RE: Question regarding TAG and TAML

["Jacques R. Durand" <JDurand@us.fujitsu.com>]

Bahareh:

Thanks for timely comments - we will look into it in the TC.

A quick response for now: machine-readable mark-up is still valuable
even for plain text Predicates, if only as a way to structure content
say for HTML rendering and navigation. 
(e.g. if some links are "formalized", like Normative Source references,
then the HTML rendering can support this linking and allow for useful
navigation  even if predicate and prereq expressions are
non-processable).

I am also reviewing your OWL ontology and will send comments.

Cheers,

Jacques

-----Original Message-----
From: Bahareh Heravi [mailto:Bahareh.Heravi@brunel.ac.uk] 
Sent: Tuesday, May 11, 2010 7:41 AM
To: tag@lists.oasis-open.org
Cc: Jacques R. Durand; David Snelling
Subject: Question regarding TAG and TAML

Dear All,

As far as I understand Test Assertions are meant to kind of formalise
Normative Statements in the specifications which is superb. 

I couldn't find a description for Test Assertion Mark-up Language in its
spec but I reckon it provides a XML representation of the Test
Assertions. However, I don't quite get the point of a machine readable
representation when the predicates are not formal.  For example in the
following, which is the only example I could find in the TAML spec:

TA id: widget-TA104-2

Normative Source: specification requirement 104

Target: widget

Predicate: [the widget] is from 5 to 15 centimeters long in its longer

dimension.

Prescription Level: mandatory

Tag: normative_property = medium-sized

 

And the XML representation:

<testAssertion id="widget-TA104-2">

. . .

<predicate> [the widget] is from LENGTH-A to LENGTH-B long in its

longer dimension</predicate>

. . .

<tag name="DefinesNormativeProperty">true</tag>

<tag name="NormativeProperty">medium-sized</tag>

</testAssertion>

 

To me it is still the same thing with not that much added semantic to
the first representation and is readable but not understandable by
machines. 

Consider the following example from ebBP v2.0.4:

"The specific roles (e.g. buyer, seller) MUST be specified at the
Business Transaction Activity level, when the Business Transaction
definition is used for a distinct purpose."

I am not an expert in writing TAs but I would write something like this:

TA id: ebBP-xxx

Normative Source: specification requirement 729 (the line number of the
sentence for your reference)

Target: Role

Predicate: [the Role] is (?) specified at the Business Transaction
Activity level, when the Business Transaction definition is used for a
distinct purpose.

Prescription Level: mandatory

 

I am not sure if the way I wrote it is correct, but if it is, the
predicate doesn't add much to the textual information in the spec and I
believe it is not easily measurable which contradicts the aim of TAs as
far as I understand. My understanding is that the XML representation
would not add that much to it either. 

Having said all these I am a bit struggling with the notion of Test
Assertions and their XML representation  and am not sure how they can
help in measuring the conformance of an implementation to a spec or
maybe even how the mark-up can help in writing test cases.

I have to say I do agree that XML representation of Test Assertions
could be very helpful, but if they have some logic base to make them
machine understandable as opposed to only readable. 

I would appreciate if you could shed some light on the matter.

Best regards
Bahareh

===================
Bahareh R. Heravi
PhD Researcher
Department of Information Systems & Computing Brunel University
Uxbridge, Middlesex, London, UB8 3PH, United Kingdom.
Web: http://people.brunel.ac.uk/~cspgbrh
<https://owa1.brunel.ac.uk/exchweb/bin/redir.asp?URL=http://people.brune
l.ac.uk/~cspgbrh>