Microsoft's OOXML Formulas Could Be Dangerous

hill101 writes "According to Rob Weir's blog, Microsoft's 325-page OOXML specification for spreadsheet formulas is deeply flawed. From basic trigonometric functions that forget to specify units, to statistical functions, to critical financial functions — the specification does not contain correct formulas that could possibly be implemented in an interoperable way. Quoting Mr. Weir: 'It has incorrect formulas that, if implemented according to the standard, may cause loss of life, property, and capital... Shame on all those who praised and continue to praise the OOXML formula specification without actually reading it.'"

So?

[ColaMan]

.....if implemented according to the standard, may cause loss of life, property, and capital..

Pffft....as if this has ever been much concern to software manufacturers before.

Every EULA has boilerplate text denying all responsibility , and you'd be mad to trust any results from software implicitly. Double check it yourself , even if it's just a few corner cases.

Typical Microsoft...

[QuietLagoon]

... put out garbage into the marketplace, and then wait for the customers to do the quality assurance work that Microsoft should have done.

The trouble is that the politicians standardizing on this spec will look only at its length and declare it to be good. Maybe Microsoft made the specification long with that intent in mind.

Meaningless statement

[Anonymous Coward]

Shame on all those who praised and continue to praise the OOXML formula specification without actually reading it.

What percentage of those who praise ODF specifications actually read it? Or any other specification? I would imagine it is a small percentage.

Surprised?

[JohnFluxx]

I doubt anyone is surprised. How can you possibly fast track a 325 page document, giving the public only a time amount of time to check it, then expect it to be perfect.

Man, I really really get annoyed at Microsoft.

Re:So?

[erroneus]

A user NOT trusting his tools is a very strange thing. If it were some sort of software engineer doubting software tools, that's one thing and it's somewhat expected. But in general:

* We trust all hand tools like wrenches and sockets to be exactly the size on the label
* We trust all of our doctor's opinions whether or not a second opinion is recommended
* We trust our math applications to do math properly
* We trust our spell checkers to check properly

In general, we trust the things we by to work as expected... as advertised. (No, I haven't seen Excel advertised to be accurate, but in a math application, it's implied by its very existence) So to say that you should re-check the results by hand is not just ridiculous, it would never happen.

I remember when the Pentium processor first came out and there was this math error in there somewhere. It was a BIG deal.

But before passing too much judgment on this too quickly, a little verification of the bugs might be helpful and let's mark our calendars to see how fast Microsoft fixes the problem... oh wait, the problem is said to be in the file specification? What does that mean if they update the format specification with regards to their ISO certification?

Yeah, I'm sure this guy is objective

[WalterGR]

Who is the author, Rob Weir?

I work for IBM, as a performance architect, as well as on various ODF technical topics. (source [robweir.com])

So a guy working on a different document format, for a company who competes with Microsoft, has unkind words? Color me shocked.

OOXML defines spreadsheet formulas, and ODF doesn't. The Microsoft boosters have been parroting the party line for quite some time.

Uh... ODF doesn't define spreadsheet formats. There's no standard for spreadsheets in ODF. How is that "parroting the party line?"

Re:Yeah, I'm sure this guy is objective

[january05]

ODF will define spreadsheet formulas, in the next version. And come on, the "IBM conspiracy" take from MS is really lame since OOXML is the one with proprietary patented extensions. I'll take any open standards company I can get, personally.

Not a good thing, because it is not a free format.

[master_p]

It may be open, but it is not free, i.e. the required changes can not be done by third parties or by a committee and then used by Microsoft. Microsoft wouldn't do anything that would hurt its embrace and extend business model, and OOXML follows that logic as well(it's so huge and flawed that no one dares using it).

I never understand why people complain so much

[pembo13]

Let MS do exactly what they want, they seem quite successful at it, if it bites them in the butt, so be it. I would just like our own software freedoms to be preserved. I have no intention on producing anything with their format, I'm sure I'll eventually have to read it, but the chances that the receiver of a document is liable for inaccurate content within that document seems very low.

What is the motivation, since I'm sure there must be a good one, to do this free work for MS?

Re:Just want to point out...

[Xiaran]

The problem is that we are talking about a proposed international standard and you are using the phrase "it's goddamn assumed by anyone...". There should ideally be *no* assumptions in a stadard... it needs to be as clear and accurate as humanly possible. Remember that once a standard is published it is translated into many laguages and possibly implemented in different cultures as mentioned in TFA. What you assume to be obvious may or may not be obvious to others.

Re:So?

[tsa]

In general:

* We trust all hand tools like wrenches and sockets to be exactly the size on the label

OK, I'm with you here.

* We trust all of our doctor's opinions whether or not a second opinion is recommended

I guess you have a good health and don't see doctors often, or you would never say this.

* We trust our math applications to do math properly

Really? I live in a scientific environment and I've never seen a colleague who put his/her full trust in a mathematical program.

* We trust our spell checkers to check properly

You're joking, right?

I myself never trust anything fully, especially if it's capable of doing more than one specific thing. Even if it doesn't have design flaws, it can break or be used in a way it wasn't meant to be used for.

no units ...

[Anonymous Coward]

From basic trigonometric functions that forget to specify units/i?

Trignometric functions are unitless to begin with. They are ratios.

Ok, but...

[hummassa]

TFA is not really impartial. Let's quote some:

First, let's take the trigonometric functions, SIN (Part 4, Section 3.17.7.287), COS (Part 4, Section 3.17.7.50) and TAN (Part 4, Section 3.17.7.313). Hard to mess these up right? Well, what if you fail to state whether their arguments are angle expressed as radians or degrees? Whoops.

Hmmm.
Someone failed the math class where they explained that an angle is a "dimensionless derived unit" [wikipedia.org] . Explaining, short version for the clicky-impaired: angles are the ratio between two measurements of length -- the length of an arc and the radius of said arc.
It got off to a bad start. For the rest of it, it moans about bad revision and wrong formulas, with some reason, but without a lot of substance.
I am pro-ODF, but this article is worthless.

Shame?!

[krygny]

"Shame on all those who praised and continue to praise the OOXML formula specification without actually reading it."

Reminds me of something I once heard a congressman rationalize in reference to a bill he just voted for containing several lame provisions (many with which he did not even agree): "Do you have any idea what reading a bill like that would entail?" I do. It would entail you doing your fucking job.

Re:Ok, but...

[jkrise]

Someone failed the math class where they explained that an angle is a "dimensionless derived unit" . Explaining, short version for the clicky-impaired: angles are the ratio between two measurements of length -- the length of an arc and the radius of said arc.
It got off to a bad start.

Technically that may be correct, but in reality, it is very common and practical to express angles in degrees. So, sin(30) = 0.5 and tan(90) = 1. Memorising the values of sin, cos and tan for 0, 30, 45, 60 and 90 degrees is a de-facto requirement to solve trig. problems in high school. Does Microsoft expect students to relearn all these convenient derievd units in radians, and go mad?

A document standard is a practical necessity to express everyday ideas in a readable format. Not to be technically accurate and practically useless. Try typing HCl + NaOH --> NaCl + H2O in Office, and watch yourself breaking the monitor.

Re:Ok, but...

[itlurksbeneath]

Dimensionless or not, in the real world (i.e. not in math class - and you really have to pick one way in math class, too), you have to pick one system of representing it and use that to send to your functions (see sin() [die.net] as an example).

That Wikipedia page you referred to us using the derived unit of "radians". There are a couple of different ways to represent that number - degrees, radians, grads. Hell, anybody that's ever used a calculator knows you have to use just one of those systems for your particular calculator.

Nice try, but do a little more research before posting and blasting somebody's article with illogical arguments.

Re:So?

[ericrost]

"* We trust all hand tools like wrenches and sockets to be exactly the size on the label"

Well, technically we trust our hand tools to conform to the nominal size specifications that go along with the size on the label and thus interface correctly with any connector that also conforms to that nominal size specification.

A 3/8" wrench is not 3/8" EXACTLY, it is some close approximation of 3/8" toleranced such that a correctly toleranced bolt that is a close approximation of 3/8" is guaranteed to be smaller (in the case of a hex head bolt).

Just your friendly neighborhood mechanical engineer. :)

Re:Yeah, I'm sure this guy is objective

[jkrise]

So a guy working on a different document format, for a company who competes with Microsoft, has unkind words? Color me shocked.

A competitor has a vested interest in exposing short-comings of his competition. So an IBM employee is the best critic of a competing Microsoft product. Why is this hard to understand? Why not criticise the views expressed, rather than the person he is OR his employers?

As for spreadsheet formats not being defined in ODF - it isn't a big deal, and the alliance seem to be working on the issue, in any case. A wrong formula is infinitely worse than No Formula.

I wonder what your vested interest is... your lack of a meaningful response and indulging in mud-slinging appears very deceptive, and your motives - suspect.

Re:Proof that open formats are a good idea?

[bjourne]

Sorry, but this is FUD of the worst kind and it is very unfortunate that it comes from IBM and the free software community. Every standard has omissions, most even glaring faults. You could find similar problems in virtually all specs. I'm to lazy to provide examples, but you can dig up lots of problems with even a venerable and industry proven spec like C89/90 too. So the spec doesn't specify whether trigonometry functions accept radians or degrees? That is what is called a "bug." Most likely, the OOXML spec will be revised to include those details. What makes or breaks the spec is whether it will be updated to fix its problems. In the meantime, you do what every other spec implementor since the time of dawn has done, rely on the reference implementation. Does MS Excel use radians or degrees? There is the answer.

A buggy spec is better than no spec at all, and ODF has no information whatsoever about its formula functions. Harping down on OOXML when ODF completely omits so much information is pretty laughable. If you want to push for ODF then please don't use fallacious arguments.

Re:Guess what?

[kryten_nl]

Just because a certain feature is the de facto standard, doesn't mean it shouldn't be included in a standards document to combat ambiguity.

Btw, comparing Excel (Excel users) to a programming language (programmers) is a stretch at best.

Someone else failed the math class

[giafly]

Someone failed the math class where they explained that an angle is a "dimensionless derived unit" [wikipedia.org].

If you must quote Wikipedia, please read it first. This article refers to the "SI system of measurement units" which measures angles in units of radians: "The unit of angle is the angle subtended at the centre of a circle by an arc of the circumference equal in length to the radius of the circle. There are 2 radians in a circle."

Other measurement systems use different units for angles, for example degrees.

In short, a thing being dimensionless does not mean no units are used to measure it.

Because the OpenFormula isn't in the spec?

[Anonymous Coward]

And the spec is going to include it. Maybe that is why ODF doesn't supply it.

So, rather than get it WRONG, they are leaving it out.

"Slashdot requires you to wait between each successful posting of a comment to allow everyone a fair chance at posting a comment.

It's been 54 minutes since you last successfully posted a comment"

Re: Circular Reference Implementation

[JetScootr]

Absolutely not. A standards specification should stand on its own, or reference other standards. "MS office" is an implementation, not a standard. It can't be used to define a standard any more than the wheelbase of your car can define what a roadway should be.
Further, if ooxml is as "free" as MS would have politicians believe, then referring back to a proprietary product destroys that "freedom". (It's really not free, anyway, but just for the sake of discussion...)

Re:Proof that open formats are a good idea?

[Karellen]

Oh, I definitely agree that open formats are a good idea, and that this does show one very good reason why.

But, the point is that MOOXML is a shitty open format. It was written in a closed environment, without a decent review by anyone, in 1/20th the time you'd expect a spec of that size to take, and is being put on a "fast-track" process to ISO which means - if it goes through - it will never have had a proper review by anyone.

Yes, having the format be open is a good thing.

But this format is utter crap, on many different levels. It's size, complexity, inconsistency, bugginess, NIH-iness, reliance on Win32, etc., etc., etc.... make it completely unsuitable to be ratified as an ISO standard.

When you're turning something into an international standard, you want to take your time and get it right. That's what the standardisation process should be about. Creating something usable by as many parties as possible. MOOXML fails completely here.

Yes, I'm in favour of them opening their document formats. I wish they'd release updated documentation for the binary .doc format as well, usable by anyone (last I checked there was a "you must agree not to use this information to create products that compete with office" clause in the (outdated) documentation download) so that people could interoperate with those formats on non-Windows platforms. But do I wish for the binary .doc format to be an international standard? Hell no!

Re:EULA?

[trianglman]

Actually, in the US, many provisions in most EULAs are illegal and not considered valid. I don't know specifically about the "no liability" clauses, but I do know that in somewhat related cases, warnings do not remove a company's liability (i.e. wet floor signs aren't protection against a company being sued when a customer falls on said wet floor). I personally don't see why holding a company liable for damage caused by their software is a bad thing. It might get us software that will actually work...

Re:Yeah, I'm sure this guy is objective

[WalterGR]

I wonder what your vested interest is... your lack of a meaningful response and indulging in mud-slinging appears very deceptive, and your motives - suspect.

I'm a Microsoft shill. They pay me a ton of money to post on /. because they know that maybe - just maybe - all it will take is one more post in their favor to convert the unfaithful.

Seriously? Come on dude. This entire story is more-of-the-same get-the-crowd-riled-up /. click-fodder. But if you want to make a show of it...

So an IBM employee is the best critic of a competing Microsoft product. Why is this hard to understand?

My post pointed out that the guy isn't objective. Then it proceeded to give an example of using the pejorative "parroting the party line" for merely stating a fact. That's it.

As for spreadsheet formats not being defined in ODF - it isn't a big deal, and the alliance seem to be working on the issue, in any case.

A lack of spreadsheet formats is a big deal if, for example, you want to create a spreadsheet.

I prefer to comment on stories and not my rhetorical technique, so I won't be watching for responses to this post. If you'd like to discuss it further, feel free to e-mail me at [e-mail].

Re:now arriving at Dallas-Fort Worth...

[RexRhino]

"The Bush administration largely gets it backwards, they say health care is expensive because of lawsuits. I say lawsuits are expensive because of our health care system." William M. Sage, Columbia University law professor and physician.

Mr. Sage might be a law professor and physician, but he is obviosly not an economist. You have to be pretty foolish to believe that making doctors pay millions in insurance premiums, and driving thousands out of the medical field for fear of lawsuits, wouldn't make health care more expensive.

Name dropping Bush is a nice touch too... usually when people can't make a coherent arguement, they throw the word "Bush" into their sound bite, because Bush is extremly unpopular and that is enough to make people lose their ability for critical thought. I mean, why didn't he mention Hitler while he was at it? Or Satan?

Did anyone notice that insurance companies donate literally billions of dollars to campaigns over the past decades to get this passed?

The Trial Lawyers of America, the political action group that represents ambulance chasers, gives 50% more campaign donations that the rest of the insurance, medical, and pharmicutical industry combined. They are the single largest donator to political campaigns in America. However many "billions" you think that the insurance companies spend, multiply that by several times and that is what the Trial Lawyers spend! http://www.triallawyersinc.com/healthcare/hc07.htm l [triallawyersinc.com]

Re:now arriving at Dallas-Fort Worth...

[node 3]

Mr. Sage might be a law professor and physician, but he is obviosly not an economist. You have to be pretty foolish to believe that making doctors pay millions in insurance premiums, and driving thousands out of the medical field for fear of lawsuits, wouldn't make health care more expensive.

Yeah, when you require such an important job be done right, *of course* it's going to cost more. At least, up front. In the long term, however, it's much, much better.

Capping medical malpractice awards is another way of saying, "limiting the rights of the victim". Aren't you conservatives supposed to be all in favor of victim's rights? If your doctor screws up, don't you think you have rights to compensation under the law? Capping compensation is like capping prison sentences. It's like saying, no matter how bad your actions are, you can only be punished so far. For a group that so strongly supports the death penalty, being against having to pay for the damage you've caused seems absurd to the extreme.

How would you feel if everything was capped like this? If your building contractor was similarly capped? Did his malpractice cost you $100 million? Tough, you can only get $200k from him, regardless of whether he was entirely at fault, and found negligent at his profession. That would be insane. How much worse when it's something to do with your health!

The Trial Lawyers of America

There is absolutely *NOTHING* wrong with being a trial lawyer. To be against them is to basically be against the JUDICIAL system. How insane is that?

You used the derogatory term "Ambulance Chasers" to refer to trial lawyers. If they have a case, if there was fault worthy of a trial, what is *wrong* with seeking to make the guilty party pay? If it weren't for trial lawyers, the US would be a much more dangerous place to live.

I have absolutely no doubt that if someone's actions caused you significant loss of health, *YOU'D* hire the best trial lawyer your money could buy.