handbook-20041122/


>All these kind of things might have influence on the work of a test 
>leader. In which ways? Maybe just a guide detailing all the things that 
>have to be organized and how to assign the tasks.

So is this "test leader" the same as the current QAH [2] notion of "QA 
point-of-contact".  We (QAWG) hassled around with that job title for a long 
time.  In fact, at one point that job title was "Test Manager".

>>Working Groups can then decide themselves whether it makes sense to have 
>>a "Test Leader" and what he would be supposed to do. Unless of course 
>>having a "Test Leader" has been properly researched and clearly
>>determined as a best practise, but then you would not need to ask the
>>questions above :-)
>
>No. I'm asking the question because I have ideas but I prefer to have more 
>input because people might think differently or will have ideas I didn't 
>have. Maybe I was not clear enough about the context. It's not related to 
>the QA WG, but to any kind of WGs.
>
>New WGs are always struggling with the same questions and they come to a 
>period of trying to define when it could be done before. :)

Bottom line:  I'm a little confused about the relationship of proposed Test 
Leader and his/her job description, on the one hand, to the advice of QAH 
(and OpsGL/ET before that) on the other hand.

Are you proposing that QAWG develop a Technique (or template or ...) to 
implement some aspects of the generalized advice of QAH?

Btw, I agree with Lynne's later comment, that some such description or 
outline of topics/considerations is a good idea.

-Lofton.

Received on Wednesday, 1 December 2004 15:28:50 UTC