Microsoft Deprecating Some OOXML Functionality

christian.einfeldt writes "According to open standards advocate Russell Ossendryver, Microsoft will be deprecating certain functionality in its Microsoft Office Open XML specification. Ossendryver says the move is an attempt to quiet critics of the specification in the run up to the crucial February ISO vote. The Microsoft-led industry standards group formally offering OOXML confirms in a 21 December 2007 announcement that issues related to the 'leap year bug', VML, compatibility settings such as 'AutoSpaceLikeWord95' and others will be 'extracted from the main specification and relocated to an independent annex in DIS 29500 for deprecated functionality.'"

deprecated but widely used by MS software?

If MS deprecates it but makes support for the deprecated features the default option in their software, they'll still be contributing to people spewing incompatible files that don't render correctly in software following the standards. It'd be better to just rip out the parts that shouldn't be there and resubmit the standard. Having to recognize and either support or report lack of support for a maze of twisty little semi-standard features for sake of backwards compatibility is not going to help the situation much,

Re:deprecated but widely used by MS software?

I disagree. If you add up all of the letters in the words "de facto" and then multiply that by the number of times the phrase "de facto standard" is written, you can see that getting certified as true standard will save massive amounts of disk space, paper, and toner.

Re:deprecated but widely used by MS software?

You are in a Microsoft Office. To the North, is a door, you hear what sounds like chairs being thrown. To the South, you see an open door, but a very dark room.

What do you do?

InnerWeb

Re:deprecated but widely used by MS software?

Do you actually believe windows developers actually want to touch OO?

Being a Windows developer (I also develop on other platforms, but this is irrelevant) I can tell you that I don't want to touch OOo or Microsoft Office. If I want to generate a document, I want the ability to do it myself if need be and so on. So far, OOXML is not really the solution I've been lucking for. The documentation is appalling, what is considered a very simple document and 'correct' in the specification does not work with Microsoft Office. It's annoying.

But in the end it makes development easier.

Honestly, generating odt, ods documents is easier.

Microsoft takes care of developers.

I don't agree. The MSDN is one huge example. It's great that it has such a vast knowledge base. Unfortunately a third of the documentation contains the wrong behavior and possibly the worst workarounds I have ever seen for a problem with lots of empty promises to fix issues in the future that haven't been fixed. That is how well Microsoft takes care of the developers.

Developers are the ones who make this who software word turn around.

I disagree. Microsoft has been buying out software companies that are a threat, launching extremely anti-competitive campaigns against 3rd party software developers that rival their own products and so on. Microsoft are the ones making the software world go round because they are manipulating everything, not the developers or the consumers.

It's about documentation, support and tools.And face it...that is a field where microsoft has no competition.

Have you even seen Microsoft's support? Staying on the telephone for hours on corporate support to report a serious bug, only to get a idiot who doesn't seem to even grasp Microsoft's own products (and this has been on every occasion I have tried to do anything with the enterprise support). Microsoft does not even provide direct support options to consumers, never mind small-time developers.

Let's see what the Linux side does.

Do they have corporate support? Yes.
Does Microsoft? Yes.
Do they have 24/7 corporate call centers and fast pick up rates? Yes.
Microsoft does not have 24/7 corporate call centers.
Do they have people who know what the hell they're talking about on the otherside? Yes -- I have even got a kernel developer at one point when it came to a serious issue.
Does Microsoft? Well, in my experience - never managed to talk to anyone who seemed to actually know.
Does the regular consumer get support? Yes - obviously paid support is better but there are a lot of free alternatives that appear to be just as good (despite the people who perceive Linux support is 'rtfm').
Does Microsoft? There is a community that provides support on issues, but there is no paid support options offered at all.

Googling up for solutions and looking them up in forums(do i hear linux users trying to say smth?) is not acceptable.

Try researching the subject a little before opening your mouth next time.

Re:deprecated but widely used by MS software?

Microsoft takes care of developers.

I don't agree. The MSDN is one huge example. It's great that it has such a vast knowledge base. Unfortunately a third of the documentation contains the wrong behavior and possibly the worst workarounds I have ever seen for a problem with lots of empty promises to fix issues in the future that haven't been fixed. That is how well Microsoft takes care of the developers.

+100.

Unfortunate reality is that M$ provides nearly complete (== always incomplete) solutions. Up side is that you can base your business on it. Down side - you are locked into M$ solutions. But you heard that hundred times already. But what everybody's missing is development side: developers working solely on M$ platforms turn slowly into agoraphobic drones who would claim that "M$ is best" just because they do not know anything better.

Many of my versity friends turned into such drones - even most reasonable ones. M$ keeps feeding them with new (presumably better) APIs and they just keep their minds piped directly into their beloved MSDN subscriptions. 5 (or 6?) data base APIs? And M$ still keep printing them. 6 IPC APIs? - OLE, OLE2, ActiveX, COM, DCOM, COM+ - but M$ doesn't stop the printing press.

"Windows is better because it has API [XXX] and [Linux/Mac OS X/etc] doesn't." Explaining people that API does solve Windows specific problem which doesn't exist on Linux nor Mac OS X just doesn't work - because they never touched them. And they will never touch them because they do not have the M$Windows' hundreds APIs. (Recent best example was ASIO [wikipedia.org] - and fact that only Windows does support it.)

Smoke and Mirrors

The result of Microsoft's manipulation should be ISO banning MSOOXML from participating in the standards process.

It's abundantly clear now that the format is critically flawed and cannot be implemented by anyone, not even the Office team themselves.

ECMA 376 is a bomb disguised as a standard. It redefines functions and components just to retain ties to the undocumented legacy formats. Therefore a number of things that should be fixed by now, thanks to better engineering, and existing ISO standards, are left not only unfixed, but even perpetuated by ECMA376.

The fact that Microsoft continues to push this fake "standard" shows how little they care about their customers and how much their business is predicated on lockin.

Re:Smoke and Mirrors

It is no more a "fake" standard than any other standard.

You may not see a pattern here. I suspect may others will.

• Maybe we could define the APIs so that they work well with NT and not the others even if they are open. Or maybe we could patent something related to this. - William Henry Gates III on ACPI, 1999
• Screw Sun, cross-platform will never work. Let's move on and steal the Java language. - Prashant Sridharan, MS Visual J++ Product Manager, 1997
• The first obligation that the ICPs undertook was to distribute Internet Explorer and no "Other Browser" in connection with any custom Web browsing software or CD-ROM content that they might offer - US District Court of Columbia on proprietary HTML extensions, 1999
• OSS projects have been able to gain a foothold in many server applications because of the wide utility of highly commoditized simple protocols. By extending these protocols and developing new ones, we can deny OSS projects entry into the market. - Vinod Vallopillil, ex-MS Engineer, 1999
• The first type of conduct found to constitute an abuse consisted in Microsoft's refusal to supply its competitors with interoperability information and to authorize them to use that information to develop and distribute products competing with its own products - EC First Instance Court, 2007
• We [Microsoft] are OASIS members but since we didn't have an interest in ODF we didn't participate in its development. - Brian Jones , MS Office Program Manager, 2007

This isn't a battle between OOXML and ISO. It is a battle between having document standards and not having them.

Microsoft is trying to wreck the concept of standards and interoperability to a point where those concepts are useless.

About right.

Sounds consistent with the way Microsoft works. Promise the moon and deliver a crater. It was their intention all along. Propose something that smacks everyones senses with a bat, then back off with something that sound more reasonable, even though it is not.

Deprecated means forever

In another move to spread more FUD, now they're trying to hide the UGLY part of the specification. But, what use is hiding it? They claim the deprecated features will be used only for the migration of old binary formats, and that they should not be used by new documents... But considering that the whole point of this document format standardization effort is to be able to open any document in 20 or 30 years time, and if the old binary format documents will be converted using deprecated features, that just means that any software implementing the standard will have to support the deprecated features anyway...

Although they keep manipulating, manipulating, and manipulating more, I still think their format stinks, they're only using it to spread FUD over other formats, and I really hope they can't pull this stunt.

Re:Deprecated means forever

And VML is used in Office 2007... see this openmalaysia blog post [openmalaysiablog.com]. Marking things as deprecated just means that it's discouraged, and it doesn't mean that there's a modern replacement (like how in HTML FONT was deprecated and it's replacement was CSS functionality). VML is still a necessary part of OOXML, so marking it as deprecated doesn't actually help developers. What would help them is if DrawingML could be used in all the places that VML can be. Infact, as the CNS (Microsoft's covenant not to sue) specifically excludes patent coverage over non-required features this means that we may now be lacking patent coverage over VML. Can anyone from Microsoft comment on this? (the Microsoft OSP might grant coverage if that "Necessary Claims" means normative, as they claim)

Re:Deprecated means forever

They claim the deprecated features will be used only for the migration of old binary formats, and that they should not be used by new documents...

Someone help me out here, for real. I think I'm missing something. What is the point of those ridiculous "backward compatibility" tags? Word's never been good enough for pixel-perfect rendering. For example, printing the same document on different printers hasn't ever been likely to give the same output. So, what on earth is the justification for maintaining a "renderLikeWord95" tag when that was never well-defined to begin with?

If the <foo> attribute originally meant "centered, bold, double-spaced", then just make the importer translate it to something like "<textblock align="center" weight="bold" height="200%"> text goes here </textblock>". Forget bug compatibility. That's a dying horse and needs killed now before we end up with something like the loose HTML parsing nightmare that browser designed are stuck with. Who cares how the document originally displayed on the original machine? MS never did before today.

Don't hide those tags - delete them. There is no rational explanation other than lock-in for having them, and as long as they're around, the IT world will know this is a joke.

A superb standard

I'd just like to remind everyone that OOXML is a superb standard.

-- Miguel

What about...

the AutoSpaceMonaLisasGapBetweenHerTeethButOnlyWhenShesSmilingLikeWord95BBQ compability?

AutoSpaceLikeWord95

> settings such as 'AutoSpaceLikeWord95' and others will be extracted

Lets just hope they keep the 'WaveYourArmsInTheAirLikeYouJustDontCare' setting.

How is OOXML good anyone but Microsoft?

I've done quite a bit of reading, and listening on the topic of OOXML, and I have come to the conclusion that there is no good (at least of the technical kind) in OOXML. Yet, people seem convinced that Microsoft is a "good" company. And a good company wouldn't actively push something that was obviously without any good for the industry... so I must be missing something. I generally just think that it is for the purpose of profit and control, but every now and then I like to give opposing views a chance - since I may be the one wrong.

This is an unsurprising move

Just like with Vista, they just drop features until it's "releasable."

Here's the obvious problem:

They will claim a feature is deprecated, or not part of the spec, but their software will continue using it. Meanwhile, other programs that try to read and write OOXML format following the "official" spec, will result in the documents created or edited by other programs not being fully compatible with MS Word. This will be seen by the user community as a deficiency in the alternative software and no as a problem with Microsoft's software.

We have seen this before and we continue to see it. People think that because a web site works with MSIE and doesn't work with Firefox that there's a problem with Firefox... Microsoft continues to damage the competition in this way and will persist in the same. I hope that the voters in the ISO decisions are aware of this potential problem.

I can almost guarantee this...

'AutoSpaceLikeWord95' and others will be 'extracted from the main specification and relocated to an independent annex in DIS 29500 for deprecated functionality.'

...that, that so called deprecated functionality will be "re-introduced" in an update to enhance the user experience and security at some later date. When this happens, part of hell will break lose and we'll be back here at Slashdot debating this and that.

How about a better description?

I'll take the grammar Nazi hit..."deprecating"? One usually deprecates oneself in every use I've ever read. While the use might be technically correct it certainly isn't current usage. I'd think "minimizes" or "plays to lowest common denominator" might be a whole lot more understandable.

From m-w.com

"Entry Word:
        deprecate
Function:
        verb

Text: 1 to express scornfully one's low opinion of -- see decry 1 2 to hold an unfavorable opinion of -- see disapprove"

bah...humbug...

Will these be removed from MICROSOFT OFFICE?

It is highly doubtful that the "deprecated functionality" will be removed from Microsoft Office. Therefore if they get the revised OOXML passed as a standard, anyone who uses Microsoft Office based on its claim to be OOXML will have been the victim of a bait-and-switch tactic.

But I suspect that was the goal all along. Orgs that just wanted to use Microsoft Office in the first place would be able to say "see, this is open" and keep doing what they were doing.

Well, at least it's somewhat documented, making it somewhat easier than .doc/ppt/xls for the free world to reverse engineer.

Sneaky?

This seems sneaky to me. Remove controvesial stuff from the standard, but put it in an Annex, that MS will implement and people will rely upon left and right, so it will become a de-facto microsoft embrace-and-extend standard.

I really try to fight the kneejerk anti-microsoft sentiment around here, but lordy, all of their moves seem so calculated and evil. It's not just single actions, it's a pattern of actions. Humans are great at recognizing patterns. And even with good moves and bad moves, one can generally see a positive attitude behind Google, for example (some may disagree, but I think the general consensus is that they're not dastardly.) But with MS, every move seems like a piece of a puzzle showing a nasty, calculated, aggressive, anti-competitive entity. Everything seems consistent with that. The way the US rolled over on everything for political reasons is shameful. Hopefully the EU will right some of those wrongs, at least in part of the world.

I guess to try and find the bright side, one could say "at least it's documented" (without an exorbitant fee and crazy restrictions, like SMB et al.)

Microsoft Deprecates ... Functionality

But, of course! What else could they deprecate?

The big problem with this

IMO there's nothing wrong with the decision to deprecate some of the most revulsive misfeatures of OOXML, but there's the very real problem that this could lead some people (in particular in the national standardization bodies that will have the opportunity in March to change their vote about OOXML) to think that these relatively minor changes somehow make OOXML suitable for acceptance as a "standard".

If you agree that this is a real risk, and you're willing to help with doing something about it, please join us at OpenISO.org [openiso.org] and help put together a "problem report" document about OOXML that explains the main issues clearly.

The debate is maturing

Nice to see that the comments thread on OOXML is shrinking as the debate matures. Of course that means that the usual trolls are either bored or on holidays but I think that we may collectively be starting to better understand what's going on.

I attended the UNSW Cyberlaw centre forum on OOXML http://www.cyberlawcentre.org/2007/ooxml/ [cyberlawcentre.org] as an interested observer and I liked what I saw. Smart people engaged in a positive discussion. Yes, the viewpoints were polar, but the words were civil and a real exchange of ideas took place.

Pia Waugh was an organiser of the event and had this to say about it: http://tinyurl.com/32zfsr [tinyurl.com]

Roll on the BRM in Geneva and may reasoned debate rule over Fear, Uncertainty and Doubt.

--

digipres, a No voter

Clear evidence OOXML is crap

The format is, in effect, still in development and there is already an independent annex dedicated to deprecated functionality. How exactly is it possible for a decent format to have deprecated features even before it exists?

While we're at it...

Well, While we're at it, why not deprecate the whole shit and get the problem solved for good?

Just wait

In a moment now he will start worrying about precious fluids.

Depracation May Make Things WORSE

Deprecating the functions doesn't mean that MS isn't going to be using them -- it just means that they don't want anybody else to use them. It's also possible that, by removing those functions to other sections, they're removing what little patent protection developers might have for implementing that functionality.... That's because MS's patent protection only clearly applies to features that are well-documented in the definition documents.

So: The results of this deprecation could be (in the worst case):

• Developers will be deluded into believing that MS isn't going to be using this functionality (e.g. how Ashton Tate was convinced, by Microsoft to focus their Lotus 123 efforts on OS/2, while MS was focusing their Excel development on Windows 3).
• Developers who develop for those capabilities could face enhanced legal exposure
• documentation for this functionality could be decreased rather than increased (which is what most people want -- given that MS is currently using this functionality).

It's getting harder and harder to deny that MS is attempting to create a standard that their software will neither read nor write properly -- but that they can claim that they're using (because they control both the trademark and the "defacto standard" implementation of).

Re:Now the waiting game...

The criticism that OOXML is basically unnecessary for anyone other than microsoft still hasn't been adequately addressed. This is like microsoft proposing MicroMiles should be an international standard because they don't control the implementation of kilometers. They could have just contributed to ODF if they were remotely interested in useful standardisation, they were given ample opportunity.

The criticism that the standard may be patent-monopoly-encumbered hasn't been adequately addressed (but that is unfortunately pretty typical of "old" standards bodies like ISO and ECMA and ITU). Really, that's not OOXML-specific, far too many "standards" are unimplementable freely (free licensing should be a basic legal requirement for any national-government-recognised standards but isn't... yet. The baby boomers are trying to get as much corruption in as possible before they become decrepit and we take over and have to clean up their mess).

Re:Now the waiting game...

Sure, because nobody cares about subtle trolling.

Seriously, the Slashdot moderation feature is there so that a "vocal" minority can hide or boost a comment because they dislike or agree with it respectively. It's a great service for those who prefer other people to do their thinking for them.

Re:Nope

In software, deprecate means to officially warn that they're planning to remove the feature in question later - after everyone else (like other software that might assume said feature is present) has had a fair chance to prepare for its removal.

In a standard, deprecated can mean something like it does for software, or it can refer to stuff that's optional or old and that usually has an alternative you're encouraged to use instead.

Re:*BSD's Final Christmas

Netcraft confirms it.

Re:Now the waiting game...

Please point out my obvious troll! I made a serious point. A lot of people on Slashdot only hate the idea of OOXML because Microsoft wrote it. Now, are you going to discuss it or hide behind the moderation system?

No, we hate it because:

1. It is not meant as a specification to be implemented openly by others -- it was only introduced to quash a competing, already ratified standard so as to maintain Microsoft's Office Monopoly.
2. It is not an open standard -- it does not promise to be patent unencumbered.
3. As provided, it cannot be implemented by anyone other than MicroSoft, since it relies on other specifications that are not open, and that are unavailable to other implementers.
4. That MicroSoft has made some things "deprecated" is a smokescreen. Many things in, say, Java, or C are "deprecated". That does not relieve a developer of the need to implement those features anyway in order to maintain backward compatibility with older documents/programs.

Aside form the above and a few other things, it's a great standard... </sarcasm>
Now -- would you care to tell us who you are and who you actually work for "Anonymous"?