다음을 통해 공유


Another word-processor supporting Open XML

It looks like AbiWord, which is an open source word processing application is now working on Open XML support. https://www.abisource.com/twiki/bin/view/Abiword/OpenXMLImport

They finished up inline formatting about a month ago, and I'm not sure what they are currently working on.

So, now we have AbiWord; iWork; OpenOffice; MS Office; Corel; Gnumeric; iPhone; NeoOffice; Palm reader (and a whole host of other applications and tools) supporting the Open XML format. Huge momentum, and we've only just started.

ISO Update

The 2nd round is now complete and the votes are all cast. I haven't seen the official numbers yet, but some of the early reports show that we already have about 74% of the voting countries in support of Open XML ISO approval. That's actually quite a bit higher than I was expecting (I thought we'd be closer to 60% at this point). There are two criteria that you must hit for final approval (2/3 or P members approval; 3/4 of all voting members approval), and it sounds like we'll be really close on both fronts.

This large scale support should set us up really nicely going into stage 3 of the fast track process. The Ecma TC 45 editor (with the help of TC 45) is now going to sort through all the comments coming in and work towards a more improved specification based on those comments. At that point we'll see a number of the countries who voted "no with comments" switch their vote over to "yes" and we should see Open XML approved after the ballot resolution meeting next year.

Between now and then though there is going to be a ton of work. There will be thousands and thousands of comments to work through. I'm confident that we'll be able to get some solid solutions in place that make it easier to sort and address the comments, but it's going to be pretty intense. I think we have some great momentum right now though which should make it easier for everyone to work together on reaching consensus.

-Brian

OpenXMLCommunity.org Quote of the Day:

Accenture Technology Consulting – Singapore

"Open XML will be a key standard to allow companies to share electonic documents. Approving this standard is key to move to true globalisation and provide services to anyone, anywhere, at any time in the world. Therefore I provide my full support and request that the standard gets approved."

- Andrew Weekes – Senior Executive

Comments

  • Anonymous
    September 04, 2007
    The comment has been removed

  • Anonymous
    September 04, 2007
    The comment has been removed

  • Anonymous
    September 04, 2007
    The comment has been removed

  • Anonymous
    September 04, 2007
    The comment has been removed

  • Anonymous
    September 04, 2007
    So basically, instead of open source developers working to implement one standard for office documents, they get to implement two currently incompatible formats (one a standard, the other a "possible" standard that will most likely be modified in the next six months).  Fun!

  • Anonymous
    September 04, 2007
    Brian, It's not like you to buy into Microsoft's spin by hiding the fact that OOXML failed to get approval this round especially given that a lot of energy were spent on both sides (and that includes you) to influence the event. While it is right that we now need to focus on the next stage, i.e., the BRM stage, to casually cast aside this big "check-point" as if it does not happen is downright not the style of the Brian Jone's I know from this blog. I'm slightly disappointed in you for not reporting that. I understand how and why MS's PR machine will choose to do that, but I did not expect this from you.

  • Anonymous
    September 04, 2007
    Wu MingShi, From my perspective the key phase of this process was always going to be dealing with the incoming comments. That's where the real technical work will be done. Reguardless of the outcome of the vote, we would have moved onto dealing with the NB's comments; and so from my perspective not much has changed. I understand why this would appear to be spin, but it really isn't. Obviously it would have been great if we already had the needed number of votes for approval, but we're pretty close and it's the work that happends over the next 6 months that will really have the greatest impact. -Brian

  • Anonymous
    September 04, 2007
    What great news! This seems like an ideal outcome for everybody: Microsoft and its users will get an international standard, ISO members will get a standard that integrates their comments, and the entire process will improve the standard and add to its legitimacy. I'm sure that higher-ups would rather not have to entertain comments--but for Brian & co., it's a golden opportunity to tie up all the loose ends they may have wanted to earlier but could not due to time and budget constraints. Incidentally, I hope the issue of platform-independence/portability in file paths gets addressed in the comment review process. (E.G., in WordprocessingML files, fields only allow absolute MS-DOS paths--adding a switch for relative paths/URIs, modifying them to use relationships, or replacing affected fields with user-friendly content controls that can refer to files outside of the package.)

  • Anonymous
    September 04, 2007
    Brian, I'm afraid I have to agree with Wu MingShi on this.  I really would expect better of you.  In particular, it is the worst sort of spin to say "In addition to this, many of the P-Members who voted no have said they will change their vote to "yes" once their comments have been dealt with."  As you know full well, there were only three ways to vote in this phase, and the "No with comments" vote ALWAYS means that you will change your vote to "Yes" if the comments are dealt with.  That does not imply a greater acceptance of the standard, and it is quite insulting to try that spin here.  You audience here seems split between those who favor Open XML and those who do not, and none need a whitewash of what the results to date are. That said, I am glad to see the focus will go into improving the standard now.  I hope the comments resolution phase is a productive one, and that substantial improvements can turn this into a standard everybody could support (except maybe that penguin person yesterday).

  • Ben Langhinrichs
  • Anonymous
    September 04, 2007
    I for one applaud Brian's commitment to go over the issues, prioritize them and fix the issues that have been pointed out.   The end result should be fantastic. There is no perfect spec and there is no perfect program, they all have bugs.   The extra QA will ironically make OOXML a more complete and rounded up specification than ODF is. Miguel

  • Anonymous
    September 04, 2007
    Hi Brian, France & New Zealand, if I recall correctly have suggested on their ballots that OOXML and ODF get merged into one format, with the best of each being made into one single standard. I believe this would be the best solution for end-users around the world. Is there any chance of this happening? Thanks.

  • Anonymous
    September 04, 2007
    The comment has been removed

  • Anonymous
    September 04, 2007
    The comment has been removed

  • Anonymous
    September 04, 2007
    Chris, PDF is easy to implement and has been in 100s of apps, OOXML is impossible unless you are a giant multinational corporation, there are 10000 pages to the format. And as for all the ODF to ooxml convertors and visa versa, they all bad and will always be a problem, especially when dealing with documents with tables, graphics and frames inserted in them. In this case there is no choice which will help innovation. They cannot even get ooxml right on a Mac!! It simply is a con job by MS, using deceptive terms like "Open" to get a bogus format approved as a ISO.

  • Anonymous
    September 04, 2007
    The comment has been removed

  • Anonymous
    September 04, 2007
    The comment has been removed

  • Anonymous
    September 04, 2007
    Now that M$ has to deal with the BRM and actually try and fix the broken standard, are they going to honour the reputed claims that they told members who vote "yes with comments" that they will be taken into account during the review and modification phase? As far as I can tell from the rules, there is no documented requirement for this to happen. Al

  • Anonymous
    September 04, 2007
    Having two standards that overlap is not beneficial to anybody in this scenario. I'm on my hands and knees, I beg both the ODF camp and Microsoft to shut up with your bickering and to collaborate on one standard for all documents. Not only will it be good for the computing world as a whole, it will help Microsoft gain the international respect it so clearly wants.

  • Anonymous
    September 04, 2007
    It looks like it is all about the P-members. 17 out of 32 is low. they will need 22 out of 33 so there is a lot of work to convince ate least 6 P-members to change their vote

  • Anonymous
    September 04, 2007
    I notice that the AbiWord support is input_only. This is scarcely "support" for MSOOXML;  it's the sort of thing one would do for a legacy document format. I generally read "support" of a file format to mean the ability to read and write, not just import.  

  • Anonymous
    September 04, 2007
    The comment has been removed

  • Anonymous
    September 04, 2007
    The comment has been removed

  • Anonymous
    September 04, 2007
    Bruce, I'm really glad to hear you say you're interested in co-operation.  The top item on my document format wishlist is a list of features the ODF TC would like to take from the Office Open XML spec for use in future revisions - what do you think the chances are of that happening, and where would I go to find it?

    • Andrew
  • Anonymous
    September 04, 2007
    The comment has been removed

  • Anonymous
    September 04, 2007
    hAL several NSBs specifically called out the conditional approval nature of their votes. It would be a mistake to assume that all of the disapprove votes are votes for rejection. IBM and buddies lobbied hard for all technical comments to be submitted with disapprove votes. I don't think 5 approve votes from the disapprove and abstentions is going to be too contentious - there's a huge amount of duplication in the comments I've seen, with dates and vml being the two most common issues.

  • Anonymous
    September 04, 2007
    The comment has been removed

  • Anonymous
    September 04, 2007
    Thank you, Brian for your response. I guess I was naive to expect large corporations to stop being so nearsighted. I guess we are stuck with another HDDVD-vs-Bluray type debacle in our future, where everybody loses in the long run. I know that deep inside you agree that co-operating with ODF to create one standard is the right choice, but as usual the higher-ups have their own agenda.

  • Anonymous
    September 04, 2007
    The comment has been removed

  • Anonymous
    September 04, 2007
    The comment has been removed

  • Anonymous
    September 04, 2007
    Is there a list of applications that are certified to write OOXML with full fidelity ?   Most of the applications you mentioned are read only OOXML / import one way, but I am unaware of any major applications that can write OOXML.

  • Anonymous
    September 04, 2007
    "Between now and then though there is going to be a ton of work. There will be thousands and thousands of comments to work through..." Pity MS is going to spend some of their own money to finish baking their own spec. Oh well, it was worth a shot to just throw 'er out there. A couple more attempts and you guys will have a nice standard! <w:fldData>CNDJ6nn5us4RjIIAqgBLqQsCAAAACAAAAA4AAABfAFQAbwBjADEANwA1ADUAMwA3ADQAOAAwAAAA</w:fldData>

  • Anonymous
    September 04, 2007
    The comment has been removed

  • Anonymous
    September 04, 2007
    The comment has been removed

  • Anonymous
    September 04, 2007
    >Another word-processor supporting Open XML To say Abiword supports OOXML is a bit of a stretch. http://www.abisource.com/mailinglists/abiword-dev/2007/Jun/0030.html You must have more confidence in the open source development model then I do.  One Google summer of code intern working for a couple of months can implement a full fledged document standard;)

  • Anonymous
    September 04, 2007
    The comment has been removed

  • Anonymous
    September 04, 2007
    The comment has been removed

  • Anonymous
    September 04, 2007
    @Alan Alan, there were actually 51 approval votes and 18 dispproval votes in the total vote. So your suggestion that they counted the abstain votes to get to that number of 51 is incorrect !!!

  • Anonymous
    September 04, 2007
    The comment has been removed

  • Anonymous
    September 05, 2007
    The comment has been removed

  • Anonymous
    September 05, 2007
    Brian,  I know I keep asking this, but I don't think you've replied (sorry if I did miss a reply).  If you do indeed like the idea of two standards and let people decide between them, does that mean that MS Office will support ODF?  Also a while ago you said that the binary format specifications are free to get.  I've gone through the process and now at the point where I have to mail the forms off.  Do you know if it's possible to get the formats online instead?  Also if I have read the terms and conditions correctly, I think I'm allowed to put the specifications online.  Can I confirm that I'm allowed to put them online once I have them? John

  • Anonymous
    September 05, 2007
    The comment has been removed

  • Anonymous
    September 05, 2007
    The comment has been removed

  • Anonymous
    September 05, 2007
    @Craig Matthews Just because an absurdity is described in a schema does not make it less absurd. Anyone who did SOA knows it. <disgression> WS-X "standards", brought to you by the evil IBM. Oh, wait, they did it with the nice Microsoft. Does that make WS-X evil or nice I wonder? </disgression>

  • Anonymous
    September 05, 2007
    MiliTux, it's "pyrrhic", for the reason Miguel stated.  The extra QA will make OOXML 1.0 a better quality spec than ODF.  You've bought yourself 6 more months to try to convince as many governments as you can that they should all move to ODF because it carries the ISO stamp of approval.  But after that, if OOXML is approved, it'll be a much better spec than ODF 1.0 (which is a pretty bad spec to begin with), and governments will likely use it instead based on the extra QA time that you've given it. This process is essentially demanding that ECMA deliver a 1.1 quality spec as the first spec.  ODF 1.0 is filled with holes that 1.1 is trying to patch.  Even that doesn't have decent spreadsheet formula support, that'll only happen with 1.2. OOXML, on the other hand will be a 1.1 quality spec, with no need to rush out another version to patch the problems of the first one.  Because that paching is going to take place NOW. You guys shot your foot off and don't even know it.

  • Anonymous
    September 05, 2007
    And why doesn't Microsoft help improve ODF instead of doubling the effort required for anyone that wants to support document formats?

  • Anonymous
    September 05, 2007
    The comment has been removed

  • Anonymous
    September 05, 2007
    Alan, You are correct.  Microsoft has no interest in putting out a good spec.  They want to do exactly the minimum amount of work to make it an ISO standard so that governments and others won't choose ODF.   Once we hit that point, the spec will rot and Office will  implement just enough of an incompatible version of the spec that it won't matter.

  • Anonymous
    September 05, 2007
    @Bruno So - let's see how this would be Pyrrhic. Perhaps the ODF camp stopped MSO-XML, but also lost their own ISO approval. No, that's not it. Maybe ODF supporters are defecting or quitting in large numbers because MSO-XML is better. Not that either. From Wikipwedia (check your own sources as you care to) "In both of Pyrrhus's victories, the Romans lost more men than Pyrrhus did. However, the Romans had a much larger supply of men from which to draw soldiers, so their losses did less damage to their war effort than Pyrrhus's losses did to his." Doesn't sound like ODF has suffered any such loss. Even if ODF is completely displaced, without the ODF effort MSO-XML would be the only choice. It would not be an ISO or even ECMA standard. You would likely have to sign the same agreement to get whatever type of documentation you are supposed to get for the binary formats now. Instead the ODF effort has exposed a weakness of dreadnaught mentality in the age of Internet-nimble. Add to that the rapid dissemination of information concerning corruption and it looks like MS was hoist by their own petard. After you are done with Pyrrhic, you might look up petard. It's under "P."

  • Anonymous
    September 05, 2007
    Yawar Amin, You may be right in your conclusions, but your reasoning is a little skewed in places.  You might be interested in reading Jason Matusow's blog, as he talks more than Brian does about the issues you seem to be interested in. You make an excellent argument for why Microsoft should be allowed to have their own file format, but their right to have a format doesn't imply that it's in the ISO's interest to certify it as a standard.  If Microsoft fails to get Office Open XML certified, they'll still have a modern, XML-based file format which works well with Microsoft Office, it's just that those few people whose needs are stringent enough to require ISO approval will know that it's not the format for them. The HD-DVD/Blu-ray argument is apt in some ways, but not others - for example, the problem with a single document format isn't that it would restrict competition, it's that it would restrict innovation.  Innovations on a standard can only occur within the areas where the standard permits, for the sake of interoperability. There's always a trade-off between innovation and interoperability, and I think there's plenty of room for people to disagree about the optimal balance.  I'm not going disagree if you believe Microsoft's got the balance right, but I do believe that many of the people that think Office Open XML shouldn't be a standard have a different opinion about the optimal balance.  I have a growing respect for the market, but I think that opposing Office Open XML's standardisation is the appropriate course of action if you believe that gaining a little more interoperability is worth sacrificing a little more innovation.

    • Andrew
  • Anonymous
    September 05, 2007
    Andrew Sayers, Thanks for your reasoned response! It was actually a pleasure to read compared to most of the comments here. As you've guessed, I do believe we can afford to sacrifice some interoperability now to reap benefits from future innovations. Look at the phenomenal coverage both formats are getting -- ODF v OOXML must be one of the most talked-about issues on the internet. This HAS to be a good thing -- more bugs are ironed out, more people are getting into the discussion and there's generally a lot more awareness. And we're already seeing that innovation which comes from competition. Microsoft is strategically positioning custom XML schemas mixed in with its documents as a competitive advantage of OOXML -- something they, strictly speaking, didn't have to do but did anyway because they saw an opportunity. You have to applaud that. ODF's response, at least to me, has been vaguely unsatisfying. What competitive advantage does ODF have? Better design, more industry backing? Whatever it is, I'm sure the public scrutiny of both formats is forcing the ODF people to work harder to improve their offering. And there's nothing wrong with that. A rising tide raises all boats. As for whether it's right or wrong to make OOXML an ISO standard, I say if anybody can afford to make their offering a standard, go for it. May the best standard win.

  • Anonymous
    September 05, 2007
    Yawar Amin,, The utility of embedding custom XML schemas is no more than interchange standards always have - it's limited to the sender and receiver agreeing before hand as to what the custom part means. The minor advantage in using XML is XML parsers. Interchange of custom information can already be done. A ccol way to do this is to use .zip format to attach interchange information to a document, or group of related documents by simply adding it to the .zip archive. Since both ODF and MSO-XML both use .zip methods, there can be no such objection to using .zip. Since this method does not rely on the format of the document(s), it's a way to empower users who don't want to convert existing documents or for documents that have no XML representation. In this way rich content can be packaged super easy with minimal tools and no conversion problems. Instead of opening up an exciting world of possibility, embedding custom XML schemas directly into MSO-XML format documents can limit choice to MS applications alone, while leaving the existing hurdle of agreeing on an interchange format for the custom data. I'll leave you to ponder the relative public scrutiny that a document 6000 pages long gets relative to a document which numbers in the hundreds pages. Finally - Look at sfgate.com/cgi-bin/article.cgi?file=/c/a/2005/10/03/MNG6GF1NU51.DTL "Standardization of equipment is vitally important ..."

  • Anonymous
    September 05, 2007
    I'm rather disappointed to see such an article on codeproject, but it's worth a read: http://www.codeproject.com/useritems/ooxml_is_defective.asp Can you comment Brian?

  • Anonymous
    September 05, 2007
    @JasonG That article is about MS Office implementation of OOXML (which is by noo means a perfect implementation) more than OOXML itself.

  • Anonymous
    September 05, 2007
    JasonG, Stephane's articles have the credibility of an editorial on a Rupert Murdoch property.   They are far from scholarly works, they are usually sloppy arguments and usually goes for volume instead of accuracy. I replied to Stephane's claims here: http://developers.slashdot.org/comments.pl?sid=279895&cid=20363627 Furthermore someone in Slashdot actually explained why the floating point issue that Stephane brought up is actually correct. Miguel.

  • Anonymous
    September 05, 2007
    The comment has been removed

  • Anonymous
    September 06, 2007
    @Bruno What you fail to understand is that with the possible exception of competing vendors no one would mind an ISO-approved OOXML. Because a good (clear and complete) spec would be easy to implement by everyone (competing Office suites, backend systems, etc) Current OOXML is not this. It includes caricatural stuff like "tag legacy elements so they're passed for interpretation to fossilized bits of old Office releases only Microsoft has access to", and many other fundamental problems. A quality OOXML would have passed 32 to 0 like ODF.

  • Anonymous
    September 06, 2007
    "Approving this standard is key to move to true globalisation ..."

  • Andrew Weekes – Senior Executive The ISO standard shipping container is key to globalization. (He's likely of Brit origin, so I'll give him the 's' and keep the 'z' for me.) That and GPS and GSM. This way goods can be packed, kept secure, moved accurately, identified, located at the destination, and using standard handling equipment, be attached with standard attachments to vehicles for transport to where the ship can't go. Check what happens when a new, streamlined approach is taken over trying to fit all the old stuff into a new form: en.wikipedia.org/wiki/Containerization Funny thing - the proliferation of container standards was seen as a bad thing, hence the ISO standard. The shipping industry has been  a steady haul to cut the number of sizes, probably to one. As to the services Andrew would like to see for anyone at any time in any place in the world, I'd first like winning lottery numbers for the next couple of multimillion dollar lotteries. OK. Scratch any time. Let's go to the Sahara. Will MSO-XML help more than a satellite phone? Scratch any place. Last chance. MSO-XML ratification will certainly help the Kurds or the Sudanese. No? Scratch anyone. This leaves the entire quote as trite hyperbole, not a reasoned statement of support. By the by. It is "electronic" not "electonic." No spell checkers on the MSO-XML side?
  • Anonymous
    September 06, 2007
    Dave S., Good example.  And for an example of what happens when the world fails to unite around a single standard, see http://en.wikipedia.org/wiki/Domestic_AC_power_plugs_and_sockets However, these are both examples where the benefits of further innovation outside the scope allowed by the standard would be very low. To take one of the big differences between ODF and Office Open XML, do you have any evidence about whether run-based XML is objectively better or worse than hierarchical XML for office automation?

    • Andrew
  • Anonymous
    September 07, 2007
    The comment has been removed

  • Anonymous
    September 07, 2007
    The comment has been removed

  • Anonymous
    September 07, 2007
    The comment has been removed

  • Anonymous
    September 07, 2007
    Looks like Brian is finally showing his stress and disappointment! It's not too late to switch sides, Brian.

  • Anonymous
    September 07, 2007
    n/a, The most astute thing I've ever heard Stehpane Rodriguez say is that Office Open XML is Brian's baby.  Brian's naturally protective of his baby, so it's understandable that he gets upset when people walk into his blog and start accusing him of having an ugly baby - which I think I just did. Brian, Using the term "new dawn" like that was sloppy of me and I apologise if I caused offence. The point I was trying to make is that (sensible) proponents of ODF seem to think in terms of starting from scratch, with no duty to support features that they dislike in older formats - figuratively speaking, like the start of a new day.  This is very different to Microsoft's approach, making incremental changes that minimise the risk of problems while upgrading. I respect both approaches technologically, and I'm not trying to insinuate that one is better than the other, but it's an important difference and I thought that "new dawn" was a nice way of summing up the mindset of the ODF camp. As to my personal opinions about the strengths and weaknesses of different formats - I'm just now getting to the point where I can form an opinion that I can express in words.  My opinion changes most days, but I've never been in any danger of declaring ODF to be a giant leap forward compared to Office Open XML.

    • Andrew
  • Anonymous
    September 07, 2007
    The comment has been removed

  • Anonymous
    September 07, 2007
    Brian, From msdn2.microsoft.com/en-us/library/bb510135.aspx re content controls. "This new functionality increases the speed with which a template designer creates documents. Not only that, but the templates are more user-friendly and robust. You can load content controls with a wealth of information by mapping to custom XML data. Content controls are predefined pieces of content. There are several types of content controls, including text blocks, drop-down menus, combo boxes, calendar controls, and pictures." I'm glad I looked. I thought content controls controlled content, as in document process/security/privacy. I'm disappointed. Now I see that content controls are a way for a lot of people to create whatever schemas (seems like the right term) they want in whatever fashion they want. Just what the world needs, more tools for the half-baked to get their hands on.

  • Anonymous
    September 08, 2007
    "AbiWord; iWork; OpenOffice; MS Office; Corel; Gnumeric; iPhone; NeoOffice; Palm reader" Aye. Abiword, the flagship of enterprise wordprocessing software solutions also supports your broken "standard", just a little bit. Why don't get real and support ISO 26300? Nobody believes your "double standards" and "user needs" lies. It stinks. -- Gerd

  • Anonymous
    September 08, 2007
    Gerd, It's also a shame that Microsoft had to drag the ISO name through the mud to push their standard.  I hope Brian is happy with his accomplishments!

  • Anonymous
    September 09, 2007
    What really counts is that OOXML is open - maybe not in the way the standard was created, but in the sense that it is based on XML, openly documents, an this allows 3rd party tools to work with the data in uprecedented ways! While the ISO process may take a while to complete, what will undoubtedly happen is that OOXML becomes a de-facto standard as enterprises world-wide upgrade to Office 2007... For more details, see my XML Aficionado blog entry:http://www.xmlaficionado.com/2007/09/office-open-xml-fails-to-win-iso.html

  • Anonymous
    September 10, 2007
    Just a related note, anybody thinking that only few Linux zealots are making a big noise about OOXML by complaining few editorial omissions in the "superb" OOXML specs I suggest people to take a look at one example of comments about the spec: http://www.jtc1sc34.org/repository/0904c.htm In the zip file you'll find tens of Word (!) documents that contain an incredible long list of comments. They undeniably underline the fact that the spec was never properly reviewed, more likely it was constructed in a hurry and knowingly omitted some key definitions. If you don't agree, just please read the docs again, the amount of comments is just breathtaking!

  • Anonymous
    September 10, 2007
    In Sweden, according to news reports, the number of voting members of SIS grew from 9 to 34. Of the 23 new companies that joined, all are Microsoft certified partners voted in favour of OOXML.

  • Anonymous
    September 10, 2007
    After reading the comments to the OOXML standard, it appears that many of the NBs are expressing concern that the OOXML can only be implemented on a Microsoft Product. Why don't you dedicate a few programmers to implementing ODF on Open Office?  A fully conforming implementation of OOXML on a non-MS based product would go along way towards swaying OOXML's detractors. Once it has been proven that OOXML can be fully implemented, all or most of the no votes will become yes votes.

  • Anonymous
    September 10, 2007
    Anonymous coward: that ZIP file is chock full of good comments on OOXML.  Most of them are dealing with OOXML's Windows+Office-specific baggage that shouldn't be in a proper spec. Good find! Some nuggets from the first file: "This element defines values for use on Windows and Macintosh platforms,  but not for any other operating systems.  " "The usage of null-terminated C- style strings is avoiding XML and  will cause the markup to interoperate poorly with XML-based tools." "The allowed values of this enumeration, EMF, WMF, etc., are Windows specific formats. No allowance seems to have been made for use by other operating systems." "This describes an "ink" element which stores "ink annotations in an application-defined format." This is apparently intended to store annotations, used with tablet input devices to add hand-written annotations to documents. These annotations are often a vital part of documents and their specification is undefined in OOXML. We are opposed to standardizing placeholder elements for entirely application- dependent proprietary formats without also specifying an interoperable format for those who with to create interoperable formats." "The formatting system described here is not comprehensive, lacking, for  example, support for Armenian, Tamil, Greek alphabetic, Ethiopic and  Khmer numerations, all in use today, as well as the various historical  systems still used by scholars." "Length is said to be “exactly 3 characters”. This is inconsistent with the example given which has a length of 6 characters."

  • Anonymous
    September 11, 2007
    Any chance I could ask you a question?  For the life of me I cannot get documents to display in Print View even if I use : <w:view w:val="print"/> Using 2003, and I've resorted to messing with the document properties of the Word Document object that is generating these, but still not working.  Can you help?  Thanks.

  • Anonymous
    September 11, 2007
    Forgot to tell you if you have time to answer my question about print view in xslt that the document is being displayed in a browser and that my e-mail if you care to write is zebedatious at hotmail.com

  • Anonymous
    September 12, 2007
    The comment has been removed

  • Anonymous
    September 12, 2007
    "I doubt that MS will be wasting too much time in trying to change the BSI's vote. " yes... but they have the "positive" ISO P-members newcomers to balance things!: Côte-d'Ivoire: inconditional APPROVAL Cyprus: inconditional APPROVAL Jamaica ( i love reggae :-) : inconditional APPROVAL Lebanon: inconditional APPROVAL Pakistan: inconditional APPROVAL Malta ( beautiful island ): APPROVAL with comments ( lightly comments indeed ) Turkey: approbal with COMMENTS Uruguay: APPROVAL with comments Venezuela: APPROVAL with comments Trinidad Tobago: Abstention ( for now ;-) ECUADOR: DISSAPROVAL with comments (UPS... what happened ? not enough gold members there? ) ( source: http://www.jtc1sc34.org/repository/0904.zip )

  • Anonymous
    September 12, 2007
    The comment has been removed

  • Anonymous
    September 12, 2007
    The comment has been removed

  • Anonymous
    September 12, 2007
    For backwards compatibility, could the spec be made flexible enough so that special tags aren't needed? For example, instead of saying do a table like word95, could it not make the table format flexible enough to support a style that is similar to the way word 95 did it?  Then "like word95" would simply become a style.

  • Anonymous
    September 12, 2007
    The comment has been removed

  • Anonymous
    September 13, 2007
    hAl,  No not like that.  I mean, for example, instead of saying that Table like Word95, you instead set the table with a certain border, etc.  So you style it to look like a word95 table. John

  • Anonymous
    September 14, 2007
    John, that would still require the specs of such a style to be published in the format specification or to leave such styles implementation defined and therefore not interoperable at all

  • Anonymous
    September 14, 2007
    It seems that ISO has now rejected the fast track process for OOXML.  Google has published an official thank you to the ISO for this.

  • Anonymous
    September 15, 2007
    This is why we can't believe Brian when he says that Microsoft respects standards: http://tuxdeluxe.org/node/255 Any Microsoft-controlled standard will be subject to the shipping code of Office (bugs and all).  So, Brian, where's the use in the standard if you won't respect it?

  • Anonymous
    September 15, 2007
    BTW, link for Google's thank-you to the ISO: http://google-code-updates.blogspot.com/2007/09/google-welcomes-iso-decision-on-ooxml.html

  • Anonymous
    September 17, 2007
    Hey folks, sorry I've been offline of late. Rather than try to reply to all the comments I'll pull together a new post. I'm also working on the FAQ idea Andrew had suggested several weeks ago. Hopefully I can get that going too. It's also a very busy week, and our puppy is sick today so I may spend most of my day at the vet unfortunately. Sorry again for not being more active lately... -Brian