[ODFPlugtest] writing implementation specific defaults

Wouter van Vugt wouter at code-counsel.net
Thu Jun 18 13:03:04 CEST 2009


I don't really care about samples. The TC can figure that out.

The point is more that implementers should *not* do something. And that
is that they should not say 'hey, this document is in A4, that's our app
default, so let's not store that information in the document itself'. As
long as we all do not say that, documents should not be different when
opened here or there or anywhere. Since all the vital layout information
can either be found in the spec, or in the document. Together they
should cover 100%.

The reason that we have this discussion is that this is what
implementers are currently not doing. Currently, the information is not
persisted, so we all have different *application* defaults. Hence,
rendering quirks occur in normal views.

Does that clarify it to a degree that we can all agree this is the way
to move forward?
Sensible defaults --> define in spec
Or --> always store in document

That's it, plain, simple, I like it!

Then there is the second story of wanting to *view* a document in other
apps like on your telephone. In which case the document might need a
change of layout. However, that is an application feature. We cannot
expect the ODF spec to cover how dynamic layout should occur. If I place
an image 20 centimeters from the left border, that was my intention and
I very much hope that someone in China sees it in that exact location
too. Except when they open a similar view to the MS Office 2k7 Full
Screen Reading layout. In that case the user expresses the intent of
changing layout for improved readability, and hence layout quirks are
acceptable. 

Code Counsel
Sharing Knowledge with Passion
wouter at code-counsel.net


-----Original Message-----
From: plugtest-bounces at opendocsociety.org
[mailto:plugtest-bounces at opendocsociety.org] On Behalf Of Jan H
Wildeboer
Sent: donderdag 18 juni 2009 12:42
To: ODF Plugfest mailinglist
Subject: Re: [ODFPlugtest] writing implementation specific defaults

Wouter van Vugt wrote:

> My main point is that the situation with sensible defaults should be
> coming from the spec, not from the app.

Please do give some insight. Tell us some examples where you see
sensible defaults.

I mostly see stuff where it doesn't make sense.

Paper size? Region specific, no sensible default possible

Borders? Paper size specific, also depends on printable size of your
printer, sometimes you want to create bleeding content for a reason etc.
- no sensible default possible

Font size? Forget it.

For interop the old rule of "less is more" holds very strong. I expect a
standard to give me posible values but bo defaults. A standard *might*
have some suggestions as informal notes, to explain the use of a certain
attribute, but the standard should not move into the values themselves.

Jan



-- 
Jan H Wildeboer          |
EMEA Open Source Affairs | Office: +49 (0)89 205071-207
Red Hat GmbH             | Mobile: +49 (0)174 33 23 249
Otto-Hahn-Str.20         | Fax:    +49 (0)89 205071-111
D-85609 Dornach/Munich   | eMail:  jan.wildeboer at redhat.com
_____________________________________________________________________

Reg. Adresse: Red Hat GmbH, Otto-Hahn-Strasse 20, 85609 Dornach bei
Muenchen
Handelsregister: Amtsgericht Muenchen HRB 153243
Geschaeftsfuehrer: Brendan Lane,Charlie Peters,Michael Cunningham,
Charles Cachera
_____________________________________________________________________

GPG Key:     3AC3C8AB
Fingerprint: 3D1E C4E0 DD67 E16D E47A  9564 A72F 5C39 3AC3 C8AB
_______________________________________________
Plugtest mailing list
Plugtest at opendocsociety.org
http://lists.opendocsociety.org/mailman/listinfo/plugtest



More information about the Plugtest mailing list