Microsoft's Office 2007 Word documents do not conform to the newly-approved Open XML (OXML) international standard. Alex Brown, who heads up the group responsible for maintaining the OXML standard at the International Standardisation Organisation (ISO), revealed the less-than-pretty findings in a blog post late last week. He …
A fast track is for something that is already basically a standard and all that is needed is to say it's an ISO standard.
So MS saying that being mostly compliant is fine means that even MS agree it should not have been a standard that went via Fast Track. Mostly because they don't have any conforming applications, it isn't used ANYWHERE and MS wouldn't expect it to be used anywhere.
Compare with the standard process (Publicly Accessible Standard?) that ODF had to go through: the spec was changed BEFORE it became the standard and only when it was set was it implemented. And setting the ODF standard took years.
There's absolutely no reason for MSOXML to have gone fast track and plenty of reason for it not to.
Want to scare Microsoft?
Start emailing all word processed documents in ODF.
Mention in the email body that you wrote the document in OpenOffice (or similar processing package which supports ODF) and they can download it from xyz.com for free, for personal or corporate use, and they can roll it out to every computer, networked or not, share it with anyone they want, totally legally. They can burn the installer to CD and take it home, again totally legally and totally free.
Mention that instead of having Microsoft Office 2007 at approx. £300 a license, they can have a FREE product which does the same thing. (At this point, mention it's totally capable of saving files in Microsoft compatible formats, and reading Microsoft format documents. Compatibility with market leaders is paramount for any company to consider.)
Viola. ODF is the new standard.
NOW GET TO WORK. We have a revolution to continue.
(Tux, because FOSS.)
Just before ppl come slamming OOXML and Office 2007
.... and I would be first to, I want to remind that this particular finding is not surprising at all, because IT standards are notoriously difficult to implement accurately. For example, the only vendor that implemented 100% of other widely acclaimed standard ISO/IEC 14882:2003 (C++ programming language) is EDG. And even there 100% is not guaranteed, IIRC.
Aren't MS the losers here?
Presumably there are legal means to stop someone claiming adherence to an ISO standard if they actually don't. So, er, don't MS end up being unable to claim support for any standard document format?
"Under the test, the spec generated 122,000 errors against the strict schemas and 84 errors against the transitional schemas.
Microsoft's Office interoperability senior product manager Doug Mahugh felt those results were good enough for the software giant to hold its head up high."
Sounds about right, eh? Only 122,000 errors? That's probably the best MS has ever done! Eighty four, then, is unprecedented even in fiction...
This is nothing new at all.
This (now rather famous) scathing attack on OOXML and how Excel fairs with it highlights just how bad MS are are meeting their own standard. It makes for interesting reading; the whole thing is basically completely buggered :
Proof, if it were needed,...
...that we don't need two standards that do the exact same thing.
And we certainly don't need a "stanard" (OXML) which is internal inconsistent! That's not a standard, that's a dog's dinner that is!
- reading comprehension that is. Go back and keep reading Alex Browns post until you understand it.
got me coat, time to take the whippet down pub
It's been said a million times before, but the hell, it's worth it.
If I e-mail all my documents to people in ODF they won't open them. Professionally speaking, that's a bit of an issue. If Office 2007 read ODF then we'd be fine, but then, on the other hand, if it did that then no-one would notice I was using anything different.
I'd rather sell my products than someone else's and thats how I stay in business, if I want to spend a few days learning a new word processor then I'll take a look but if I'm working then I'd rather make money.
Besides Office 07 was out before this standard was and since schema change in order to become standards it didn't stand a chance maybe the next version (which will be backwards compaible) or maybe a patch.
TextPad supports RTF, which I believe is a standard.
Oh yes I would so love to be able to do that. Unfortunately, I feel it would be severely career-limiting if I sent an email like that to my biggest client who have just swallowed the whole Redmond bloat package (Exchange, OCS, MOSS, etc), regardless of my position in my company.
AC for obvious reasons
How to *really* scare Microsoft ...
Implement an office suite that uses OXML correctly. Wouldn't it be hysterically funny if the first OXML-compliant application was FOSS? :) To be honest I doubt it's worth the hours (years?) that would go into it but the idea of beating MS to an implementation of their own pseudostandard tickles me.
"IT standards are notoriously difficult to implement accurately"
That may be, but when the standard that Microsoft want implemented
a) doesn't work with their most recent product
b) is so vague ("it'll do it like Office 2000")
c) is not patent-proof
then if you don't mind, I'll be fucking skeptical.
Für immer und ewig ist dieser Standard ganz mein!
Bronek Kozicki: "I want to remind that this particular finding is not surprising at all, because IT standards are notoriously difficult to implement accurately. For example, the only vendor that implemented 100% of other widely acclaimed standard ISO/IEC 14882:2003 (C++ programming language) is EDG."
Maybe - but I strongly suspect it wasn't that EDG submitted the standard for consideration in the first place, it wasn't that EDG was trying to fishtail an earlier, already approved C++ standard, it wasn't that the standard approval process did not involve more raised eyebrows than a US presidential election and finally it wasn't that the only one that knew how to implement the standard (and the only one to be able to do so legally, actually) was EDG.
Thus, the cases are not comparable.
Paris, because she might have unsuspected standards.
Right, a couple of points here:
1. "...does not adhere to the latest specifications of the draft standard (ISO/IEC 29500)." That'll be the word DRAFT then....
2. Um, who does support standards? The ISO's website doesn't even adhere: http://validator.w3.org/check?uri=http%3A%2F%2Fwww.iso.org&charset=%28detect+automatically%29&doctype=Inline&group=0
@Ash: More like how to scare your clients
I honestly have no idea, why people keep sending word processor files to me. Usually, all I have to do with them, is to read them and provide some kind of response.
Now, I don't care what looser word processor you use to make the text as long as it does the job, and I get at easily to read document. And bloody hell no, am I going install anything in order to do so. If I have, you *will* get "deprioritised".
So, there are a couple of standard format, which I am happy to accept: PDF, PS and ASCII text.
So take your OXML, ODF and even XSL-FO and stick them! None of them are mature enough to become real standards yet.
I think you might need to clarify your use of the word 'standard' in respect to RTF. Microsoft call it a standard, yet its specification seems to change in lockstep with Office releases.
Certainly saving an Office document as RTF is no guarantee that TextPad can display it as Office will.
Get a free set of stake knives with every ISO standard purchased.
"There's absolutely no reason for MSOXML to have gone fast track"
Sure there are, lots of them... They all start with $
Isn't ASCII (the American Standard Code for Information Interchange) a ISO standard? And I believe M$ supports that... although I suspect that's only by necessity, and if M$ had their way MSCII would be the new standard.
@Alistair and others
I dunno. Consider this: there are a number of offices out there still running Microsoft Office 2003 while new computers are coming with Microsoft Office 2007. For the 2003 people to use 2007 documents, either 2007 people have to Save As a 2003 format (unlikely,) or the 2003 people have to install the Office Compatibility Pack, which is a Microsoft download.
Now, let us consider that we sent an ODF document to a customer or colleague using Microsoft Office 2003 (for the sake of argument, we will ignore 2007 for right now.) The ODF plug-in for Microsoft Office is also a Microsoft download.
Given that both scenarios would require a download, I could easily imagine the receiving end being ok with doing so, especially since both downloads are available from Microsoft, so they should be trusted by the IT department or end user, right?
Paris, trusted by IT departments the world over.
They're scared already
@ Ash, re: want to scare Microsoft?
I have previously done exactly what you suggested about a year ago. I only use ODF, and build complex documents containing macros which I cannot be bothered to translate into M$'s proprietary protocols. I sent out some ODF forms to about 300 people at work, and included a note that if they wanted to try this out they could download OpenOffice from www.openoffice.org. Guess what? Hundreds of people wound up switching over to OpenOffice. So your suggested technique works well, to their benefit in obtaining free supported software (instead of using unsupported possibly copied MSoffice, with the associated security risks and vulnerabilities).
Just one correction: Microsoft is ALREADY scared, hence their disastrous rush to ram the ridiculous OOXML though ISO. Shot themselves in the foot again. Stupid, that.
@Thomas Vestergaard: instead of being so indignant upon receiving an ISO standard, editable format that actually works and is in common use throughout the rest of the world, here are some options you could consider:
1) just use GoogleDocs to open it, and edit it and read it on-line, without installing anything. Google supports and uses ODF. Google is a pretty big company, isn't it? And IBM? And Sun? Canonical? Novell? And so on?
2) Download the ODF converter for MSoffice from http://sourceforge.net/projects/odf-converter/ Boy, I wonder why MS didn't write that little ap? And why they tried unsuccessfully to reinvent the wheel with OOXML?
Of course, OpenOffice or IBM Symphony, or Abiword are all free, support ODF as well as MS formats, and work great, but M$ will always be ready to take hundreds of dollars from those who want to support their worthy cause (to make you poor, and them rich).
A agree with you on that one... though I'll add HTML in that pile as well...
So when is Office 2007 going to morph to the more compliant Office 2012 , or will that be included in Office 2007 service pack update due in or around the same time to comply with EU regulations drafted in 2009 ?
As most would say , seeing is believing !
Just to make this a three-way fanboy fight:
I just use LaTeX, which can generate BEAUTIFULLY typeset pdf (even use it for posters and slides).
But then I am an inveterate nerd, with WAY to many equations for MS-Word to handle. I found that 128 equation objects was a limit in earlier MS-Word versions(6.0 and '97). Maybe that has improved (perhaps even up to to 256 ;-) ).
Scientist icon because LaTeX will never become mainstream.
@ Thomas Vestergaard
"I don't care what looser word processor you use to make the text as long as it does the job, and I get at easily to read document. And bloody hell no, am I going install anything in order to do so. If I have, you *will* get "deprioritised"."
So you do care what word processor they use? or is it that you do care as long as it is Microsofts shitty crap?. Now, the big problem for me is not whether we have to waste our time installing things or not, but that someone sends me a document format I cannot open and I then have to pay to open that format. (I know Microsoft do supply readers for their office formats which I suppose is good but what if I have to do a minor edit on them?)
On another note, maybe you use OpenOffice as I hear they are pretty loose. They give their product to anyone! :o)
Paris. Because she looks better than OXML & Thomas Vestergaard, and can conform to my standard any day!
As a contractor I use Open Office that has a very handy "Export to PDF" file menu item, that works a treat. This provides professional looking documents that can't be edited. This is great as I am rarely collaborating on documents with people.
If I need to collaborate then I can turn on track changes, save as a .doc and email it. This function is cross compatible with MS Word and works very well.
It has to be said that Open Office is not without its bugs. Particularly with dates from Excel to Open Office spreadsheets. When copied or referenced the dates are years apart.
However, instead of shaking my fist and saying a "prayer" to Bill, I shake my fist and accept that I got it free and it works pretty darn well. (Though I really wish they'd fix it)
I'm not sure the bug list is any shorter in Open Office than Ms Office. But at least it's free, in the sense of "beer" and "speech".
(For clarity's sake I'm using Neo Office on OS X)
I would move to Open Office in a heartbeat, but...
...for one member of the Open Office suite - Impress (the Powerpoint equivalent).
I use word processing and spreadsheet functions to a very limited level (as do most end users), but with Powerpoint I push it to the absolute limits.
When Microsoft release Powerpoint 2007 they totally screwed up by introducing the "ribbon" interface. Regardless of how good or bad this interface is, it turned their existing customers from "experts" to "novices" overnight.
I felt this was a fabulous opportunity for Open Office to establish itself as a serious alternative to Office. Open Office apps that retain the familiar look and feel of pre-2007 Office, and offer the same functions and performance should be snapped up by the market - especially because it's free!
HOWEVER...Impress is essentially unusable for anything but basic text slides. This is apparently (according the the Impress blog) because it does not take advantage of graphics hardware acceleration on PCs.
The result is that slide transitions and animations are jerky. Text and objects are initially displayed without anti-aliasing, and then visibly anti-aliased on screen for the audience's delight and amusement. I assume that the people who write Impress somehow don't notice, but to presentation professionals it's impossible to ignore.
I think the problem is that the people developing Impress tend to be developers (gosh, there's a surprise). But the people who use Impress are presenters.
But those developers also present slides from time to time. You've seen those presentations - they're the ones with a couple of hundred words on a slide, and lots of cartoon clipart images - good thing I can reduce that font size down to 8-point eh!
But Powerpoint has evolved not just as a text slide presentation package. It now contains a rich set of animations such that you can create quite complex, interactive "modules". As I say, I don't think I'm typical of the average Powerpoint user.
When I raised this issue in the Impress blog I was told "you should be using Flash if you want fancy animations". While I'm sure the person who wrote that is very clever at programming - he (probably a he) has completely missed the point.
There's a very real opportunity, given the gross incompetence and arrogance that Microsoft has displayed with Vista, OOXML, and Office 2007 for Open Office to establish a strong base of users. But to compete with Office you have to track both the features, and the performance for the whole suite, while returning the useability we enjoyed pre-2007.
I happen to be a Powerpoint "power user" (or I was before 2007 turned me into a bumbling eejit). I'm sure there are Word power users, and Excel power users who have similar concerns about the other Open office apps.
What's needed is a ODF reader
99% of the time a document emailed from outside the users organisation only requires reading or, at the most, printing. A subset of open office to just read an print ODFs could be made small, therefore quick and easy to install. It wouldn't upset people who have paid a fortune for a word processing package and don't want to believe that a similar cheaper option existed.
Like Acrobat reader it would just be seen by casual users as a 'document browser'. It is required that web businesses offer certain legal documents etc in a format that is available to all users. ODF documents and a link to reader offer the cheapest and only 'future proof' way of doing this (apart from HTML, which isn't ideal for the purpose). Once a critical mass has been reached where a significant proportion of users are prepared to open ODF docs then users will start to use the format in ernest.
So come on somebody, whip together a small and easy to install ODF reader!
No icon chosen as they do not adhere to an ISO standard.
How really really to scare microsoft
Get them at the bike sheds after school and finish off with opening
a sauce bottle over their heads ;)
£417.48 for office 2007 pro (Amazon) ?
On yer bike I can buy a bloody computer for that and have it packed with FOSS !
M$ - a monopolist robbing you blind !
@ Thomas Vestergaard
"Now, I don't care what looser word processor you use"
Preferably one with a working spell-checker ? ;P lol
Sadly a spell-checker wouldn't have helped as although looser is the wrong word, it's actually spelt correctly although I much prefer tighter word processors.
Impossible to tell w/o looking at the validator output file whether its error-recovery is poor and/or it actually does what it's supposed to, i.e. compared to 6000 pages of spec. I hope nobody has just stuck a 16-bit int counter in a naff spot in the FSM (and forgotten to initialze it). But frankly it is beyond a joke.
The basic idea of XML when it was SGML was to encode classes of documents with a flexible but formalizable structure - not to formalize every possible permutation of elements and entities that a word-processor can output, which achieves what?
M$ buy themselves respectability. The cost would have been prohibitive for any business without its almost limitless cash or obsession with hegemony. They did it only after government bodies started making noises about ummm well, it's getting to be a rather big recurring cost, and ummm perhaps it is better not to be tied to one supplier, and ummm...
Which goes to show - as Tuco said "If your're going to shoot, shoot...". And every ideal turns into an idol.
OK so before I start let me clear something up. I'm against the way OOXML was forced through the ISO process, I'm horrified by the stunts MS pulled to get it through. I use OO.o as my office suite and I run Ubuntu as my primary OS.
Now that's out of the way, can you squawkers just tell me how exactly you expect a piece of software published before a standard is created to conform to that standard? The whole point of the BRM was to fix as many issues found in the ECMA standard (which Office 2007 does conform to) as possible. Now I'll agree that not enough was fixed, but that very process of fixing will by definition mean that Office, released before the changes, *can not* conform until it's patched. It's a logical impossibility.
Engage brains before opening mouths please.
@Just to make this a three-way fanboy fight:
I agree, LaTeX is wonderful :).
The only problem is that the learning curve is a bit steep, so people aren't too happy in the beginning... A it hard to introduce to new persons, if they aren't writing something *highly* technical.
*Also a geek*
If you send me documents in OXML, I won't read them.
In fact, probably 80+% of people CAN'T read them.
So your point was what?
Re: Ugh, embarrassing
You forget that this standard was passed (unlike ODF) through the Fast Track process.
That process is intended for things which are ALREADY a well recognised defacto/pro-forma standard.
So when Ethernet Frames were given an ISO (if they were) it was already an IETF standard implemented in hundreds of products with a damn good BSD (and therefore COMPLETELY OPEN: this was before software patents) implementation. It really could afford to go through on the nod.
NONE of the other standards that aren't 100% completely implemented ANYWHERE went via fast track.
And if MS thought this was a legitimate Fast Track standard, IT MUST HAVE ALREADY BEEN IMPLEMENTED!!!! At least on Office. And the "transitional" can only mean "well, it was the standard before people saw the bugs we couldn't see" and MS Office doesn't even do THAT right.
That is what you missed.
A Fast Track MUST ALREADY BE IMPLEMENTED to become eligible for that process.
This, most obviously, wasn't.
Please engage at least brainstem before posting.
| Right, a couple of points here:
| 1. "...does not adhere to the latest specifications of the draft standard (ISO/IEC 29500)." That'll be the word DRAFT then....
| 2. Um, who does support standards? The ISO's website doesn't even adhere:
| (13 failures)
Ok, first off the W3C html recommendation is not an ISO standard. In fact that's the meaning of the dash (-) in the first quote of the DOCTYPE declaration:
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
It tells us that the organization that follows (W3C) is not a registered ISO organization.
Now if you test www.w3.org against the recommendation, I bet it will pass (in fact I know it will)
Did I say this would happen when they shoved the standard through the ISO approval? It figures that MS can't even make it's own product conform to it's own freakin' standard,
LaTeX is wonderful, but only if you can accept treating your letter as CODE which needs to be COMPILED, and more often than not DEBUGGED.
Of course, I use it... And it looks beautiful... But damn, user-friendly it is not. Basically, if you want to do anything special, you better pray that somebody else tried before and posted the solution on the web...
TeX is also a very interesting programming language. In the APL sense of "interesting".
No Shit Sherlock
What! Microsoft doesn't conform to a standard ( even one they bribed and corrupted and entire process to deliver )... elementary my dear Watson, a convicted predatory monopolist cannot change its spots...
"Wouldn't it be hysterically funny if the first OXML-compliant application was FOSS?"
Since both standards are XML-based, it ought to be possible to write a piece of XSLT to transform one into the other (ho ho!) and I think I've read somewhere that such an effort is underway. You may yet enjoy that particular piece of hysteria.
only requires that you treat the document as code if you look at the source. Same deal with OOXML, HTML etc. A LatTEX-eating WYSIWYG word processor would hide that all from you.
Personally I find OO does everything I need. I generally send out PDFs (no need for clients to modify the docs). If the client does need to modify the docs, and they are WinTards, then I'll send out one of the retro .doc formats that OO supports or html.
Charles, there's something called LyX which is a near-wysiwyg interface for LaTeX. But why bother anyway? (though it might be usefull for beginners who want to use LyX output as a starting point).
To add a bit of spice to the discussion, I do prefer Lout: smaller (full install: less than 9 Mb), more simple, but not less powerfull.
Almost forgot about the topic: I am not surprised at all about the 122 000 errors against the standard. That's probably the closest they ever got to a standard (the fact that they pretty much wrote it might explain...). 84 errors against their very own description of the actual document? That would be hillarious, if not for the fanbuoy crowd who keeps shouting "MS rulez".
When PDF first came out...
were there lots of people saying to the early PDF adopters, "none of this PDF for me thank you, send me plain text or send me my Microsoft formats,or be gone with you"
What happened to make PDF ubiquitous?
What stops the same happening with this? (What would it take to make the same happen with this?)
@ Steven Shuster (w3)
Now let's have fun and compare how the websites for two major office suites and a word processor compare (w3c validator)
www.microsoft.com : 26 errors, This page is not Valid HTML 4.01 Transitional!
MS office page: 55 errors, This page is not Valid XHTML 1.0 Strict!
www.openoffice.org: This Page Is Valid XHTML 1.0 Transitional!
www.abiword.com: This Page Is Valid XHTML 1.0 Transitional! (1 warning though: The character encoding specified in the HTTP header (utf-8) is different from the value in the XML declaration (iso-8859-1). I will use the value from the HTTP header (utf-8).)
(just for the fun: www.debian.org: This Page Is Valid HTML 4.01 Strict! whereas www.ubuntu.com: 54 errors This page is not Valid XHTML 1.0 Strict! - sounds like the Ubuntu guys designed their site with FrontPage!)
Many fail to see Microsoft is reasonnable here
For once, and only on this very, very precise situation, actually Microshit is right and many people bash without even thinking, because there's "Micro$haft"named here, so they're in automatic mode.
Re-read please, many lose all credibility and do real, thought-upon Microsoft bashers harm by criticizing anything even when MS is right.
The article, for those who obviously haven't understand, says the following:
- Microsoft had Office 2007.
- They tried to make a standard out of it
- They failed
- What they did succeed is having their spec be modified in small ways and get THAT adopted as a standard.
What do you expect? What happened is that to accept that as a standard (which I fully agree they should never have done but that's something else), ISO required some changes.
What is now standard is NOT what Microsoft proposed but something "quite close".
Hence, it is obvious that it's impossible that they conform to the standard, it's a matter of definition of the word "changes". And it's not related to Microsoft at all, for once.
Please keep the bashing for when there is reasno for it. It's often enough that we don't need it when for the first time in maybe a year, Microsoft can't be reproached anything here (oh, they can about the previous process itself, just not about being non-complaint right after the spec was changed).
In fact, it would be if they WERE compliant that it would be scary. It would mean the ISO body was so corrupt they did not even ask for any change to what Microsoft had proposed (or even worse, that they asked for changes which Microsoft had suggested to them so they could keep face).
A bit unfair to expect
"the tweaks made to the spec at an ISO ballot resolution meeting (BRM) in February this year."
It is a bit unfair to expect code released prior to this date to be compatible with these "tweaks".
I am not MS's greatest fan but this is nonsense.
@Greg, @ Zap, et al
That is the appropriate response for a standard NOT taken through the Fast Track process.
IF MS had not forced this through the Fast Track or, on having found that their Office format NEEDS changing, they could have withdrawn the standard from the Fast Track process and taken the PAS route (as ODF did).
If MS had done that, then calling them to shit WOULD be unwarranted.
MS DID NOT DO THIS.
Ergo, they deserve the shitpile being dumped on them.
Has everyone forgotten - MS promises not to sue only if your implementation conforms to OXML.