RE: [xml-dev] Microsoft buys the Swedish vote on OOXML?

[Len Bullard <len.bullard@uai.com>]

From: Jonathan Robie [mailto:jonathan.robie@redhat.com] 
 
>Len, there's no allegation of bribery in the article, there is only an 
>allegation that Microsoft had something to do with gathering most of the 
>new members together, and the author of the article supports this 
>allegation by pointing out that 18 of the 23 are either Microsoft Gold 
>Certified Partners or Microsoft Certified Partners. 

You mean the title "Microsoft buys the Swedish vote" suggested they went to
Wal-Mart to pick up a discounted toy?    Of course MS partners vote for MS
submissions.  That's practical business.  Did MS gather them up?  I can't
say and neither can you nor can the author of that article.  They probably
did.  Are Google, Sun and IBM not doing that as well?  If not, they are not
being very smart.  Only in the slightly goofy period of the W3C trying to
dissuade others from going to other bodies did anyone ever think you don't
have to show up on time with your own posse.

>The article does not say anything specific about what kind of communication
>may have occurred between Microsoft and its partners. 

Right.  It says they bought Swedish votes.

>The article does suggest that it's pretty unusual when a standards vote is
>determined primarily by people who were not members of the group the day
>before, and are partners of a company who was going to lose the vote. Do
>you think this was just a coincidence? 

No.  I think it rather weird that everyone else shows up late in smaller
numbers.   Do Sun, Red Hat, IBM and Google have so few partners they can't
afford to join?

>Do you think this is the way important votes should be held in standards
>bodies?

It's been my experience that late comers are pretty common in this voting
game just as they are in American primaries.  It's low turn out until the
vote counts.  What you really want to know is who is reading the minutes and
the other published papers as they are being issued.  Most of what you need
to know to cast a vote can be known without going to the meetings.

> Google's paper is specious.  It claims wide use for ODF without the 
> numbers to back that up.  

>Here's what they say: "Counting the number of documents found by doing 
>Web searches for different document types the older Microsoft Office 
>formats dominate, but the second most widely used format is the existing 
>ISO standard ODF. As translation is needed anyway it would make more 
>sense to convert to ODF, the existing ISO standard for editable document 
>types." 

And they say:

"Many companies have announced they will support OOXML-and several have
announced translators for the new formats. This is only to be expected-as
Microsoft is a major vendor in the office automation space. Wide industry
support doesn't necessarily make a good ISO standard although it definitely
helps. What matters more for a good interoperable standard is multiple
implementations."

So only if the RIGHT vendors do it for the RIGHT format is it a good thing?
Do the multiple contradictions there come across as loudly as they should?
And if there are millions of .doc, etc. files out there, why should a
customer go to ODF where the translations will be worse because they will be
down translations?  This comes down to 'one standard and it will be ours'.  

>You suggest that Google does not have the numbers to back that 
>up? Er, this is Google, after all ....

Umm... this is the same company that sends me ads based on pay-to-play
words.  That is not exactly the source I trust and the numbers they have (12
implementations) don't really stack up to the switching costs for 'millions'
of documents.

>Google's argument seems to be this:

>1. ODF already exists, and can represent Microsoft documents as well as 
>the documents from other vendors. 

Once translated.

>Blessing OOXML as a second format 
>results in two mutually incompatible formats that can each claim to be 
>standard. Google invokes memories of Betamax to suggest this is bad.

Betamax lost because of shelf space stocking problems for low-end Mom and
Pop video tape vendors, not quality or quantity.  It was a supply side
problem and had Sony spent some time looking at that, they could have
repackaged.  A one standard product market is the same as a one shoe size
mall.  It isn't bad for the store but it is bad for the feet.

Put another way, how many of you want to give up your Audi's, BMWs and
Lexuses to drive VW Beatles made in Mexico?  After all, a car is a car and
if the emission standards could be met, well, you would have to give up the
fourth kid and maybe slim down yourself, but all to the common good, yes?

>2. OOXML is too large and complex to be easily reviewed, and does not 
>play well with existing standards. "The OOXML standard document is 6546 
>pages long. The ODF standard,
>which achieves the same goal, is only 867 pages. The reason for this is 
>that ODF references other existing ISO standards for such things as date 
>specifications, math
>formula markup and many other needs of an office document format 
>standard. OOXML invents its own versions of these existing standards, 
>which is unnecessary and complicates the final standard.", "Considering 
>that OOXML has only received about 5.5% of the review that comparable 
>standards have undergone, reports about inconsistencies, contradictions 
>and missing information are hardly surprising."

Ummm... isn't that the "SGML is too hard" with pages Jon tossed to the floor
argument.  A decade later, bloat and more bloat in XML as the gaps were
filled and we still hate namespaces and XSD.  IOW, for those that want to
get their product from a single source or a source and its partners, it may
not be "unnecessary".   As for the 5.5% number, based on what I am reading
either a lot of review was done or a lot of people are complaining about
specifications they haven't read.

>3. OOXML is not yet widely adopted, at least on the Web. ODF is the most 
>common XML format for office document data, and it would make more sense 
>to translate data from proprietary formats into a simpler, existing 
>standard format that has already received extensive review.

Office documents, Jonathan, not 'web documents'.  The web has become a
cancer according to what supporters of open source tell me ("can't be
stopped; it just pops up somewhere else").  That "Web IS Holy" argument
doesn't wash anymore.  Packing a a doc file into an attachment and sending
it or opening HTML in Word is an everyday occurrence.                    

Web documents are rarely that complicated and if they are, they are in PDF
for which if I want a version that doesn't crack or break, I still have to
buy it from Adobe (yes, I have Foxit; it screws up cut and paste, displays,
and so on).


>4. There is not widespread adoption of OOXML outside of Microsoft. Most 
>of the OOXML implementations are from partners of Microsoft who have 
>contractual agreements to implement OOXML software.

So the partners of Microsoft are inside Microsoft and the partners of Red
Hat are outside Red Hat?  Are you the Red Queen?  That's really bogus.
Everyone is partnering with someone and they vote their self-interest.  That
is the business side of this.  The technical side is a different issue.  You
insist on down translation so your customers will lose information so they
can pay to switch to do business with you?   What kind of deal is that?

>I guess I should express my own views too. I think OOXML is a good 
>thing, and I'm very glad Microsoft produced it. I don't think it should 
>be a standard, because it is designed precisely to represent one 
>vendor's office documents, not as a general purpose office document 
>format created by a group of vendors.

That is a MUCH better deal than if it isn't a standard.  Because of
political gerrymandering, if it isn't a standard, the legal conditions are
such that very large customers have to switch and thee me and the rest of
the taxpayers foot the bill.  Where will that money go?  Well... to Red Hat,
Sun, IBM, and indirectly, to Google.    

So the bottom line is their customers pay you to free them from Microsoft by
taking the same information and making it less informative?  Huh?

>Yawn. I'm not very good at arguments about who are the good guys and who 
>are the bad guys, but I don't find this kind of language terribly 
>convincing. Maybe I've heard too much of it.

Try.  It isn't that hard to figure out who be which if the rich get richer
and the poor pay for the free lunch.  I don't mind capitalism but this is
Prince John taking money from the villagers for a ransom he doesn't intend
to deliver to the man he persuaded to hold his brother hostage.

len
This email and any files transmitted with it are confidential and intended solely for the use of the individual or entity to whom they are addressed. If you have received this email in error please notify the sender. This message contains confidential information and is intended only for the individual named. If you are not the named addressee you should not disseminate, distribute or copy this e-mail.