Oliver,
Apologies but I have been mostly offline for the last day or two as this
thread has developed.
I think there is some mis-understanding that I was requesting
*normative* definition of default renderings for ODF.
As several people have pointed out, locales have a great deal of impact
on actual rendering and so a *normative* definition for a single locale
would not be terribly useful.
However, having said that, consider the puzzlement of a new implementer
of ODF. They want users to see their product as providing "interchange"
with Lotus Symphony or OpenOffice but unless they know the details of
the "default styles" being applied, that is going to be rather
difficult. I suppose one could say that they should incur the cost of
studying those programs to learn what is already known to others but
that seems contrary to the notion of an "open" standard.
I suppose not specifying the details of "default styles,"
non-normatively, does make the standard shorter but it also makes it
less precise.
I am not suggesting that we define defaults for all possible locales but
we certainly should know enough of the top 2 or 3 locales to say what
those "defaults" would be, at least non-normatively.
A consequence of not doing so is to create a barrier to those who might
want to implement the standard but lack the resources to develop the
defaults for the locales where we do define non-normative defaults.
I am not suggesting that is why specifying defaults is being opposed but
do think we need to consider the consequences of not specifying those
defaults.
I will have to think about how "default" styles are mentioned in the
current text because I suspect that in several cases knowledge of those
"default" styles influences what is being said in the text.
My tentative position is that if we know some "default" style,
particularly if knowledge of that default style for any locale will make
it easier to implement the standard, then simply noting that locales
vary is a poor excuse for not saying what we know.
Yes, presentations will vary from locale to locale but that missing the
fact that most interchange occurs *intra*-locale. And there users expect
appearances to be relatively uniform (Or at least I do. Admittedly a
universe where n = 1 but I am sure there are other examples that can be
cited.).
Hope you are at the start of a great day!
Patrick
Oliver-Rainer Wittmann - Software Engineer - Sun Microsystems wrote:
> Hi,
>
> Michael Brauer - Sun Germany - ham02 - Hamburg wrote:
>> On 04/14/09 23:41, wrote:
>>> If you define the attributes for a default style then you will need
>>> to make a specific statement about many locale dependent items such
>>> as text direction, decimal separators, even typeface. Locales that
>>> use the defaults stated by the standard will then be able to have
>>> more concise markup, since they could assume the defaults. And
>>> locales that differ from the defaults would require more verbose
>>> markup since they would need to override those defaults 100% of the
>>> time.
>>
>> I agree. And in fact, even in OpenOffice.org, there is no single
>> universal default style, but the default style is generated based on
>> locale information. Further, the user has the option to specify some
>> of these defaults manually.
>>
>> Which is to say that I'm in favor of keeping the default styles
>> implementation dependent.
>>
>>
>
> I agree to Rob's and Michael's opinion.
>
> Best regards, Oliver.
>
>
> P.S.: We have proposal "Add default values" -
> http://wiki.oasis-open.org/office/Add_default_values - in progress in
> order to define at least a part of default values for certain attributes
>
--
Patrick Durusau
Chair, V1 - US TAG to JTC 1/SC 34
Convener, JTC 1/SC 34/WG 3 (Topic Maps)
Editor, OpenDocument Format TC (OASIS), Project Editor ISO/IEC 26300
Co-Editor, ISO/IEC 13250-1, 13250-5 (Topic Maps)
