다음을 통해 공유


Responses now complete for all 3,522 comments

Yesterday night Ecma TC45 and the Editor of DIS 29500 were able to publish the final drop of our responses to the 3,522 national body comments. The Ecma status report just went live today, and you can access it here: https://www.ecma-international.org/news/TC45_current_work/Proposed%20dispositions%20for%20National%20Body%20comments%20on%20DIS%2029500%20complete.htm . Many of the national bodies have already been following along with the progress over the past couple months, as Ecma continued to post the latest responses; and I think they'll all be happy to see the final draft ready.

It's been a ton of hard work over the past several months, and it really feels great to move onto the final stage of this process (I need some sleep). It's unbelievable how much work we've been able to accomplish within TC45. Similar to how we moved from a 2,000 page spec to a 6,000 in 2006, in 2007 we were able to respond to 3,500 comments and generated a 2,300 page document (a bit less that a page per comment) where I believe we were able to successfully handle the national bodies comments.

The most impressive part about TC45 was the diversity of skills and experience we had on the group. We had a large number of conference calls and a week long meeting in Japan to get to this point. Many people on TC45 actually worked through the winter holidays and through their own personal vacation time (thanks again everyone!). There were folks like BP who have huge amounts of documents in the corporation and their key concern is access to those documents and ensuring that they could use different tools on different platforms to access those documents. The US Library of Congress had a huge amount of experience in document archival, and the level of detailed review they brought was huge. The folks from Novell and Apple were both really helpful as they had already built Open XML implementations and they were able to provide the point of view of an implementer. It's not possible to name all the folks who helped make this happen, but I know that they are all as happy as I am with the results. We were able to help the editor pull together an even better specification, and we're all excited to continue moving the standard forward over the coming years.

-Brian

Comments

  • Anonymous
    January 14, 2008
    Congratulations on completing this stage of the work.  Sounds like a huge job.  I have one question, which I know is probably premature because this is just a proposal, but how do you see the deprecated part working?  Not in the format so much as in MS Office.  In other words, if a binary document which uses a deprecated feature is opened with MS Office 20xx (whenever that is released), do you foresee changing the contents to match the non-deprecated logic, or simply using the deprecated features for that document?  I can see pros and cons for either approach for MS Office, but it seems like for the sake of OOXML as a standard, you would really want to convert to non-deprecated features if at all possible.  Any thoughts on that (and again, not so much what the format will require as what you think MS Office might choose to do, assuming the standard does not require one approach or the other)?

  • Anonymous
    January 14, 2008
    having seen +1000 pages of one of the +40 PDFs generated by ECMA (Microsoft), only one word come to my mind and my heart: pathetic this DIS 29500 is becoming an amorphous thing ... very worring ... keep devaluating standardization, Microsoft ! Congratulations Brian, you, Microsoft and ISO are making history ;-)     carlos

  • Anonymous
    January 14, 2008
    Carlos, Maybe you can enlighten us exactly as to what you found to be pathetic.  

  • Anonymous
    January 14, 2008
    Again, the same frustration of not being able to see the alteration because I am not NB members. Any chance of an white-out/black-out/non-classified version?

  • Anonymous
    January 14, 2008
    >Carlos, >Maybe you can enlighten us exactly as to what you found >to be pathetic.   may be you could enlighten yourself, finding out about the process and deliverables of true and serious standards:  C, C++, PDFs, etc.   good luck ! :-)       carlos

  • Anonymous
    January 14, 2008
    Carlos, the standards you mention are only "serious" standards because they have been in use for years. They went through, and are still going through, a standardization process. You need to give OOXML some time in the industry before you make a claim of pathetic. Otherwise you look like a foolish prognosticator trying to garner attention.

  • Anonymous
    January 14, 2008
    This is wonderful news! I hope they give you extra vacation time for all your hard work!

  • Anonymous
    January 14, 2008
    The comment has been removed

  • Anonymous
    January 14, 2008
    Francis, Thanks! There is still a lot of work to do this winter and spring, so my vacation is probably going to have to wait until closer to April. My wife's trying to get me to clear at least a few days so we can get away for a long weekend though... (we just got another new weimaraner puppy so we want to take our two dogs out somewhere with snow) -Brian

  • Anonymous
    January 14, 2008
    The comment has been removed

  • Anonymous
    January 14, 2008
    >Carlos, >Did you actually read through the responses? yes >I'm extremely proud of the work we did in TC45 congratulations for your optimistic view of reality. I would be ashamed to have to change so much of a proposal ... this is a sign of the lack of quality ( i.e: rush ) of the process >It shows the effiency of TC45, and the willingness >to continue improving on the standard. you aren't  improving the standard, you are correcting it you have demostrated that you don't know how to draft a standard. Generating +10000 pages of changes [1] you are actually devaluating the words "standard", "standardization" and "fast-tracking" >That's how standards and software work, >they continue to evolve and improve over time. this is how standards work when money and marketing needs are the drivers of the standardization process. Proper review and technical merits are third-class citizens.  Carlos [1] http://www.ecma-international.org/news/TC45_current_work/Proposed%20dispositions%20for%20National%20Body%20comments%20on%20DIS%2029500%20complete.htm

  • Anonymous
    January 14, 2008
    The comment has been removed

  • Anonymous
    January 14, 2008
    The comment has been removed

  • Anonymous
    January 14, 2008
    Where are the corrections so that I can see the probably ugly face of AutoSpaceLikeWord95? It seems that ECMA is totally incompetent to setup a public website to publish its comments?

  • Anonymous
    January 14, 2008
    Thank you Brian for considering a redacted version. I agree it is a big job. I will be satisfied if the the latest uptodate specification, i.e., the one that will be made public as the ISO standard assuming BRM goes ahead and no changes there. After all, there is a lot of people out there will do a "diff" of both documents. Sure, there will be a lot of noise from the opposition camp, but hey, the more scrutiny the standard receive and the better the standard will be (to their chagrin) In the long run, it is not the comments that are important, but the actual ISO specs itself. It is true that by not publishing the comments we will not see the argument behind them. But I am sure there will be noises about them. However, the noise level can indicate to us which points are important and we can then concentrate on these. With this, we can discuss (repeats) points on the perceived important points without falling foul of the confidentiality rule

  • Anonymous
    January 14, 2008
    Rien à voir avec la bataille Thermophyles ou les spartiates n'étaient que 300. Le processus de standardisation

  • Anonymous
    January 14, 2008
    @Brian Jones, Unsurprisingly, the major issue about a migration format such as ECMA 376 remains unaddressed. The migration format document does not contain the migration tables, therefore Microsoft is excluding others from reliably migrating existing files to the newer file formats. This is in direct contradiction with what  standards are for. PS : the so-called growing number of Open XML related projects is hypocrisy at best. It does not cost more to start a project related to binary formats, and there are gazillions such projects out there. Since those projects are just started, they always allow Microsoft to conveniently ignore the main issue : there is no serious and reliable non-Microsoft implementation out there at this point. A little more genuine words don't hurt.

  • Anonymous
    January 15, 2008
    Am actually surprised that Brian even bothered to reply to Carlos at this point. It is clear from his various posting that we got a troll in our hands, and one that probably saw the page count and figured "time to post!" considering his lack of analysis or his reply. And of course, Carlos has probably never participated in any other standard efforts.   OOXML has improved thanks to this process, and will become a much better standard as a result and this seems to bother a lot of people as it might mean that they have lost a year long FUD battle.

  • Anonymous
    January 15, 2008
    S, Am not sure that you can blame Microsoft for a lack of "serious and reliable non-Microsoft implementation out there at this point".   Most products implementing OOXML are like the "beaujolais nouveau" wines of software products, they are just coming out.   It will take some time before they become vintage wines.    These products will mature over time, they will improve based on the needs of their users and will eventually be as tasty as any other product.

  • Anonymous
    January 15, 2008
    The comment has been removed

  • Anonymous
    January 15, 2008
    The comment has been removed

  • Anonymous
    January 15, 2008
    @Cmdr Cmdr said "Am not sure that you can blame Microsoft for a lack of "serious and reliable non-Microsoft implementation out there at this point".  " What about Office 97 then? Hasn't Microsoft made all binary formats available (officeff@microsoft.com)? How come there is no known reliable way to use a competing product? Office 97 shipped in 1997 right? That's eleven years ago. Even if implementors are quite slow, I think they've had enough time. Well obviously, not having the migration mapping tables does not help. That's why it will be even harder to compete with something based on "OOXML" with Microsoft's reference implementation (which itself is a moving target). We know this : it's called fire and motion, and it has everything to do with making it impossible for the market to play on equal footing. This by the way describes what the Office group has been doing for two decades now. Now can you please answer this question : without the migration mapping tables, how do you intend to compete with Office 2007 ?

  • Anonymous
    January 15, 2008
    Jiří Kosek, the official collector of the Czech Republic's 75 OOXML objections, has posted his evaluation of the ECMA's responses to those objections: http://xmlguru.cz/2008/01/ecma-response-to-czech-ooxml-comments It's a very thorough evaluation, with each response scoring a "Satisfactory Resolved (GREEN)", "Partially Resolved (YELLOW)", and "Not Satisfactory Resolved (RED)". And the results?  I'll just quote him: "ECMA already provided proposed resolution for 75 comments (out of total 75 Czech comments). This means that 100.00% of Czech comments were handled by ECMA. 90.67% of comments were satisfactory resolved. 8.00% of comments were resolved only partially. 1.33% of comments were not satisfactory resolved. ... In fact I was really surprised how many “green boxes” are there at the end. I was expecting that ECMA will properly address only part of our comments. The vast majority of Czech comments was addressed by ECMA so it is time to say yes to OOXML." (I looked at the lone "Unsatisfactory Resolved" issue, and it's just a disagreement on the wording of the spec (that a portion is too verbose for Czech Republic's liking), not on technical issues, and Jiří is fine with letting that get resolved in the future.) Sorry, anti-OOXML fanboys, but the writing is on the wall.  It's time to face the facts.  All your tremendous effort to FUD and derail OOXML has done is to create an ISO OOXML 1.0 spec that is actually of 1.5 quality, complete, and polished (unlike ODF 1.0, which is woefully incomplete and not polished at all).  Just as Miguel predicted, I might add.

  • Anonymous
    January 15, 2008
    "Sorry, anti-OOXML fanboys, but the writing is on the wall." The arrogant in you probably does not understand that no one seriously following the OOXML saga believed that this would fail. And everyone, except you and a few other shills apparently, clearly saw that contributions such as IBM's Weir have made great service to Microsoft to improve the specs. This alone contradicts what you are saying. Let's go further. I think the entire issue left is that there isn't and probably will never be a second reference implementation that can ever compete with Microsoft's implementation for the simple reason that if you correct mistakes in the specs, you are not discussing all what is MISSING in said specs. And an example of that is the migration mapping tables, which in their absence ensures that competing products are excluded from doing what Office 2007 does. This isn't what standards are for. No matter how you see it, this is largely immoral to parade on OOXML's so-called "success" when you know that the specs that are provided guarantee exclusivity, therefore no interoperability. Sure, you will be able to unzip a file, and read/write angle brackets. But that's not what we are talking about. Last but not least, if there are important changes mades to the specs, then a logical conclusion is that there will be no reference implementation. What an accomplishment indeed... (Of course, I'm playing devil's advocate here, everyone knows that Microsoft was well on its way to remove VML in Office documents and that the reason we have discussed VML in this whole 2 years is only the consequence of Microsoft's own incompetence in shipping Office 2007 devoid of VML).

  • Anonymous
    January 15, 2008
    The comment has been removed

  • Anonymous
    January 15, 2008
    The comment has been removed

  • Anonymous
    January 15, 2008
    Una noticia muy interesante es que ECMA respondió a todos los comentarios que se hicieron en todo el

  • Anonymous
    January 15, 2008
    Una noticia muy interesante es que ECMA respondió a todos los comentarios que se hicieron en todo el

  • Anonymous
    January 15, 2008
    The comment has been removed

  • Anonymous
    January 16, 2008
    S, If that's the case, then do you think he will now support the format since we've fixed the vast majority of issues raised? -Brian

  • Anonymous
    January 16, 2008
    The comment has been removed

  • Anonymous
    January 16, 2008
    The comment has been removed

  • Anonymous
    January 16, 2008
    S, You didn't answer my question

  • Anonymous
    January 16, 2008
    Hoping that you will answer mine then (i.e. migration tables), I think Rob will continue to be against OOXML (February 2008 edition) because it :

  1. does not prove interoperability across applications and platforms for any substantial scenario. Where are those scenarios, if any?
  2. does not show evidence that a second standard is needed so far. OOXML (February edition) looks more like ODF now. For Microsoft, or anybody, to have a case, they need to come up with something way better than the current standard. Something so much better that it's a no brainer that it's going to advance the state of the art. You can ask him directly.
  • Anonymous
    January 16, 2008
    The comment has been removed

  • Anonymous
    January 16, 2008
    Len,
    Couldn't agree with you more. And I'm definitely looking forward to the snow :-) -------------------  S, So is your main gripe then with Open XML that you would like to also see the binary documentation (which is already available BTW), as well as a mapping that lets you take files in the binary formats and move them into the Open XML formats? Aren't those issues seperate from the standardization of Open XML? -Brian

  • Anonymous
    January 16, 2008
    S, Let me know if this helps: http://blogs.msdn.com/brian_jones/archive/2008/01/16/mapping-documents-in-the-binary-format-doc-xls-ppt-to-the-open-xml-format.aspx -Brian

  • Anonymous
    January 16, 2008
    The ECMA comments are here: http://www.yousendit.com/transfer.php?action=check_download&ufid=0E674E3F139C8C27&key=fc5831420f5bfeed9453494e4edb71c5e11f1c33

  • Anonymous
    January 16, 2008
    The comment has been removed

  • Anonymous
    January 17, 2008
    >>The ECMA comments are here: >Take a look and let me know what you guys think. having read the changes proposed, i must say: the same trend continues: rushing and rushing... you are making a mockery of the ISO fast-tracking process: changing on-the-fly the specification, without proper review and consensus and all this only to achieve "ISO status" for the format of a software product, leaving interoperability, implementability and standards reusing as a second-class citizen. IMHO, with all this corrections, Microsoft demostrated that this draft is in an unaceptable state for fast-tracking and need careful reviewing. Compare it with ODF and PDF fast-tracking: they didn't abused the system. But the game is clear here: just send some response to the 3500 comments, to conform NBs ... for example: substitue the word  OLE  by "linking technology" ... shameful

  • Anonymous
    January 17, 2008
    @marc These same NBs weren't bothered about the use of "OLE" in the ODF specs (ole-object-count)I notice. They probably never reviewed them at all. Rushing by doing a massive amount of review and refinement is better than rushing by just releasing with no-one reviewing it. Gareth

  • Anonymous
    January 17, 2008
    The comment has been removed

  • Anonymous
    January 17, 2008
    @marc Man hours = number of people x hours worked per person. I would still contend that more man hours were expended on OOXML than ODF. The value of a standard is surely not to be judged by who is involved in it's production and the speed of its gestation. Maybe you can just cut Microsoft a little slack for being new to this, rather than some other organizations that have been gaming the system with perhaps more finesse over the years. If you feel standards should be entirely free of commercial involvement, then who will drive them?  If they are not out-and-out commercial organizations, then who will pay these people? Show me an existing standard and follow the money.  I doubt you will find any "clean" ones. Gareth

  • Anonymous
    January 18, 2008
    Marc, Have you looked at the ODF committee's responses to the NB comments that came in during their review? They just completely blew off a number of very relevant comments. TC45 spent a huge amount of time sorting through each comment and pulling together what we feel is a very good set of responses. The difference between the two processes is night and day. -Brian