[odf-discuss] Three kinds of tests

[M. Fioretti]

On Wed, May 07, 2008 00:34:25 AM -0400, robert_weir at us.ibm.com wrote:

> In my mind, when think of the ideal team for this, I think we need the
> following:
...
>   - A "toolsmith", someone with excellent scripting ability (e.g.,
>     Python)and knowledge of XML in general to help create
>     time-saving automation tools that will make this effort
>     possible.
>   - Someone with a formal quality assurance background who can guide
>     the team into creating high quality test cases.

>   - Someone with good outreach skills, especially into organizations
>     that have adopted ODF.  Especially on the real-world
>     interoperability tests,

While surely I'm not the best person for neither of the tasks above, I
believe I could probably be of some use with all of them. How much and
how constantly depends on several things, we'll see.

On a technical/political note (just curious, I agree there are more
urgent things, like finding out who can actually do the work):

>     No sense in us spending months on text styles if spreadsheet
>     formulas are the things that are causing the most problems.

technically speaking, it makes perfect sense: what creates problems to
_real_ end users (ie the people who _actually_ use ODF in the field
for their work) has top priority. But this is not a technical battle,
is a political one: I mean that any ODF-capable application,
especially desktop office suites, would probably be asked by
indoctrinated burocrats to have "100% conformance or nothing" to be
considered in any public tender. How would this be handled? Again, I
agree this is not priority, and am not even sure if there is (can be)
a real answer to such a question, but I thought I'd just ask anyway,
to brainstorm a little if nothing else.

   	      	       	  Marco

-- 
Your own civil rights and the quality of your life heavily depend on how
software is used *around* you:            http://digifreedom.net/node/84