Slashdot Mirror

Have a read of ISO OOXML convener: Microsoft's format "heading for failure", from which I quote:

Microsoft's failure to adopt the standard version of the format after two years has drawn criticism from Alex Brown, the convener of ISO's OOXML subcommittee (SC34). Brown was consistently supportive of Microsoft's push to obtain fast-track approval for OOXML during and after the ISO review process, but his optimism appears to be waning. In a recent blog entry, Brown contends that Microsoft is not fulfilling its commitment to adopt the ISO's edited version of the standard.

I would quote directly from Alex's blog, but sadly the "Microsoft .NET Framework Version:2.0.50727.3634; ASP.NET Version:2.0.50727.3634" serving the blog post has NullReferenceException.

Some other thoughts come from OOXML is defective by design, where this post Backwards compatible? One more lie by omission is worth reading.

The nightmare wrt google docs is writing a doc in word, passing it to google docs for somebody's editing,

That's the nightmare for ANYONE trying to inter-operate with Microsoft.

And since it's the result of deliberate efforts by Microsoft to fight open standards, it should result in them being banned from government tenders.

http://www.adjb.net/post/Microsoft-Fails-the-Standards-Test.aspx
http://www.consortiuminfo.org/standardsblog/index.php?topic=20051116124417686
http://blogs.computerworlduk.com/open-enterprise/2012/04/1how-microsoft-fought-true-open-standards-i/index.htm

As far as I can tell, the problem here is that the article is not bending the truth to match the usual reality-distorted pro-ODF bias expected by slashdot users and other FSF goons.

Let's start with this statement:

In the ODF article, Alex Brown bends the truth to make it seem like no one is supporting ODF, and that it is a flawed and incomplete standard.

It seems to be like he doesn't fail to bend the truth. It's a flawed and incomplete standard, in some ways it is vague, in others it's simply inconsistent.

Let's take tracked changes for instance, a feature in ODF 1.1 which pretends to be complete. The reality is that the standard is so vague and broken that the most popular implementation, Google Docs, ignores the standard entirely, implementing changes in their proprietary system. Microsoft simply solves the problem by disabling the functionality in order to avoid future breaking.

http://blogs.msdn.com/dmahugh/archive/2009/05/13/tracked-changes.aspx

Let's not talk about ODF 1.2 either, since its only a working draft.

So Microsoft was somehow able to do a perfect by the letter implementation of the ODF 1.1 (the current standard) spec, and yet they haven't got full interoperability with OpenOffice? It sounds an awful lot like Sun took a very liberal interpretation of their vague standard and are now standing by their wonky mess of source code (Ooo) as the standard-- similar to Solaris and POSIX. Thats unacceptable, ODF passed the standards bodies, not OpenOffice.org.

The fact that Microsoft could create one of the only correct implementations of the ODF standard and still break interoperability suggests that the ubiquity of this standard is largely overstated:

http://adjb.net/post/Notes-on-Document-Conformance-and-Portability-4.aspx

There are arguments to be made on the subject of digging through Sun's source code to make this vague standard work, but then ODF violates the FSF's very quote bashing MS-OOXML:

"For any standard it is essential that it is implementable by any third party without necessity of cooperation by another company"

Source: http://fsfe.org/documents/msooxml-questions

So, you can't make interoperable ODF without referencing OpenOffice because it is vague and incomplete... but it's not a complete standard unless you don't have to rely on the assistance of a certain corporation (Sun) to implement it properly?

It sounds to me like ODF is locking functionality to Sun's software the same way DOC locks functionality to Microsoft Office. MS-OOXML may be wordy, but it turns out that you need a lot of words to make a complete office standard. ODF is a paper tiger, end of story. The problem slashdot points out here is simply a lack of reality-distorting pro-ODF bias... this "whisper campaign" might be the seeping shadow of "reality" in the reality of writing a complete and interoperable standard escapes Sun--leaving them with something terse but heavily marketed with a vicious and aggressive activist campaign by angel advertisers who fancy themselves freedom fighters.

Before you judge on this issue, it helps to read comments by various involved parts - those raising the issue to attention, MS people who have implemented ODF, and informed commenters outside this dispute. So, here's a bunch of links to start with.

First of all, a series of blog post by OASIS' Rob Weir (who's criticizing MSOffice) and Microsoft's Doug Mahugh (who's defending it) that evolved into a kind of a public discussion on the issue. Here they are in chronological / meaningful reading order:

http://www.robweir.com/blog/2009/05/update-on-odf-spreadsheet.html
http://blogs.msdn.com/dmahugh/archive/2009/05/05/odf-spreadsheet-interoperability.aspx
http://www.robweir.com/blog/2009/05/follow-up-on-excel-2007-sp2s-odf.html
http://blogs.msdn.com/dmahugh/archive/2009/05/09/1-2-1.aspx
http://www.robweir.com/blog/2009/05/battle-for-odf-interoperability.html

Then there's some outside commentary. I've taken the following links from comments in Doug's blog posts, and they tend to either be neutral or side with MS on this, so it may not be a representative sample. If you have any representing informed argument for the other side (e.g. by members of ODF committee, or ODF implementers - in general, people who know the ins and outs of the spec, and can accurately judge on its wording and intent - not random blogosphere FUD from either side), please mention them in replies.

http://ajg.math.concordia.ab.ca/?p=4
http://adjb.net/post/Notes-on-Document-Conformance-and-Portability-4.aspx
http://broadcast.oreilly.com/2009/05/odf-11-formula-support-in-office-sp2.html

The point of standards is interoperability. MS is legally required, due to their monopoly influence, to make a valid attempt at interoperability. Hiding behind the standard doesn't make their actions any more legal or ethical.

Hiding behind the lack of a standard doesn't make the ODF Alliance correct.

You're being obtuse or you don't know enough about software coding to know what a "reference implementation" is. A published standard tells you how to write to the standard. A reference implementation is an open source implementation of that standard that can be used when there is doubt about how the standard is to be implemented.

OpenOffice is a reference implementation? Are you sure? It has far too much legacy cruft to fit that bill... if their reference implementation is a massive pile of unix garbage hacked up from a 90's office suite, they've got bigger problems on their hands than Microsoft interoperability. OpenOffice.org is not the reference implementation of ODF in the same way going on a forum or IRC channel does not constitute documentation.

Is it, however, reasonable to assume the way OpenOffice, KOffice, Symphony, Google Docs, and Sun's ODF plug-in implement it is the most interoperable way?

The standard is actually pretty open-ended, so the various implementations are not really consistent. Microsoft's implementation of the spreadsheet, for instance, is fairly on track with the way Symphony implemented it.

I don't believe this is true. They both broke compatibility and screwed up parts of the implementation making it not quite conforming.

I don't think the other parties really want to take on Microsoft in terms of conformance with ODF 1.1:

http://adjb.net/post/Notes-on-Document-Conformance-and-Portability-4.aspx

Who cares what Sun wants? This is about having useful interoperability among office suites so we can have some real competition and choice going forward. No one company should be able to hold up progress. MS needs to be stopped from playing these games and hurting the industry in the process.

I would argue that parties like Sun and Google are maintaining a hacked and inconsistent basis under the mask of standards. If they want to play the standards games, they should write standards that are both complete and that they are able and willing to adhere to.

Sun matters because OpenOffice is a Sun product and OASIS is basically Sun.

In a meeting, except as otherwise specified in these directives, questions are decided by a majority of the votes cast at the meeting by P-members expressing either approval or disapproval. (9.1.4)

Then at the meeting he used different rules and allowed non-P-members to vote too ... these non-P-members were the late-joining African nations that favored Microsoft. The ISO organiser even admits he changed the rules,

[Update 2008-03-06. This was the wrong clause. In-meeting Fast Track BRM voting is for resolving the comments of a constituency determined by the combined voting procedure (O-members + P-members) as per the JTC 1 Directives 9.5, and that is the understanding of the "normal JTC 1 procedures" in 13.8.]

Source docs

I find that most of your claims are false or incorrect.

(1) The 6,000 pages includes a tutorial, and the 1,500 pages actually are clarifications. If you had read any of those clarifications you would have seen that the majority of them are small editorial changes so it hardly amounts to your implied "7,500" pages.

(3) It is not "underspecified". Had you bothered to actually look up the actual issues, you would have seen that things like "lineSpaceLikeWord95" were fully specified as it was one of the major issues that people raised.

(4) There is no risk of "Microsoft patents". The whole OOXML is covered by the OSP, and Microsoft has announced that their commitment to the OSP will go beyond the current spec and will continue as OOXML is evolved, this is the identical claim that Sun and IBM have made for ODF regarding patents.

(5) There are not "the thousand or so comments brought to the BRM and not individually addressed", every comment was addressed. There were an incredible number of repeated issues submitted (people copy-pasting from Groklaw, IBM talking points or members from one country editing the content for other countries). Once sorted and made unique they came down to a thousand. Every single issue came with a proposal to clarify or address the issue.

ECMA met with various countries (the invitation was open to every country) a month in advance of the BRM to walk countries through the comments, get their feedback and update the proposed text /in advance/ of the meeting in Geneva.

All the resolutions, statistics and the actual comments are available here:

http://adjb.net/comments.php?y=08&m=03&entry=entry080306-082306

Going back to why Slashdot hates OOXML. It is very clear, because its Microsoft.

Alex Brown, the BRM convenor, talks a little in his blog about the issues in SC34 and how things are not quite as they are portrayed in various discussions on the internet.

http://adjb.net/index.php?entry=entry071215-123604

In fact it is only 1000 distinct comments if you read the blog of the Alex Brown who will be convening the ballot resolution meeting http://adjb.net/index.php?entry=entry071024-180256.

The convenor of the BRM on his blog stated there were only 1030 distinct comments left after checking the comments for duplicates. This is not strange as certain minor editorial comments have up to 100 (nearly) identical instances submitted
http://adjb.net/index.php?entry=entry071024-180256

The large number of identical or near identical duplicates (almost 3000) are caused mostly by for instance IBM creating a large list of comments which was send to a lot of national bodies and by several of those submitted (almost) fully in their ballot response whilst some other national selected only certain comments from them.

Reading Brian Jones blog it also says that Ecma has already responded to 622 of the comments and in most cases altered them according to the suggested comment by the national bodies