Re: [oiic-formation-discuss] What is interoperability anyway / really?

[robert_weir@us.ibm.com]

"Dave Pawson" <dave.pawson@gmail.com>
wrote on 06/09/2008 03:48:00 AM:

> 
> Taking this idea further, would it be helpful to try and define it
> as a list of
> things. We can all agree on some items, e.g. instance validation against
> the schema.
> As we move to things like pagination, it becomes less clear.
> 1..10 murky to clear?
> Perhaps these are Robs 'features'?
> My two extremes would be:
> Instance validation   10
> pixel perfect layout   1
> 
> If we can find some of the things that would really bug Davids users,
> then we can determine whether they are in|out of our definition
> of interop?
> 
> 
> Is it the feature list? I could be convinced it's even wider than
that?
> A vendors interpretation of feature X could be sufficiently at
> variance with another vendors to cause an end user reaction.
> 

It isn't even clear to me that pixel perfect layout
is providing greater interoperability.  A GIF file will give you a
perfect, interchangeable pixture, but good luck trying to extract information
from it.  Similarly, a PDF has fixed layout, but that one man's "fixed"
is another man's "inflexible".  In some uses, like on devices,
the ability to freely flow and repaginate is essentially and this improves
interoperability.

I forget who it was, Roger Bacon or General Robert
Lee, but some wise man said, "In order to be able to do some things,
you must be able not to do other things".  It is important that
we respect the design center of ODF, that it is not a fixed layout markup
language, and that this brings certain benefits while also involving certain
trade-off's.  We can push the envelope a little, but we're not going
to turn ODF into PDF.  We don't need to.  We already have PDF.

-Rob