Ed Brill

Post a Comment

1. 1  Stephen http://notes2self.net | 3/19/2008 5:19:23 PM

   Ed maybe you should tell Doug Heintzman, "Director of Strategy for IBM Collaboration Technologies". He's quoted here

   { Link }

   IBM favors ODF as a file format because it is "truly open" and technically elegant, Heintzman said.

   But IBM will support Open XML, which is the current document format in Office 2007, in its Lotus collaboration and portal products. IBM already supports older versions of Office.

   Let's be honest, Lotus' OpenXML support seems to be at least as close to market as your iPhone stuff right ;-)

1. 2  Ed Brill http://www.edbrill.com | 3/19/2008 5:22:17 PM

   Stephen, good to hear from you, I thought you were busy with all the blogging yourself.

   Stephen, as is the classic Microsoft way, you have taken something out of context as a verbatim. The statement you cite is NOT a direct quote from my colleague Mr. Heintzman, it is what the reporter wrote. I suspect Doug said what I said -- that we will support Office 2007's file formats in certain products.

1. 3  Stephen http://notes2self.net | 3/19/2008 5:35:18 PM

   Cool, I think that's the right thing to do for customers and shareholders.

   So to take the text you have an issue with

   "For example, Novell, Corel, Apple, IBM, Sun, and others have already adopted (or announced adoption of) Open XML (OOXML) in their products on a variety of platforms -- including Linux, Windows, Mac OS, Palm OS, Java, and .NET."

   that would seem to be true if you yourself have said IBM "will support Office 2007's file formats in certain products" (I wasn't aware you'd said that too)

   Office 2007 implements Ecma376.

   Chris' letter confirms that Microsoft will implement any ISO version (That would be IS29500)in its products too.

   Also, Rick Jelliffe's responded to Rob's latest contribution here { Link }

1. 4  Ed Brill http://www.edbrill.com | 3/19/2008 5:43:23 PM

   Stephen, you and I both know that what is implemented in Office 2007 is NOT what was submitted to ISO, and that when you implement whatever comes out the other end of this process, it won't be the current document format that is in Office 2007. They are two different things, and one need only look at the challenges with getting OOXML right in Office 2008 Mac for evidence of those challenges in the MS product line.

1. 5  Stephen http://notes2self.net | 3/19/2008 6:07:52 PM

   >you and I both know that what is implemented in Office 2007 is NOT what was submitted to ISO

   So here we will have to agree to disagree Ed. Office 2007 underwent changes during development to stay up to date with Ecma376 as TC45 did its work. If you've found any differences you could always identify where there are - I assure you they will be treated as bugs.

   Is that your beef? Would you be happier with "IBM have announced adoption of Office 2007's file formats in some of their products"?

1. 6  Ed Brill http://www.edbrill.com | 3/19/2008 8:21:05 PM

   @5 Stephen, you know full well what I am talking about...articles like this one: { Link }

   The spec as submitted does not equal both the read and write format in Microsoft Office 2007.

   Saying that "IBM...have...adopted Open XML" as in Caposella's open letter is disingenuous. It's not the primary or default or anything other than a format some IBM products will support. We didn't "adopt" Oracle just because Lotus Connections can use an Oracle database.

   One of the main criteria for standards to be adopted at the ISO level is for there to be multiple examples of implementation as "prior art". When even Office 2007 isn't conforming to what's in those 7000 pages...how can anyone think that this is the right time and version to ratify?

1. 7  Nathan T. Freeman http://nathan.lotus911.com | 3/19/2008 9:01:47 PM

   @3 - "Office 2007 implements Ecma376."

   Excellent. Then we can take all deviations of Office2007's output against Ecma376 as the list of unreported errata in the Office product line! Good, that will finally give an objective measure of Microsoft's quality engineering output.

   Thanks for the heads up! Now we have a metric!

1. 8  hAl  | 3/20/2008 2:28:42 AM

   That is kind of amsuing. Ed refers to a post that states Office does not write OOOXML because it writes stuff that is not in the spec like macro's as well.

   That is amusing. Especially as seeing that OOo and Lotus write a lot of stuff that is not is the ODF spec as well. Invluding macro's but more important even including spreadsheet formula's. It is laughable that people describe OOXML as not fully documenting what Office is writing whilst ODF is not even documenting basic office functionality let alone related and /or embedded functionality.

   If anything is clear form that is should be that ODF is totally unasuable untill it is at least at the same level of completenesss as OOXML is in describing what is in the ODF office documents.

1. 9  Nathan T. Freeman http://nathan.lotus911.com | 3/20/2008 12:58:59 PM

   @8 - Of course, the topic isn't ODF. It's OOXML. Don't change the subject.

1. 10  Luc Bollen  | 3/20/2008 5:37:02 PM

   @Nathan, I'm afraid that the current tactic of hAl is to comment on each blog about OOXML and MS by talking about ODF and IBM.

   @hAl: Is this because you have nothing sensible to say to defend OOXML based on its own merits ?

1. 11  Hassan  | 3/20/2008 5:53:27 PM

   @Luc & Nathan

   No, it's just to show the hypocrisy of anti-OOXML people who turn a blind eye to elements in ODF that are identical to the ones they complain about in OOXML.

1. 12  Mike Brown  | 3/20/2008 6:35:45 PM

   From the Microsoft Handbook: How To Discuss OOXML:

   "Talk about ODF. Talk about IBM. Do *not* under any circumstances discuss the OOXML spec itself. Severe penalties will apply to anybody that breaks this rule.."

   Cheers,

   - Mike

1. 13  Stephen http://notes2self.net | 3/20/2008 7:38:12 PM

   @6 Ed ... so let me see if I've got this straight:-

   i)You contend, based on the linked article in "Fanatic Attack", that Office 2007 doesn't implement Ecma 376.

   ii) Consequently you think that when you personally say IBM/Lotus "will support Office 2007's file formats in certain products" this is not the same as saying that IBM/Lotus "will support OpenXML in certain products"

   iii) You blame Microsoft when people fail to make this distinction.

   It seems clear that the crux of the matter is whether i) is true or not. I've stated in plain enough language that it isn't. I'd suggest that that's the consensus opinion.

   If you don't want people to equate "Office 2007 file formats" and OpenXML you're entirely free to point out where they differ, and I've explained that Microsoft will treat any such instances as bugs.

   Then you say "We didn't "adopt" Oracle just because Lotus Connections can use an Oracle database."

   In a sense you did, "Adopt: choose and decide to use something: to take up something such as a plan, idea, cause, or practice and use or follow it - decided to adopt a wait-and-see policy { Link }

   Then you say "One of the main criteria for standards to be adopted at the ISO level is for there to be multiple examples of implementation as "prior art"."

   This is just wrong. As a case in point there were no implementations of IS26300 when it became a standard. IETF require multiple implementations, but IETF also requires wide adoption and there are many examples of competing ways of doing things (SPKI/SDSI/PKIX for example)

   Anyway, I'm pleased to hear you confirm there will be "Office 2007 support in some products". My view has always been that the world's better when IBM competes { Link } and worse when it throws in the towel { Link } .

1. 14  Stephen http://notes2self.net | 3/20/2008 7:40:48 PM

   @7 Nathan, you're welcome. I suspect you'd already worked that out for yourself though.

   No-one in Microsoft is disputing that ECMA376 changes things, and that if DIS29500 is approved things will change even more.

1. 15  Stephen http://notes2self.net | 3/20/2008 7:42:27 PM

   @12, Mike - so talk about it then. Do you have an opinion or do you just chirp?

1. 16  Luc Bollen  | 3/20/2008 8:02:21 PM

   @Stephen: as I see that you are more keen to comment on Ed's blog than on yours, I will repeat here questions I made on your blog on 18-Mar-08, and still awaiting a reply.

   (@Ed: Sorry to bring this here.)

   - "we are committed to supporting the Open XML specification that is approved by ISO/IEC in our products." Does that means that in the next version of MS Office, there will be a "Save as ISO 29500" option (i.e. with strict conformance, and without extensions like DRM, macros, SharePoint tags...) ?

   - While fighting FUD : what about an update of the OSP asserting that it will remain applicable to any version of ISO/IEC 29500 published by ISO, including for products developed under GPL V2 & V3 ? This would be a wonderful way of stopping what you call FUD on these subjects.

   See also { Link }

1. 17  Stephen http://notes2self.net | 3/20/2008 8:22:11 PM

   @16 Luc, ouch - I've answered now :-)

1. 18  Ed Brill http://www.edbrill.com | 3/20/2008 8:29:45 PM

   @16 cheers Luc, I poked Mr. McGibbon for the very point of spending time here instead of answering there. Whatever it took to get his attention.

   Stephen, so I can now claim that Microsoft "adopted" Lotus Domino when you created your own version of an Outlook connector to the Domino server? Cool!

1. 19  Stephen http://notes2self.net | 3/20/2008 8:57:22 PM

   I wondered what your comment yesterday meant!

   Well you could certainly say that the "Microsoft Office Outlook Connector for IBM Lotus Domino" adopted NRPC (assuming that's what it used?)

1. 20  Luc Bollen  | 3/20/2008 9:03:14 PM

   @Ed: sure, IBM adopted and supports OOXML, and Microsoft adopted and supports Lotus Domino and ODF and IBM Symphony (they fund converters using the same file format).

   And the Unix 'cp' command adopted and supports OOXML (it is both a conformant consumer and producer.) And any company selling or distributing Unix or Linux adopted and supports the 'cp' command and OOXML.

   Thanks to Microsoft, welcome to a brave new world knowing only love, peace and support of everything by everybody.

1. 21  Christopher  | 3/21/2008 3:01:56 AM

   It seems that most of the complaints here are about versioning. Microsoft released Office 2007 with the spec as it existed at the time (minus any errata which anyone is encouraged to submit) and changes have been made to address comments during the stardardization process. Further changes will be made in the future that will make the Office 2007 implementation not fully compliant with the updated spec. How is this any different from any other standard? If anything, updatability makes it a better standard.

1. 22  Nathan T. Freeman http://nathan.lotus911.com | 3/21/2008 5:16:15 AM

   @14 - I had not, actually. I'd never seen any cases in which Microsoft made public statements about the specifications to which their product(s) would adhere. But I believe you are being quite clear in your statement that if and when Office 2007 produces output in the OOXML format, any fashion in which that output does not match the ECMA376 specification is due to an unintentional bug, and is considered by Microsoft to be the support responsibility of Microsoft.

   This is an unusually clear statement on support plans from your organization. So unusual, in fact, I wonder whether you are in a sufficient position of authority to make it.

   @11 - Hypocrisy? How so? Let's pretend for a moment that I agreed that ODF has a laundry list of inadequacies as an ISO specification. What would be non-hypocritical course of action? To suggest improvements to it? To campaign to have it DEcertified? Or to say "aw f%&k it... might as well pass any old crap through ISO now?"

   If I were someone that had a problem with the ODF specification, the course of action which would display consistency would be to fight against OOXML EVEN MORE, because we shouldn't let ANOTHER bad spec into ISO.

   Somehow you seem to be suggesting that a person of integrity would say "two wrongs make a right, so let's pass that sucker!"

   This is all hypothetical, of course, because the comparison is merely an attempt to distract attention from OOXML itself. And it's not even a subtle attempt. But you're welcome to keep pitching the idea that a bad spec is okay now because bad specs have been okay before. That's certainly a more honest proposal than claim OOXML is, y'know, a decent piece of engineering.

1. 23  Nathan T. Freeman http://nathan.lotus911.com | 3/21/2008 9:01:40 PM

   Bored or defeated?

   Well, whatever... thanks for letting me have the last word, gentlemen.

1. 24  Ian Scott  | 3/22/2008 7:45:52 AM

   @Nathan - I was going to chime in simply to concur entirely with you although I was going to take it a little step further and ask two very specific questions as follows:

   @Stephen - in the very clear statement you gave in respect of Microsoft taking on the support responsibility for any OOXML output produced in any fashion by Office 2007 that does not match the ECMA376 specification were/are you providing an official statement on behalf of Microsoft - one that Microsoft can be henceforth held to - or were you offering a personal opinion?

   @Stephen - If you were making an official statement on behalf of Microsoft can you clarify if this support will extend beyond Office 2007 - ie to future releases of Microsoft Office?

1. 25  Douglas Heintzman  | 3/22/2008 9:36:31 AM

   Unfortunately I don't think that the debate about what supports what is terribly productive. If we have to I suppose we could talk about support with a lower-case “s” and Support with an upper-case “S”. Microsoft is certainly grossly abusing the facts, and manipulating the debate when it argues that IBM and other companies are Supporting their file format initiative by “supporting” it in various products. Just because you can put an OOXML file into one of our databases or content management systems doesn't mean that we are “Supporting” their format. That is like saying that Microsoft is “Supporting” ODF because I can save all my ODF documents to their file system in Windows on my hard drive. The fact that Microsoft is publicly arguing that there is broad industry “Support” and momentum for their specification using the examples that it does is a gross abuse of the truth.

   Let's be clear. IBM believes that the monopolistic domination of the office productivity market has inhibited investment and innovation. We believe that XML based file formats have tremendous potential to improve workflow, activities, and team collaboration. We believe that a modern XML based file format will allow customers to perform business analytics, generate automated semantic information, improved searchability, and make the information contained in documents much more socially and process aware and increase its overall value to the customer. We think that documents will become much more seamlessly integrated into business applications and processes.

   At the end of the day we don't care what specification is used or what it is called as long as it has some basic characteristics. It needs to be technically elegant (which includes simplicity of design) and pragmatically implementable (which includes the pragmatic use of existing well defined standards where appropriate). It needs to be truly open and managed in an open, and by extension predictable, way, not controlled by any one company. Ideally it needs to be unencumbered by by the intellectual property of any company. We have communicated this to Microsoft repeatedly for many years and they have been invited repeatedly to participate in the industry's collective efforts to build such a specification.

   It is very true that OOXML and ODF have slightly different design points. ODF is a forward looking specification. It endeavors to do as good a job as possible in providing high fidelity access to legacy documents stored in proprietary formats. It endeavors to do an excellent job of being forward looking and having the feature set and pragmatically implemental design to unlock innovation and allow developers and users to leverage document content in ways that were not previously possible. We have a major strategy project underway and have many very promising research projects that are showing tremendous potential. Many other companies are doing the same. OOXML in contrast was designed to provide excellent reverse compatibility with legacy documents but makes many compromises in design to achieve this objective which compromises its usefulness in doing many of the new things we believe are possible. The ODF TC at OASIS has been grappling with these tradeoffs for years.

   It is no wonder that Microsoft on the one hand, and IBM and Google and Oracle and Sun and many others on the other hand have chosen to pursue different strategies. Microsoft wants to preserve its monopoly and its massive market insertion power. I certainly don't fault them. But the interests of monopolies are always in tension with the interests of their customers and the market in large, both in terms of reduced freedom of action (lock-in) and the associated monopolistic pricing, and in terms of the potential innovation and improvement in capability that has been forgone.

   Of course I recognize that some people just want to write a letter or build a simple presentation, and they don't care what the underlying format is, and they may very well be happy to pay Microsoft's asking price for what I happily admit is excellent software. I also recognize that as a whole we are all paying an awful lot of money for a very mature technology in a market space that, in large, hasn't seen much innovation for a very long time (the talking paper clip notwithstanding). I believe that we are on the cusp of a “Beyond Office” era with much more socially aware and semantically rich information that organizations and individuals can leverage to create value in unprecedented ways. We believe that we need a specification that is technically elegant, pragmatically implemental, and truly open.

1. 26  Vid http://www.svaksha.com | 3/22/2008 3:43:29 PM

   Just wanted to say Thankyou for not supporting non-free formats.