OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.


Help: OASIS Mailing Lists Help | MarkMail Help

office message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]

Subject: Re: [office] [Fwd: Re: [pr] Help for a marketing and technical message]

On Monday 09 April 2007 14:10, Dave Pawson wrote:
> PIcking up on a couple of these points.
> On 09/04/07, Thomas Zander <zander@kde.org> wrote:
> > We are basically talking about;
> > http://testsuite.opendocumentfellowship.org/summary.html
> > Which is part of;
> > http://testsuite.opendocumentfellowship.org/
> >
> > The test suite is an initiative of Intel which hired several people
> > last year to write tests (based on the specification) and test the
> > two main ODF implementations for compliance. I did the proofing of
> > tests and testing of KOffice as well as automation of  the framework.
> > So I have at minimum read each test and compared it to the actual
> > spec. :-)
> I'm curious what steer Intel gave these people. Seemingly to test
> implementations
> by assessing the man machine interface?

I'm not sure what you are referring to here.
The ODF specification is clear enough to know how a feature is suppost to 
behave in an application, and that can be tested.

> > * ODF focuses on long term storage and openness of data. Never loose
> > sight of that. Specifying how features work before the different
> > implementations (re)invent the wheel in different sizes is in line
> > with that thought.
> This is my point of interest. Do the various implementations lay down
> on disk XML that is 'compliant' (clearer definition needed) with ODF,
> for each feature.

I'm sorry, I again don't follow what you mean. It seems you are mistaking 
ODF with a DTD or schema.  ODF is more than that and being able to read 
ODF means being able to interpret the meaning behind the XML.

Look at it this way; I can write a poem and you can then define being able 
to read it at different levels;
1) you can read the words (know the language)
2) you can understand the sentences (know the grammar of the language)
3) you can understand the concepts and ideas recorded in the text.

What you described is level 2.  ODF instead expects level 3. The 
implementation has to understand what the XML structures on disk mean and 
only when it does that can we say it is compliant with a certain feature.

For example; when I write in a file that this paragraph is a numbered 
paragraph; I expect the application to place the right number in front. 
Only when it understands the concepts and ideas needed to come to the 
right number and when it actually does place that number in front, that's 
when we say it fully supports this feature.

I'm absolutely sure this is not something I'm just making up; it is how 
any and all specifications work.
Thomas Zander

PGP signature

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]