Re: sc2.5.1_l1_002 step 2: Structure review

Quoting Carlos Iglesias <carlos.iglesias@fundacionctic.org>:
> As per my action item from the lasT teleconference here we have the
> structure review (step 2) of the second test sample with some
> comments 
> 
> * contact information of the submitter is available;
> 
> Where is supposed to be recorded this information?

I thought this would be recorded in the Wiki or Bugzilla, once we start 
using it. (At the moment, we only have CVS history and this mail 
archive.)


> * all necessary files follow the naming conventions;
> 
> The test samples metadata document specifies that within each
> technology directory there should be a sub-directory with the files
> for the actual test sample content and a sub-directory with the files
> for the metadata.
> 
> There is a "testfile" subdirectory for the test sample content and a
> "metadata" subdirectory for the metadata, but the proposed name in
> the usage document for the test sample content is "content" instead
> of "testfile"

The original name for the directory was "testfiles" (plural!), but 
Shadi proposed to change this to "content". 


> Additionally the test sample makes use of embedded scripting, should
> this have been externalized under a script subdirectory?

The script is only a few lines long; that is probably why it is not in 
a separate file. Do you think this should be required?

> * all the metadata restrictions are applied, for example:
> 	o dates and other values use the correct format; OK
>       o copyright notices and other values are correct;
> 
> The version metadata is  $Revision: 1.3 $ but there is a comment in
> the test sample content that reads $Revision: 1.2 

That's because the metadata were changed (due to changes in the schema) 
but the sample file was not. It's very common that metadata are changed 
without changes in the corresponding sample file (at least, it is very 
common in BenToWeb), so I think that revision numbers between metadata 
and content files will often be out of sync.

>       o techniques and locations are referenced correctly;
> 	
> According to TCDL spec "The locations element type specifies the
> location or locations where relevant (especially faulty) code
> occurs..." but it's not clear for me where should the locations point
> in a pass case, should we restrict locations to faulty use cases in
> the Metadata Vocabulary?

Not necessarily, because techniques often have clearly identifiable 
"locations" (e.g. labels for input fields, alt attributes, etcetera).


> Additionally there is a "requiredTests" element which is not part of
> the Metadata Vocabulary

The "requiredTests" element currently just records the test mode 
("testMode" element). It is still there because I thought we wanted to 
get rid if the scenarios, but not the test mode. Did I get that wrong?

Best regards,

Christophe

-- 
Christophe Strobbe
K.U.Leuven - Departement of Electrical Engineering - Research Group on 
Document Architectures
Kasteelpark Arenberg 10 - 3001 Leuven-Heverlee - BELGIUM
tel: +32 16 32 85 51
http://www.docarch.be/ 

Disclaimer: http://www.kuleuven.be/cwis/email_disclaimer.htm

Received on Tuesday, 2 January 2007 12:08:48 UTC