decoration decoration
Stories

GROKLAW
When you want to know more... 		decoration
For layout only
Home
Archives
Site Map
Search
About Groklaw
Awards
Legal Research
Timelines
ApplevSamsung
ApplevSamsung p.2
ArchiveExplorer
Autozone
Bilski
Cases
Cast: Lawyers
Comes v. MS
Contracts/Documents
Courts
DRM
Gordon v MS
GPL
Grokdoc
HTML How To
IPI v RH
IV v. Google
Legal Docs
Lodsys
MS Litigations
MSvB&N
News Picks
Novell v. MS
Novell-MS Deal
ODF/OOXML
OOXML Appeals
OraclevGoogle
Patents
ProjectMonterey
Psystar
Quote Database
Red Hat v SCO
Salus Book
SCEA v Hotz
SCO Appeals
SCO Bankruptcy
SCO Financials
SCO Overview
SCO v IBM
SCO v Novell
SCO:Soup2Nuts
SCOsource
Sean Daly
Software Patents
Switch to Linux
Transcripts
Unix Books

Gear

Groklaw Gear

Click here to send an email to the editor of this weblog.

You won't find me on Facebook

Donate

Donate Paypal

No Legal Advice

The information on Groklaw is not intended to constitute legal advice. While Mark is a lawyer and he has asked other lawyers and law students to contribute articles, all of these articles are offered to help educate, not to provide specific legal advice. They are not your lawyers.

Here's Groklaw's comments policy.

What's New

STORIES
No new stories

COMMENTS last 48 hrs
No new comments

Sponsors

Hosting:
hosted by ibiblio

On servers donated to ibiblio by AMD.

Webmaster
Microsoft Office 2007 Fails OOXML Conformance Tests, Alex Brown Admits, Hopes For the Best
Monday, April 21 2008 @ 09:42 AM EDT

This takes the cake. Alex Brown has just admitted on his Griffin Brown blog and further to ZDNET UK's Peter Judge that Microsoft Office 2007 has failed two OOXML conformance tests he ran. First ZDNET:
In a blog posting this week, Alex Brown, leader of the International Organization for Standardization (ISO) group in charge of maintaining the Office Open XML (OOXML) standard, revealed that Microsoft Office 2007 documents do not meet the latest specifications of the ISO OOXML draft standard.

"Word documents generated by today's version of Microsoft Office 2007 do not conform to ISO/IEC 29500," said Brown in a blog post recounting the process of testing a document against the "strict" and "transitional" schema defined in the standard.

Color me surprised. Say, France, you might want to slow down adding this "standard" to your list until it actually works. All you folks who voted for it need to tell us why you accepted it before it was done. Because what this means is that OOXML was just approved as an ISO standard, on the allegation that it was necessary for interoperability with Microsoft documents, and it turns out it doesn't even do that. It also means no one can interoperate successfully with Microsoft Office 2007 except Microsoft. Neato. Back to Go. Do not collect $200. Isn't the Fast Track supposed to be for already *implemented* standards?

I know. After ISO was captured by Microsoft, nobody cares about rules any more. Brown hopes Microsoft will be a good doo-bee and fix this.

What are the odds?

In a statement sent to ZDNet.co.uk on Friday, Brown said that, although he was hopeful that Microsoft will update its Office products to stay in line with the version of OOXML approved by ISO, it is not guaranteed. "The question behind the question, for a lot of the current OOXML debate, seems to be: can Microsoft really be trusted to behave? We shall see," said Brown.

Commentators, including Tim Bray, the inventor of XML, have suggested that Microsoft is unlikely to bother to keep conformant with the OOXML standard as it develops within ISO, but Brown was more optimistic: "Given Microsoft's proven ability to tinker with the Office XML file format between service packs, I am hoping that Microsoft Office will shortly be brought into line with the [ISO/IEC] 29500 specification, and will stay that way," he said. "Indeed, a strong motivation for approving 29500 as an ISO/IEC standard was to discourage Microsoft from this kind of file-format rug-pulling stunt in future."

Hilarious. We saw how well that worked already, watching the process by which Microsoft got this "standard" approved in the first place. Brown adds with the usual dose of snark that he hopes to test OpenOffice against ODF sometime. He persists in pretending that Microsoft is not a monopoly or that such a position doesn't mean anything. Here's what he concluded from the tests he ran:

Such a test is only indicative, of course, but a few tentative conclusions can be drawn:
* Word documents generated by today's version of MS Office 2007 do not conform to ISO/IEC 29500

* Making them conform to the STRICT schema is going to require some surgery to the (de)serialisation code of the application

* Making them conform to the TRANSITIONAL will require less of the same sort of surgery (since they're quite close to conformant as-is)

Given Microsoft's proven ability to tinker with the Office XML file format between service packs, I am hoping that MS Office will shortly be brought into line with the 29500 specification, and will stay that way. Indeed, a strong motivation for approving 29500 as an ISO/IEC standard was to discourage Microsoft from this kind of file format rug-pulling stunt in future.

"17MB (around 122,000) of invalidity messages" in the strict test; less in a "transitional" model, meaning one no one on the planet will be using, since the entire point of the BRM was to fix stuff and none of those fixes are yet incorporated into Microsoft Office 2007. And by the time they are, will Microsoft Office 2007 have moved on, so we can continue to play catch up with Microsoft forever and a day? Isn't that what standards are supposed to prevent? Tim Bray, the man who invented XML, told us already not to hold our breath for Microsoft to fervently fix OOXML:

I suppose they’ll probably show up to the meetings and try to act interested, but it’s going to be a sideline and nobody important will be there. What Microsoft really wanted was that ISO stamp of approval to use as a marketing tool. And just like your mother told you, when they get what they want and have their way with you, they’re probably not gonna call you in the morning.

He calls it ISO's Fantasy, and Brown, sitting by the phone, has it bad. Here's the part he's not mentioning: OOXML is headed for maintenance mode, so even more changes are going to happen. And it allows for proprietary extensions. So it's a perpetually moving target.


  


Microsoft Office 2007 Fails OOXML Conformance Tests, Alex Brown Admits, Hopes For the Best | 408 comments | Create New Account
Comments belong to whoever posts them. Please notify us of inappropriate comments.
Corrections here
Authored by: Erwan on Monday, April 21 2008 @ 09:46 AM EDT
If any.

---
Erwan

[ Reply to This | # ]

News Picks Discussions here.
Authored by: Erwan on Monday, April 21 2008 @ 09:47 AM EDT
Please don't forget to quote the article.

---
Erwan

[ Reply to This | # ]

Off Topic
Authored by: feldegast on Monday, April 21 2008 @ 09:48 AM EDT
Please make links clickable

---
IANAL
My posts are ©2004-2008 and released under the Creative Commons License
Attribution-Noncommercial 2.0
P.J. has permission for commercial use.

[ Reply to This | # ]

Other MS Standards
Authored by: Anonymous on Monday, April 21 2008 @ 10:05 AM EDT
I don't know any facts or figures about MS' past history in implementing and
maintaining standards. It would be interesting to see of the standards that
they've been involved in/pushed-for, how many of them are kept current, and what
software implements them.

For instance, I'd be particularly interested in compliance with ECMA 234 -
http://www.ecma-international.org/publications/standards/Ecma-234.htm -
Application Programming Interface for Windows (APIW) (December 1995).

I wonder if Vista implements this. And if it does, what differences there are
between the Vista implementation and the spec. Similarly XP.

Regarding spec maintenance, it would be interesting to know which software
implemented the spec, and then stopped implementing. Or which grew extensions,
whilst the spec was not updated, making it impossible to guarantee
interoperation for software built upon spec+extensions.

JeffV

[ Reply to This | # ]

Effective way to prevent false claims of OOXML compliance
Authored by: Anonymous on Monday, April 21 2008 @ 10:28 AM EDT
In Germany, they will NOT advertise MS Office as OOXML compliant until it
actually is - should they dare, the day after they publish respective nonsense,
they will catch a Cease & Desist letter (Abmahnung).

This has worked with SCO fud and stopped it very quickly in Germany, it will
work here.

I love my countries competition laws :-) While they are sometimes a little bit
hard against small business, they have sharp teeth to bite companies such as
Microsoft Deutschland GmbH and prevent companies from spreading fud quite
reliably.

For other German readers here: If you get aware of such claims or
advertisements, please report them to a Verbraucherzentrale or Linux-Verband.de.


If you are in the software business yourself, you may also contact an attorney
who can make some quick money with such a D&D letter.

(NB: I have no relation to Linux-Verband Deutschland or any MS competitor)

[ Reply to This | # ]

When has Microsoft ever been standards compliant?
Authored by: Anonymous on Monday, April 21 2008 @ 10:29 AM EDT
When has Microsoft ever been standards compliant? Just because they wrote the
standard wasn't going to change that, was it?

[ Reply to This | # ]

Microsoft Office 2007 Fails OOXML Conformance Tests, Alex Brown Admits, Hopes For the Best
Authored by: PolR on Monday, April 21 2008 @ 10:30 AM EDT
He persists in pretending that Microsoft is not a monopoly or that such a position doesn't mean anything.
Nowadays, reading Alex Brown on such topic is like watching the dead parrot sketch. Check the second comment in this thread of consortium.info to see what I mean.

[ Reply to This | # ]

"Sitting by the phone" borders on a personal Attack
Authored by: Anonymous on Monday, April 21 2008 @ 10:35 AM EDT
Obviously, it is better for Alex Brown personally to put on his CV, "I
chaired OOXML committee, and it became the world standard", than "I
spent a whole spring on some little format that Microsoft tried to force on
everyone, but the format has disappeared now". This will no doubt colour
some of his blog posts.

While he may not be a part of the open source world, I don't we think should
resort to personal attacks.

"Brown, sitting by the phone, has it bad" is a joke, but it might seem
on the border of a personal attack to some.

If he is sitting by the phone then it is probably to answer his poor customers
who have problems using Microsoft's nonsense format.

[ Reply to This | # ]

Microsoft Office 2007 Fails OOXML Conformance Tests, Alex Brown Admits, Hopes For the Best
Authored by: Anonymous on Monday, April 21 2008 @ 10:42 AM EDT
I am sorry if I an asking a question which may be blindingly obvious, but here
it goes.

Does this mean that Microsoft cannot claim that their Office product meets their
own ISO standard?

Cant help it. Snort...Splatter...Honk.

I am getting all emotional.

[ Reply to This | # ]

Both ODF and OOXML are irrelevent at the moment
Authored by: TiddlyPom on Monday, April 21 2008 @ 10:44 AM EDT
At the present moment in time, both ODF and OOXML are completely irrelevent.

Big statement I know but there are a *vast* number (I would say large majority) of users out there still using Microsoft's old binary [doc/xls/ppt] formats as a de-facto standard. As an OpenOffice user myself, I store my own documents in ODF format but end up having to send documents to other people in [doc/xls/ppt] formats or render them as PDF if I know they are only for reference.

I do not know *anyone* in business in the UK who uses either Microsoft XML formats (of any type) yet (although there must be some). In every case it is assumed that you will send [.doc] format files if they are a word processing document.

Why state this?

If users are to move from binary formats to *any other* format then there is huge inertia to overcome (and millions of documents will have to be converted into new formats). I have come across this time and again whilst evangelising about OpenOffice. The standard question is always
"Can I use Microsoft documents with it?" (meaning the current/old binary formats)
Obviously this is why Microsoft want OOXML to be the new standard but they will have a pretty hard task on their hands forcing users to migrate over to the new (unknown) format - I would say as hard as the FOSS community convincing users to migrate to ODF instead.

It is going to be ages yet before the current binary formats are consigned to history and once they are, Microsoft's hold over users is going to be even more tenuous.

---
Open Source Software - Unpicking the Microsoft monopoly piece-by-piece.

[ Reply to This | # ]

OpenDocument validation
Authored by: Anonymous on Monday, April 21 2008 @ 10:44 AM EDT
Hi!

Does someone know where I can find official ODF
conformance tests, so that I can check whether my files
are truly in an ISO-approved format without variations?

Recently I started to code a PHP website that generates
OpenDocument files. While doing so, I used the information
made available by the OpenDocumentFellowship. They have a
nice tutorial for OpenDocument and an even nicer ODF
validator: http://opendocumentfellowship.com/validator

Unfortuately the validator is still "very much alpha
software right now":
(http://www.cyclone3.org/documentation-of-XUL-CMS/addons/extensions/.svn/ODFvali
dator/)

The validator reports errors on all OpenOffice 2.3 files I
throw at it. The validator targets the ISO-approved
OpenDocument 1.0, and maybe OpenOffice follows the (not
yet ISO-approved) OpenDocument 1.2 spec. Or maybe this is
simply because of bugs in the validator.

BTW, the ODF standard is badly written and it is extremely
difficult to code to it, but the OOXML standard is even
worse. I quickly gave up with OOXML completely - that file
format is really strange and un-XML-ish.

Olaf

[ Reply to This | # ]

Hopes For the Best? Arnaud Le Hors knows the future
Authored by: Winter on Monday, April 21 2008 @ 10:47 AM EDT

Copied from GrokDoc:

A History of Microsoft and Standards

Quote:

Arnaud Le Hors posted some personal experiences "straight from the horse's mouth" in What Microsoft’s track record tells us about OOXML’s future. I think MS' employees should know best.

This kept me intrigued until one day my peer at Microsoft told me: "For us, the submission to W3C is the end of the road. What happens after doesn’t really matter."
I didn’t need to ask why. The explanation was obvious. Once the specification is submitted to W3C Microsoft can tell its customers that it is a standard. Technically it’s not, but if any customer ever cares to ask it’s easy enough to put their fear to rest by explaining that the process is started and it’s simply a matter of time. By the time the standard eventually comes out, customers are already using Microsof’s technology and no longer have much choice if they find out that Microsoft doesn’t even bother adhering to the actual standard. They are locked-in.

Rob

---
Some say the sun rises in the east, some say it rises in the west; the truth lies probably somewhere in between.

[ Reply to This | # ]

Microsoft Office 2007 Fails OOXML Conformance Tests, Alex Brown Admits, Hopes For the Best
Authored by: Anonymous on Monday, April 21 2008 @ 10:50 AM EDT
We saw how well that worked already, watching the process by which Microsoft got this "standard" approved in the first place.

Oh, PJ don't be silly.

M$ will introduce special option on by default "Produce ISO compliant XML files" which would also cripple some crucial functionality. So officially they will be off the hook ("Office does produce standard-compliant files by default!") but everybody would be immediately turning the option off to not to lose access to all features.

This way it will remain the way M$ want it: proprietary non-compliant file format would remain mainstream, while standard-compliant files will be literally unusable.

P.S. Also, We should rather brace ourselves for astroturfing by M$ partners all in one voice screaming that "Office2007 produces OOXML and OOXML is XML and you can read it with Notepad and there is ISO standard with same name thus it is standard compliant".

[ Reply to This | # ]

Be Careful What You Wish For
Authored by: Anonymous on Monday, April 21 2008 @ 10:57 AM EDT

I have shared PJ's interpretations of Alex Brown's past decisions as indicative of a push from the top to short-circuit deliberations (especially regarding real- world realities) and get a standard by any means necessary. That said, some of today's pull quotes make me wonder if we aren't seeing Headmaster Brown laying down some discipline and introducing Microsoft to his inversion of the shop-owner's rule, i.e., "You bought it, you can't break it."

[ Reply to This | # ]

What's Next?
Authored by: dio gratia on Monday, April 21 2008 @ 10:57 AM EDT
From Alex Brown's blog:
To repeat the exercise with ISO/IEC 26300:2006 (ODF 1.0) and a popular implementation of OpenDocument. Will anybody be brave enough to predict what kind of result that exercise will have?
It's nice to have the luxury of multiple implementations from which to select 'a popular one'. Can we guess which one will be tested?

[ Reply to This | # ]

The conspiracy theories get legs (repost)
Authored by: Winter on Monday, April 21 2008 @ 11:00 AM EDT

In the previous article I commented about Marbux' conspiracy theories regarding back-room deals. It only gets better with Alex Brown's admission which increases the pressure on MS.

In a rather rambling post, Marbux presents his thoughts on a Private deal to approve OOXML? More evidence surfaces

I vaguely remember that this has been discussed in GL before. But given the "scandals" that have surfaced in the mean time, is there any news?

What I make of this post is that NBs have forced MS to put OOXML under full JTC1 control for their last minute switch. The French response already hinted at veiled threats. MS were very vulnerable, so some "diplomacy" could force a lot of concessions out of them. And if anything, the French are GOOD diplomats.
Actually, coming to think of it, in this situation I would have to EXPECT governments to use their position to get every concession out of MS they could. So, indeed, MS could not "lose", but might regret winning later.
(to stay in style, Hitler dearly regretted his deal with Chamberlain, as the 2 year delay did cost him the Battle of Britain, and maybe the war)

Furthermore, there could have been a deal to force harmonization between ODF and OOXML. This seems odd as OOXML needs only 2 new tags to get full two way compatibility with ODF. ODF already can store almost everything OOXML stores. Therefore, harmonization would mean that OOXML would be made irrelevant.

So maybe this WAS a Pyrrhic victory for MS. It seems to be a feature of good diplomacy to let your opponent get the public victory when they really lost completely. Then they lose and won't even get the public sympathy of the underdog.

Any thoughts?

Rob

---
Some say the sun rises in the east, some say it rises in the west; the truth lies probably somewhere in between.

[ Reply to This | # ]

Why were conformance tests only run AFTER approval of the standard?
Authored by: Anonymous on Monday, April 21 2008 @ 11:07 AM EDT
After all, isn't application compliance a fair indicator of a vendor's
commitment to a standard?

[ Reply to This | # ]

Alex Brown is missing one huge point
Authored by: DMF on Monday, April 21 2008 @ 11:17 AM EDT
Rather than locking down MS' internal format to the new "standard" (which isn't going to happen anyway since the standard seems to allow proprietary extensions), the standard itself is a moving target. It was adopted with the explicit understanding that it will be changed.

The change process can be stretched out over years and each new release of Office can be marketed - indeed, governments may be forced to upgrade - on the basis that the new release is more compatible with the "new evolved standard" than the last one. A properly controlled change process can also be sufficient justification for new releases of Office all by itself!

Far from what Brown claims, all ISO has managed to standardize is the Microsoft must-upgrade money machine. The effect on open document access will be negative.

[ Reply to This | # ]

Is is possible for someone to be as naive as this?
Authored by: Anonymous on Monday, April 21 2008 @ 11:21 AM EDT

"The question behind the question, for a lot of the current OOXML debate, seems to be: can Microsoft really be trusted to behave? We shall see," said Brown.

"can Microsoft be trusted"? Are there really people in the IT industry who think the answer might be yes?

Alex Brown has not been living on Mars for the last 10 years. In fact, he has probably seen as much of Microsoft's maneuverings as anyone. Does he really expect us to believe he is still in doubt about this?

[ Reply to This | # ]

Vendor lock in, revisited
Authored by: HockeyPuck on Monday, April 21 2008 @ 11:58 AM EDT
Like I said in a recent post, their game is out in the open. No one can support
OOXML today; then how can it possibly be a standard? Not even the inventor can
do it.

The game now is in Microsoft’s favor (unfortunately). They are the only vendor
that can state they support all formats. There are ODF plugins for MS Office
(Open Office already stated they will not support OOXML and others will follow,
even if they could and that too will be brought to others attention by MS).
They will state that there is already a lot of legacy data that only they can
properly render. Of course they will fail to mention they caused this mess in
the first place and OOXML will continue this tradition. They created terms
surrounding OOXML that prohibits GPL like licenses to implement the
specification. So those products are excluded (but they will only use
statemensts like from Open Office Org).

The only answer I see is to send letters and emails to our respective government
representatives urging them to reject OOXML as a standard until MS opens it up
to all developers (not just private). Explain to them ODF is an established
standard, with all the documentation open to all (including Microsoft). OOXML
has public documentation, but it has problems. There is no application today
that supports the specification properly. Explain that even MS has problems with
their own specification. Tell them this could lead to more expensive vendor lock
in, curropt and lost information and we do not wish to pay for the extra expense
involved with OOXML (you must have M$ Office, therefore Window$). ODF and PDF
pretty much supports all platforms, regardless of vendor. They need to force MS
into making it work and open up to all if they insist on shoving it down our
throats. Short of that; it will be rejected as an open spec or never implemented
even if it is on paper.

Think if it this way; would you eat certain meat that is not cooked? Then why in
the world would you “eat” OOXML?

[ Reply to This | # ]

Microsoft Office 2007 Fails OOXML Conformance Tests, Alex Brown Admits, Hopes For the Best
Authored by: Anonymous on Monday, April 21 2008 @ 12:39 PM EDT
The problem is that many people will wrongly believe that there is now an ISO
Standard for the Microsoft Office file format. These people don't understand
that there is now a standard that would be slightly similar to something that
closely resembles something not entirely completely different from the
Microsoft Office file format, if they had done a bit more work and removed a
few
thousand known faults in that standard, and that Microsoft is not
interested in
(1) fixing that standard, and (2) changing the Microsoft Office
file format to be
compatible with that standard.

[ Reply to This | # ]

Microsoft Office 2007 Fails OOXML Conformance Tests, Alex Brown Admits, Hopes For the Best
Authored by: Anonymous on Monday, April 21 2008 @ 01:15 PM EDT
Does anyone remember Cool Hand Luke where the Warden tells Luke “What we have
here is a failure to communicate.”

That is what Microsoft has here a failure to communicate to the open source mob.
What Microsoft defines as standard is what is standard simply because Microsoft
has the power, 90+% market domination. There will be no other standard but
Microsoft as long as Microsoft has a 90+% monopoly. Thus, Microsoft is right and
if you do not believe that ask Microsoft it is they who have the 90+% market
penetration.

Meanwhile Red Hat is busy announcing to the world that it is avoiding desktop
Linux.
http://slashdot.org/index.pl?issue=20080417
http://www.press.redhat.com/2008/04/16/whats-going-on-with-red-hat-desktop-syste
ms-an-update/

and the Linux Foundation is no help either
http://linux.slashdot.org/article.pl?sid=08/04/14/168233&tid=163
http://www.linux.com/feature/132203

[ Reply to This | # ]

This is not a problem
Authored by: overshoot on Monday, April 21 2008 @ 01:18 PM EDT
With IS-29500 headed into maintenance, any problems with the specification can
be corrected to bring IS-29500 into compliance with the standard.

[ Reply to This | # ]

My suggestion to ISO - change the extension..
Authored by: Anonymous on Monday, April 21 2008 @ 02:46 PM EDT
Well, ISO, now that you are in control of the ISO/IEC 29500 format, why not
start with changing the filename extension into something else? And then demand
that any'thing' using that extension to be 100% standards compliant.
The name of the extension could be anything, .ooxml, .isoxml or whatever. Or
even something like "ISO Open Document Text XML format" (.odtx)...
don't think too many would have problems with that.. ;-)

Roy Amundsen

[ Reply to This | # ]

Microsoft Office 2007 Fails OOXML Conformance Tests, Alex Brown Admits, Hopes For the Best
Authored by: Anonymous on Monday, April 21 2008 @ 03:01 PM EDT
But this illustrates why Microsoft is going to have a bit of trouble translating
ISO approval of OOXML into $$$. The purpose of this whole maneuver was so e.g.
Massachusetts could add OOXML to the list of approved open standards, and
Microsoft could back to business as usual. But for this to work, they must
implement the ISO version of the standard. Otherwise, all it takes is one state
senator who's taken an interest in the issue to keep sending letters to the
bureaucracy asking why they're approving a vendor that isn't following their
rules. I know a lot of people here think Microsoft can just buy everyone off,
but you have to remember what Lincoln said: you can't buy all the senators all
the time. (OK, that's what Lincoln should have said.) If Microsoft is
blatantly not following the standard, it'll turn into an unending war of
attrition, and if nothing else endless bad PR for them.

[ Reply to This | # ]

Microsoft Office 2007 Fails OOXML Conformance Tests, Alex Brown Admits, Hopes For the Best
Authored by: nickd on Monday, April 21 2008 @ 04:18 PM EDT
A correction of a small mistake that many people --including PJ in this
article-- have made.
The wrong reasoning goes like this:

fact#1: The Fast Track is supposed to be for already implemented standards
fact#2: MS Office doesn't implement ISO 29500
conclusion: Fast Track was not suitable for ISO 29500

Now, although the conclusion is right, it's not because strictly speaking MS
Office doesn't implement the final ISO29500. You see, specs are indeed supposed
to be improved a little bit as they go through the Fast Track process and
getting improved clearly means that the original --already implemented-- specs
will not be the same as the final ISO specs. The conclusion however is truly
right but because of another reason: the specs are assumed to be improved in
DETAILS and NOT EXTENSIVELY. That is not the case with OOXML / ISO29500 because
the specifications that MS Office implement are miles away from the
specifications in ISO 29500. In other words when (and only when) some already
implemented specifications are assumed to be in quite good shape and are rather
widely accepted by the industry as they are then ISO assumes it's safe to take a
shortcut from the lenghty procedure (aka Fast Track).

Sadly although this was not the case here (and this was evident even before the
BRM) few cared and now we have this crazy situation were the final standard (ISO
29500) is FAR FAR away from the already implemented specification (OOXML). More
over we don't know if the final spec is even internally consistent. That's
because a lot of technical changes were chosen by voting without time to
evaluate their compatibility and the impact of all the changes when taken
together on the final spec (ask any engineer if voting is a good idea when it
comes to technical matters and he will probably tell you that this is the
stupidest idea he ever heard: "hey boys, we must select a resistor for this
circuit but why bother calculating it? Everybody propose a value and let's vote
to approve or disapprove it..." after a few minutes... kaboom!)

---
PS: obligatory excuse: I only take the time to correct PJ (at least I think I am
correcting her) because I admire the high level of her work

[ Reply to This | # ]

Good news?
Authored by: GLJason on Monday, April 21 2008 @ 04:41 PM EDT
Now France can go ahead and add OOXML to their list of approved formats. They
won't be able to purchase Microsoft Office 2007 though because it doesn't
conform to EITHER OOXML or ODF.

[ Reply to This | # ]

  • Good news? - Authored by: Anonymous on Monday, April 21 2008 @ 05:38 PM EDT
  • Good news? - Authored by: Anonymous on Monday, April 21 2008 @ 08:20 PM EDT
    • Good news? - Authored by: Anonymous on Tuesday, April 22 2008 @ 03:28 AM EDT
    • Good news? - Authored by: Anonymous on Tuesday, April 22 2008 @ 05:08 AM EDT
A different spin on Alex Brown's blog entry
Authored by: bugstomper on Monday, April 21 2008 @ 04:46 PM EDT
I really don't see 84 instances of mostly the same error in a 100MB complex
document as being a big deal by itself. If a release of OpenOffice did that well
on a similar ODF document it would merit a bug report or two and a statement
about how well the software was coming along.

What this really demonstrates is how well Microsoft orchestrated the standards
process to keep OOXML as close as possible to a specification of what is in MS
Office 2007 as opposed to a standard for interoperability. The Transitional spec
is the one that sacrifices interoperability and implementability in order to
stay compatible with MS Office 2007. The mere 84 errors show that it succeeds in
its goal.

I have every confidence that Microsoft will make the minor changes to the Office
format to fix those, and then will be able to claim ISO conformance. However,
conformance to the Transitional spec allows MS Office to use all those features
that make it hard for anyone else to implement the standard in a way that would
be interoperable with MS Office. As Alex Brown demonstrated, MS Office is very
far from compliance to the Strict spec, and is likely to never be compatible
with that. Microsoft does not have to conform to Strict to meet their dual goal
of getting government procurements while still locking out other
implementations.

[ Reply to This | # ]

Microsoft Office 2007 Fails OOXML Conformance Tests, Alex Brown Admits, Hopes For the Best
Authored by: Anonymous on Monday, April 21 2008 @ 05:42 PM EDT
Please remind me, when ODF passed ISO, was OpenOffice ISO compliant ?

No it wasn't either.

But you still manage to make all this fuss about Office 2007 not being ISO
compliant 4 weeks after the standard was approved.

You often complain about FUD PJ, but you're a master of the art.

[ Reply to This | # ]

Wasn't the point of the BRM to CHANGE the standard?
Authored by: Anonymous on Monday, April 21 2008 @ 06:52 PM EDT
OK, I do not get this. I thought the point of the BRM was
to make changes to the standard proposal such that it
becomes acceptable to the NBs. Now, we know that 80+% of
the issues were not even discussed, but some changes were
approved anyhow. So that would make it pretty obvious that
any software written before the BRM (like Office-2007)
could not possibly be compatible with the final accepted
version of the standard.

So what was the point of this test ?
Of course, there is no software compatible with the
standard at this point.
The better question is: will there ever be one? If yes,
when ?

Zs.Zs.

[ Reply to This | # ]

Microsoft Office 2007 Fails OOXML Conformance Tests, Alex Brown Admits, Hopes For the Best
Authored by: Anonymous on Monday, April 21 2008 @ 07:19 PM EDT
Potentially on the bright side, MS may have shot themselves in the foot. After
all when a standard is published, applications are expected to conform to that
standard. If they don't there are organisations who will not adopt the
non-conforming product as a result.

We may see OpenOffice, KOffice, maybe say...Wordperfect incorporating the OOXML
Standard. If they do we would assume they would conform to the standard as
preserved by ISO and that could become a selling point.

Let's say OpenOffice conforms to ISO OOXML Standard and passes its conformance
tests with, say 98% -100% pass rate .If this happened it would become a selling
point for OpenOffice etc, and may become a stumbling block for MS Office.

[ Reply to This | # ]

122000 errors, is it a record?
Authored by: Anonymous on Monday, April 21 2008 @ 07:49 PM EDT
I just want to know, is failing the ISO test with 122,000 errors a record?

Or has anything been worse?

seriously. n/t

it's gotta be up there!

crap when I read that news I spewed my drink all over my keyboard, M$ owes me a
new one!

[ Reply to This | # ]

Fix what? The Standard or Office?
Authored by: Anonymous on Monday, April 21 2008 @ 09:23 PM EDT
PJ wrote: "I know. After ISO was captured by Microsoft, nobody cares about
rules any more. Brown hopes Microsoft will be a good doo-bee and fix
this."


Fix what, the Standard, or the Product?

Seems easier if they just fix the standard. Isn't that what Brown suggested
that Maintenance mode of ISO standards is for?

[ Reply to This | # ]

ODF Compliance
Authored by: Anonymous on Tuesday, April 22 2008 @ 04:38 AM EDT
Alex Brown misses some important points about the compliance of Office suites
with the ODF standard.

The first is that by picking on OpenOffice he shows his true colours as
anti-FOSS. There are several other heavily used suites that also use ODF,
GoogleDocs springs to mind.

But that fact that there are several implementations of ODF with no one
implementation having priority mean that there is a competitive edge to being
compliant with the standard.

At present with OOXML there is no point in anybody (other than MS) trying to
implement it. The standard will change to favour MS leaving everybody else out
in the cold. For other vendors to implement the standard would effectively mean
cutting their own throat by locking themselves into an attempt at compatibility
with MS (ho ho ho what fun).

I also suspect that MS will find it in their own interest to be almost in
compliance withthe standard, but not quite, like HTML.

[ Reply to This | # ]

Microsoft Office 2007 Fails OOXML Conformance Tests, Alex Brown Admits, Hopes For the Best
Authored by: Anonymous on Tuesday, April 22 2008 @ 11:16 AM EDT
All right. Fast track is for an ***ESTABLISHED*** product.

*****THERE IS NO PRODUCT THAT CONFORMS TO OOXML******

words fail me

[ Reply to This | # ]

  • Why - Authored by: Anonymous on Wednesday, April 23 2008 @ 02:31 PM EDT
Microsoft Office 2007 Fails OOXML Conformance Tests, Alex Brown Admits, Hopes For the Best
Authored by: Anonymous on Tuesday, April 22 2008 @ 12:59 PM EDT
call me dumb and old country hick but,

wouldn't you try this before approving it as a standard?

[ Reply to This | # ]

A Brown's Red Herring Effort
Authored by: Jose on Wednesday, April 23 2008 @ 01:43 PM EDT
PJ,

The gist of a particular discussion here
http://www.linuxtoday.com/news_story.php3?ltsn=2008-04-21-014-26-NW-MS-LL is
that schema conformance is a terribly low bar to pass or approach. I can write
something that conforms to a schema without even being able to render or process
a document in any way. Alternatively, you can write something that produces
valid schemas but does nothing at all like the standard dictates. In other
words, what might pass as syntactical rules (schema conformance) vs what might
pass as semantics (how those syntactical chunks map to behavior or are further
restricted based on content that would be impractical to convert to syntax
restrictions) are generally considered to be independent.

The end portion of this reply
http://www.linuxtoday.com/news_story.php3?ltsn=2008-04-21-014-26-NW-MS-LL-0005 ,
also from the same thread as above, suggests that Microsoft can implement all
the secrets of their current closed formats (if they wanted to, though I expect
they have revealed some of those secrets and will now go for new secrets) while
maintaining schema conformance or at least getting very close.

I have not looked at the actual OOXML spec, but it takes very little in any XML
schema (iirc) to allow a conformant app to get away with murder. I would find it
virtually unbelievable a claim that OOXML provided less than a bounty of ways to
get away with such murder.

Alex Brown seems to be performing the part of an MS accomplice very well.
Perhaps that blog entry was created to make up for his transgressions, ie, for
the MS negative comments he made that were caught on groklaw a little earlier
(the blog may have even been directly suggested by someone closer to
Microsoft).

The quoted portions below are a bit long. If you want the short version, it's
just what I stated above (the quoted comments cover more items and also a
running example, reproduced here for those with time on their hands).

*****************
>> Well, I couldn't say what Alex B's motivations are, but I think you're
missing the point.

I understand your point, but I'm afraid that SOSOXML/UhOhXML approval by ISO
effectively gave Microsoft that checkbox regardless of exactly how faithful they
are to the standard. [This doesn't mean there won't be some fighting up ahead,
but the hard part probably is over for them if the game plays out at all like in
the past.]

Maybe Alex meant well, but I am very suspect. In the real world, you can wreck
havoc while following the standard's syntax, and Alex played up conformance to
the syntax as the only game that matters. That would be like forgetting that the
same legal page renders differently in IE vs Firefox [though most IE pages have
extra content as well] because the prior Monopoly browser simply rendered the
standard elements in a "buggy" fashion. Conformance defined solely on
a schema is basically a joke if the intention is faithfulness and
interoperability (interop not in terms of exchanging docs but in terms of
matching specific syntax to specific behavior so that the rendering is constant
across competing products).

Plus, Microsoft will argue that 98% of OOXML (or 9x% + extra extensions) is
better than 100% of ODF "because ODF is inferior," etc. Microsoft
yielded on some items because they know they just have to be close. In fact, a
healthy percentage of bugs in OO.org (which likely has many outstanding bugs at
any given time) are likely to be bugs that violate the standard in some way
however insignificant. Microsoft will be able to point that out and then
"conclude" that 98% faithfulness is normal for software. They'll use
the old "no one is perfect argument" to justify them being that much
worse than the competition in practice. And most people will actually be
attracted to that kind of argument, btw.

So while I am not saying that we should ignore that Microsoft is not conformant,
we should not put too many eggs in that basket lest we take value from what
really matters. This is why I am saying that Alex Brown effectively is helping
to change the focus of the game. This would allow Microsoft to get a free pass
when they conform to the Schema, leading most people/customers to overlook the
real crimes.

This approach *may* work [heck, old Al C was caught on tax evasion], but it is
very risky IMO because schema conformance can be very easy.. especially since
Microsoft controls what ends up being the official schemas. In fact, I think
they are likely trying to play off that only the transitional schema is likely
to ever be properly implemented by anyone. That low hurdle is the one they will
hold themselves to, hoping that third parties waste their time shooting for the
much higher schema hurdle since the standard will change again "just in
time" for Microsoft to remain in the lead or at least competitive.

Again, Microsoft has always used the checkbox effect for software that has not
complied (maybe some release was good enough to pass inspections, but they
immediately shipped something else). They know the game. Those that go to the
trouble to require checkbox' probably don't care to do rigorous, costly,
annoying, better-not-be-failed, periodic testing, testing that doesn't even
exist today or for some time because of the huge pile of manure that is UhOhXML.
[and MS's test suite will be tuned so that they leave enough holes and so that
it matches their software.]

OK, well, Alex B can continue to hold MS' feet to the "fire". That may
help; however, note he ended the blog by pointing out how much worse Openoffice
is likely to be. That path puts MSO on a higher plane than all the competitors.
This focus attempts to shift the game to OOXML schema conformance and puts
pressure on the Openoffice people to run the rat race instead of becoming
stronger and growing ODF. That blog entry, like most things coming from
Microsoft, should be quickly put in its place with the bait not taken.

>> And frankly, near as I can make out from all the commentary I've read
on OOXML, it's practically impossible for anyone, *even Microsoft*, to actually
comply with it.

It appears there are at least 2 schemas.

The thing is that Microsoft is not likely to ever comply with the full
*behavior* and components as specified by UhOhXML even though they mostly
control it. But that is not what is being measured by Alex B. By focusing on
schema compliance (the syntax), he just removed the real obstacle to MS
compliance (behavior) while simultaneously pointing out that MS is likely to be
leading in that (by comparison) insignificant race by a large margin over
Openoffice.

Microsoft will out market OO.o and use blog content of dudes like A B in the
process.

The worst thing OO.o can do is to get distracted. Microsoft will probably close
on the checkbox buy they made within 2 months' time no matter what OO.o does.
But now it's back to improving OO.o and showing that Freedomware is the best buy
out there in terms of overall value and safety of documents.

Focus on all the way MS fails the customer and OO does not. Alex B should not be
trying to turn the game away from what matters or lowering the "pass"
hurdle for Microsoft.

[I would not feel that badly if A B had stated, a la Rob Weir, that
"despite the meaninglessness of simply meeting the schema of their own
standard, Microsoft had failed in just that. Meanwhile OO.o is doing a much
better job meeting industry approved schemas and behavior of the older more
established ODF." But no, A B is setting the stage to let Microsoft off
easily while making OO take one on the chin. Disgusting is what comes to mind.]
*****************

App A speaks English (OOXML).

App MSO speaks English+French (MSOOXML).

Some user uses MSO to create a document of eating of cake and drinking of orange
juice.

The document is encoded by MSO as English and French (the "French"
shown here as English within brackets to simplify things for LT readers). It
says:
"Eat a rat [no, make that a cake], then drink some blood [well, not blood
but orange juice]."

The MSO reader understands English and French so can render the proper document
on screen to show the eating of a cake and drinking of orange juice as the
author intended. App A only understands English and will get very confused,
rendering instead the eating of rats and drinking of blood plus possibly some
gobbledygook ..or not since that stuff (written in "French") may
translate into a comment in English so is not shown on screen or perhaps
translates into a command that adjusts the color a little to make the blood
appear darker red than it might otherwise appear. Whatever the English
translation of the French, the document will be rendered by A very differently
than intended.

Yet, the document can be syntactically accurate according to some schema that
allows anything that looks like a word in any human language to pass as legal.

In reality, there is no need for MSOOXML to understand and use a full
"secret" language like French. It simply needs to understand a
Perverted English that appears like English in almost every way but where some
phrases mean different things in Perverted English than in English with the
occasional nonEnglish word thrown in as allowed by parts of the schema.

Thus MSO follows the schema. Further, it might even follow the semantics in most
places, just not everywhere (Perverted English).

Of course, you can follow a schema while introducing a whole new
interpretation/rendering. Thus MSO can range from (case A) appearing to
implement OOXML almost correctly all the way to (case B) clearly violating the
spec (but while still "conforming"). This gives MSO's author maximum
flexibility to fool others into running the rat race thinking they can keep up
(case A), while still being able to preserve the Monopoly amid heavy competition
by twisting the standard so that no one can follow it yet claim some sort of
conformance (case B).

And they can even do both at the same time. For example, a simple flag might be
turned off for most documents, giving the impression that MSO almost complies.
Later, a slight difference can give the impression MSO is even closer to
complying; however, another slight change can lead to the document being
rendered with no real accuracy except by MSO.

The flag tells MSO what mode it should use to interpret the rest of the
document, and this tiny flag would be hidden among the entire of the huge file.
The change can occur for documents in general over time as Microsoft requires
for competitive reasons, or it can happen even for old documents once they get
opened (even if the author didn't want to save the changes). In fact,
controlling Windows/Vista, means that Microsoft can "adjust" all your
documents while you sleep one night. This might affect archived documents and
everything else (heck, it might corrupt ODF documents).

Let me return to the example above:
"Follow path A. Path A is: Drink orange juice and eat cake. Path B is:
Drink blood and eat rats."

Both apps render the above properly; however, with a slight change, we get back
to only MSO rendering as intended by the author.

"Follow path B [no, A]. Path A is: Drink orange juice and eat cake. Path B
is: Drink blood and eat rats."

Another example, might be instructions on getting to the South Pole (to find
southern penguins). The instructions could easily be changed to lead you to the
North Pole instead. It just takes a tiny item (a negation of a crucial
direction) to end up traveling north instead of south. This can happen while you
sleep during the night before the trip (if you run Monopolyware on your PC). And
the document would not appear any different if you look at it through MSO the
next morning. It might even appear very similar if you analyze it under a hex
decoder. But it will render very differently on XYZOffice. Yikes! Hope you
learned Northern Penguin English since the laptop you are taking on the trip
uses XYZOffice, a different "OOXML conformant application."

The possibilities truly are limitless. [Go ahead and play around with the
above.. go ahead, really.. feel like a powerful dirty Monopolist for a day.]
..And that is why you won't catch me using a closed source Monopoly product
except under extreme circumstances, standard or no standard.

And Alex Brown appears to me to be trying to distract us from recognizing the
big elephant in the room by shouting... "Hey, did you guys notice the new
doorbell! Wow, it plays a few different notes than the old one. Let's see how
close we can get it to the old doorbell chime by changing the room temperature.
Let's play a game and keep trying until we get it just right, OK? We all want
that, don't we? Of course we do."
*****************

[ Reply to This | # ]

ODF or OOXML?
Authored by: Jose on Wednesday, April 23 2008 @ 07:52 PM EDT
Simple, if one values competition and/or the preservation of archived material,
go with the format best implemented by the open source apps that provide the
features you need.

If you then use a closed source application, make sure there is a way to limit
the output to the features of the standard (which presumably are or soon will be
implemented by the open source applications), and make sure the documents you
care about were tested and render appropriately within the open source app you
want as a backup.

In short, if you care about preservation of documents, you need to
open/view/test all documents created with closed source software from the open
source software to verify that the documents render (and function)
appropriately. This should be done each and every time you save the document
since the closed app can change things around on a later save in a way that
cannot be duplicated by the open source apps [I believe there are documented
cases of Microsoft apps being caught changing things around gratuitously in a
way that breaks something else]. Instead of going through the trouble every
time, consider getting in writing certain guarantees from the vendor.

Needless to say, you should not be using a closed source operating environment
if you can help it since the document can be changed at any time without you
being aware. For example, the OS may share and use the same software libraries
used by the proprietary office suite (if owned by the same vendor) to open/check
documents while you sleep at night as part of some security check. During any of
these document accesses, the file might change if the libraries so allow it (eg,
to update the document with a timestamp perhaps). That process could break
compatibility, bringing to an unfortunate end all the prior work done to ensure
compatibility for the affected document or documents.

[ Reply to This | # ]
Groklaw © Copyright 2003-2013 Pamela Jones.
All trademarks and copyrights on this page are owned by their respective owners.
Comments are owned by the individual posters.

PJ's articles are licensed under a Creative Commons License. ( Details ) 		Site layout based on Woodlands theme by Bryan Bell.
Groklaw logo by John Crowley.
News Picks logo by Ted Thompson.
Powered By GeekLog
Created this page in 0.06 seconds