Microsoft Blasts IBM Over XML Standards 323
carlmenezes writes "Ars Technica has up an article discussing Microsoft's latest salvo against IBM. Microsoft's open letter to IBM adds fresh ammunition to the battle of words between those who support Microsoft's Open XML and OpenOffice.org's OpenDocument file formats. Microsoft has strong words for IBM, which it accuses of deliberately trying to sabotage Microsoft's attempt to get Open XML certified as a standard by the ECMA. In the letter, general managers Tom Robertson and Jean Paol write: 'When ODF was under consideration, Microsoft made no effort to slow down the process because we recognized customers' interest in the standardization of document formats.' In contrast, the authors charge that IBM 'led a global campaign' urging that governments and other organizations demand that International Standards Organization (ISO) reject Open XML outright."
Microsoft is being disingenuous (Score:5, Informative)
This might be true, but when Massachusetts decided to adopt this standard they raised holy hell, and used every trick in the book to make Massachusetts take it back.
10 Billion In Revenue - You'd Be Pissed Too (Score:2, Informative)
Even a modest loss of that revenue would bring dramatic changes to Microsoft as a company and how it operates.
yea sure (Score:5, Informative)
"When ODF was under consideration, Microsoft made no effort to slow down the process because we recognized customers' interest in the standardization of document formats."
Yep you bet no effort to slow down the standardization process because they refused to be involved. However they have made every effort possible and will continue to do so in the future to slow
the adoption and deployment of this standard by any means necessary.
Ars missing something (Score:5, Informative)
OASIS submitted ODF (Score:5, Informative)
This emphasis on ODF is to strengthen the parent post's claim on the importance of ODF being unencumbered.
The only standard that stands still is dead (Score:3, Informative)
Okay - the subject is probably overkill. Standards change all the time. Or rather, standards gain extensions and new features all the time. I work with DRDA (a database networking protocol to encapsulate data passed from client to server and back) and it is constantly being added to to cope with new situations and requirements. That's not to say it's a bad standard - the core is solid and does (mostly) what database people need it to do. When we need it to do something new, we make proposals. The DRDA review board takes a look at it. Other people who use DRDA get an opportunity to make changes or block it entirely. We make changes to the proposal and it goes around again. Eventually, once consensus is reached, it gets formally written up and becomes a part of the next iteration of the DRDA standard.
When a standard stops evolving, it is because people no longer need it to do something new. That can be for entirely good reasons (it does everything one could conceivable need) but it does mean that that standard has reached it's natural limits.
ODF continues to evolve because people keep needing documents to do new stuff. Collaboration, equations, macros, formulas are all areas of change. A good standard recognizes that change will happen and builds that change right into the core structure. ODF has an extensions mechanism for precisely this reason. You will still be able to open an ODF version 1.2 document with an editor that only supports ODF version 1.0. Any features that are not supported by the ODF 1.0 editor won't be usable, visible or editable but that won't stop you getting at the rest of the data.
Cheers,
Toby Haynes
you're both wrong (Score:1, Informative)
Code that's been used more than once is reused (in the second instance). Code could be
reusable and never have been reused. Reusable indicates the ability to be reused.
Similary, portable indicates only the possibility of movement across platforms, not any
actual history thereof.
And, yes, IAAL (I am a linguist).
Re:They both suck. (Score:5, Informative)
Re:Ars missing something (Score:5, Informative)
IBM or MS who to trust (Score:1, Informative)
If for some reason you think the big guys opensource stuff, just to feel better trust me they don't, they do it because they get more money out of it, so they are willing to sink money into it.
Re:Whose format is whose? (Score:5, Informative)
- OpenOffice
- Star Office
- Google Docs & Spreadsheets
- KOffice
- Scribus
- Abiword
- ajaxWrite
- Zoho Writer
- Ichitaro
- IBM's Lotus/Domino
- IBM Workplace
- Mobile Office
- Gnumeric
- Neo Office
- Hancom Office
- WordPerfect???
So it is just Microsoft who is trying to frame this as a MS Office vs. OpenOffice argument, when it really is an Open, multi-vendor format vs a single vendor, obfuscated format argument. Argue formats, not software.Re:They both suck. (Score:3, Informative)
Re:One True Format (Score:3, Informative)
Call me crazy but having two different standards doesn't really capture the idea of having Standards at all. I thought the point of standards was to make it so we (the developers) only have to implement one thing.
I disagree. I don't think there is any problem with having multiple OPEN standards because it is easy to translate between them and it allows competition among them for the best feature set and easiest to use, etc. The fundamental objection is what MS has come up with that they claim is an open standard. We've seen this same crap from them many times. Customers demand feature/behavior X because they want certain benefits it brings to them that do not benefit the software developer. Eventually, to keep customers from moving to something else to get those benefits, MS releases their own version of feature/behavior Y that has some of the same characteristics of feature/behavior X and which someone who does not understand the benefits or how those benefits come about might mistake for the same thing. At the same time this feature/behavior Y undermines and removes as many of the benefits that are not in MS's best interests as possible, then spends millions on marketing to try to tell people they are the same thing or that feature/behavior Y is better. Half the people that were demanding feature/benefit X have enough people in their organization confused that the move to feature/benfit X does not take place. MS does not want to give customers features, they want to give them bullet points.
In this particular example, why do people want an open standard document format? Why are customers and governments demanding it in the first place? What are the benefits? Well, first it means you can't be locked in by a vendor and multiple companies can all easily create tools, with different specialties that can interoperate and compete. This means lower prices and more innovation. Since it is a standard, you no longer have to worry that different versions of the software will be unable to read one another's documents.
Okay, lets look at MS's "Open"XML. The standard is very large, which makes it harder to implement exhaustively, meaning different version of software may well have incompatibilities. More importantly, the spec does not define all the behavior of all the things contained within it, instead referencing outside, closed software behaviors that have to be reverse engineered and which can never be done perfectly. When it says, make this table behave like Word98, no one but MS know what that means, meaning no one else can completely adhere to the spec so things are likely to break when moving between different software. That means it costs extra money when moving to a different tool, in order to fix incompatibilities. Many of the features are coded to be tool specific for interoperability with one specific program instead of generically with programs of that type, thus making it hard for users of the spec to interoperate with anyone but MS's partners. Finally, the last I heard MS's license still only specified vendors are protected from patent lawsuits when they are providing the current, latest version of the OpenXML spec, thus creating tools that are backwards compatible with old files and programs is dependent upon MS's behaving well, which no one in their right mind should expect. Does that pretty much castrate all the reasons people want to move to an open standard in the first place?
This is just like their "shared source" initiative. Customers demanded open source because that development method provided significant benefits. One benefit was many people reviewed the code for security holes One was all the "free" features that were added to the software by other companies and hobbyists. One was the fact that the code could fork, preventing lock in by one vendor. What did MS produce in order to confuse customers? Shared Source. Only people who pay and sign NDAs can see it, removing the benefit of many eyes. Only MS can contribute removing the benefit of free code. Only MS cont
Re:They both suck. (Score:5, Informative)
Wow, insightful indeed. To add some metrics to the rebuttals in this thread: create two files with some bogus XML, one with very short tag names, the other with long ones. Then compare their uncompressed versus zipped sizes:
File 1:
<r>
<c>Value</c>
<c>Value</c>
</r>
(1000 total copies of the <c> element)
Uncompressed: 16,009 bytes
Compressed: 190 bytes
File 2:
<thisIsTheVeryLongRootElementTagName>
<andThisIsTheVeryLongChildElementTagName>Value</a
<andThisIsTheVeryLongChildElementTagName>Value</a
</thisIsTheVeryLongRootElementTagName>
(1000 total copies of the <andThisIsTheVeryLongChildElementTagName> element)
Uncompressed: 92,079 bytes
Compressed: 525 bytes
So yeah, let's create really obscure and non-intuitive file formats to save ourselves the wasteful redundancy of XML.
Re:Ars missing something (Score:2, Informative)
Microsoft Standards (Score:3, Informative)
First: They have no idea how to document file formats, this is mostly because of their file format model. I worked as a contractor, indirectly, for Microsoft a long time ago. Their file formats are not "documented" per se'. They are program structure based and can change on the whim of a developer, their name at the time was "chunk format." This works well if you don't expect anyone to use your document format or you supply the access library.
At its core it is because they do not design formats, they code them as needed. Need a feature or special case? Just add a struct, an ID, and a chunk of read/write code and it works. How the hell do you document the outcome of that process? This isn't a bad methodology for internal state or temporary files, but it is a disaster for any sort of long term accessibility and interoperability.
Microsoft develops software like a small company because as long as they have the monopoly, they don't *need* to supply document format information in order to compete. Everyone else has to understand their formats and they aren't going to help at all. Their 'XML' format shows they have not changed one bit. Rather than "design" the document format, they are merely documenting what they have which is just a bunch of special cases.
Second: A true open office document format, usable by everyone, will spawn amazing amounts of innovations. Everything from document searching to intelligent document processing. When anyone can read and create documents on any platform or programming language than everyone else's programs can use as well, just think of what people will come up with. If that's going to happen, Microsoft has to make sure that they are the only benefactor, because except for the monopoly, Microsoft has no inherent value in the face of Linux and OpenOffice.org. At least Apple makes a nice computer.
Here's why (Score:1, Informative)
Authored by: schestowitz on Friday, February 16 2007 @ 09:42 AM EST
,----[ Quote ]
| From: Bill Gates
| Sent: Saturday, December 05, 1989 9:44 AM
| To: Bob Muglia (Exchange); Jon DeVaan; Steven Sinofsky
| Cc: Paul Mariz
| Subject: Office rendering
|
| One thing we have got to change is our strategy -- allowing Office
| documents to be rendered very well by OTHER PEOPLES BROWSERS is one of the
| most destructive things we could do to the company.
|
| We have to stop putting any effort into this and make sure that Office
| documents very well depends on PROPRIETARY IE capabilities.
|
| Anything else is suicide for our platform. This is a case where Office has
| to to destroy Windows.
`----
http://www.iowaconsumercase.or
Share the knowledge, folks.
Re:They both suck. (Score:1, Informative)
Puh-leaze (Score:3, Informative)
Microsoft blasting IBM over standards is another paranoid delusion of MS. IBM and 20 countries did not object to the its OOXML standard because MS proposed it. They objected because the standard is fundamentally flawed. The arstechnica article doesn't go into depth about the objections but Groklaw had a better analysis. [groklaw.net]
My personal opinion is that MS did a poor job of the standard on purpose. They propose their standard so that technically they are working towards interoperability if anybody asks. However, they do it so badly that it could never be adopted. Then they can point to that reason as why they chose not to open up their format.
And why not? (Score:3, Informative)
Now on the other hand, ODF is already standardised, having a new incompatible standard will simply fragment the industry, which is precisely what standards seek to prevent. What microsoft should do, and what ISO should tell them to do, is either use the existing standard, or go through the proper channels to propose updates to it.
Any deficiencies microsoft believe ODF to have, are entirely their own fault... microsoft have long been a member of OASIS, and were more than welcome to contribute to the original drafting of ODF, they made the decision not to in the hope that it would never get anywhere and be forgotten about.
Couple of WTFs... (Score:3, Informative)
Sure, Open XML was designed to address the need for Microsoft to maintain control over desktop office suites, while ODF was actually designed to be an open standard.
No, really, WTF is this supposed to mean? Would Microsoft mind pointing out some part of ODF that's insufficient? Better yet, offer a suggestion as to how to improve it -- they were, after all, part of OASIS for awhile...
Anyone who's been on Slashdot for awhile should remember how much lobbying Microsoft did to try to prevent ODF from taking root in Massachusetts. So, technically, Microsoft didn't try to slow down the standardization process, they merely tried to slow down the implementation process.
Yeah, note the copyright notice at the bottom of the page. Astroturf, anyone?
And from Ars Technica...
And ODF has to support all the features of:
(ripped off directly from a post by this comment [slashdot.org].)
So there you go. I suppose it's possible Word 2007 could have more features than ALL of those, but somehow, I doubt it. The spec isn't bloated because Word is so great, the spec is bloated because Microsoft is afraid of interoperability.
The fact is not that it's impossible -- it could be done, if you want to reverse engineer about five or six generations of Word. It would be difficult, but not impossible, to support enough of the standard to be liveable -- after all, we've done that with the binary Office formats for years.
No, the problem is that it's prohibitively, deliberately difficult for third-parties to implement perfectly, since it references specific quirks on specific versions of Microsoft's products, and the products of others, and doesn't even try to explain what those quirks are, only that you should support them properly. I would say that Microsoft is being deliberately unhelpful here.
If you're going to make it 6000 pages and unhelpful, why not make it 12000 pages, but actually spell out what we're supposed to do? At least then, we could not only duplicate the features in ODF, but we could do them better, the way they were meant to be done. For example: Instead of saying "Emulate Word 95 Full-Width Character Spacing", Microsoft could actually specify how Word 95 implements full-width character spacing. Then, we'd implement specifications that allow the implementation of any kind of spacing you want.
Let me put it this way: In HTML, we could've had, for example: <slashdot-link story_id="07/02/16/1334234" />. That would've been pretty damned convenient for the Slashdot people, but annoying for everyone else, who would have to go to Slashdot to find out how they did it, and in any case, it's much more limited than our current <a href> style which lets you actually link to anywhere. Standards are not about coddling sp