[odf-discuss] Microsoft wins prize for "Best Campaigner against OOXML Standardization"

[marbux]

On 10/3/07, Daniel Carrera <daniel.carrera at zmsl.com> wrote:
>
> marbux wrote:
> > And I sincerely
> > doubt that a company that has added more than 150 proprietary extensions
> > to ODF in its own implementations would be all that concerned about a
> > very small amount of dark matter that in no manner affects application
> > performance except by enhancing round-trip ability.
>
> I think that there is a difference between OOo's extensions and dark
> matter. The difference is that OOo understands its own extensions. I
> make no claim that OOo's extensions are right or wrong, but rather that
> they feel different for Sun:
>
> 1. If they could modify OOo in a way that makes it read files from MS
> Office better, but requires extending ODF another 150 times, I'm sure
> they'd go for it.


Domonstrably inaccurate. Sun was willing to kick in its $350,000 to fund the
Foundation to finish ODF plug-ins for all three major Office apps on a crash
basis, as requested by Massachusetts ITD, but was insist on terms that would
have allowed them to make sure the plug-ins never saw the light of day
outside Massachusetts. And that required only 5 generic extensions.

2. If all they can do is insert a dark matter blob, that doesn't help
> OOo read files from MS Office any better. So they can reasonably decide
> that there is no benefit.


The dark matter blobs are for the less than 1 per cent of the Microsoft
metadata in its IMBR that have not yet been decoded. As I said, even if Sun
didn't create dark objects in the resulting ODF they could still achieve
over 99 per cent fidelity using the extensions we proposed by simply
destroying the blobs. The difference between 80-85% fidelity and 99+%
fidelity is not an insignificant benefit.99+% approaches what is necessary
for business processes in the enterprise market. 80-85% is a non-starter.


> Plus I can't see why a company that is so demonstrably dead-set against
> > high-fidelity round-trip interop with MS Office
>
> Let's be careful to not express that opinion as a fact (see the Code of
> Conduct) (this is not a warning, I'm not the moderator). We know that
> the TC members from Sun, IBM and KDE have voted against your dark matter
> proposals.
>

KDE does not vote on the TC. And it is not just the "dark matter proposals"
that went down in flames. Our metadata proposals did as well, but a proposal
from Sun that blocked our ability to use metadata to do the job also went
down in flames, and was compounded by a an ambush attack by Sun just before
the Metadata SC work went to the TC work that stripped the requirement of
preserving RDF metadata created by other apps. The Sun proposal was clearly
identified by Bruce d'Arcus as one that would break interop  among ODF apps
yet it sailed through the SC and the TC without a single use case
demonstrating any need for the discretion. To boot, the TC archives are
rampant with Sun posts unequivocally expressing opposition to interop with
MS Office.

If discussion of the fox in the ODF hen house is off topic, then the
Fellowship is deliberately blinding itself to a very serious problem.

This doesn't mean that they are against round tripping
> improvements in all its forms. For example,


They are unquestionably opposed to round-tripping in non-lossy mode.


*I* oppose the Foundation
> proposals that I saw, but that doesn't mean that I will also oppose a
> later proposal. One thing I tried to convey early on (months ago) was
> that I felt your proposal was impossible "as stated".


What can I say? It works and the only thing that was necessary was to get
the extensions into the spec so others could use them and to get Sun to stop
trashing foreign elements and attributes.


That is, a
> different proposal that achieves a similar outcome may be possible. In
> fact, I have some ideas, but I haven't come up with anything that can
> always survive editing in another application :-(.


I'll be lodging an interop proposal with the TC in the next few days that
takes yet another different tack. I fully expect Sun to vote it down. But
I'm making a record that the request was made for later use at ISO. I will
post a link to the proposal when it is made.

Best regards,

BUCK "MARBUX" MARTIN
  Director of Legal Affairs
  OpenDocument Foundation
  Contact:
<http://www.opendocumentfoundation.us/contact.htm>
-- Universal Interop Now!
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.opendocumentfellowship.com/pipermail/odf-discuss/attachments/20071003/ef8bb33d/attachment-0001.htm