Talk:OpenDocument

Sun's Patent Pledge

The recent addition of Sun's patent pledge as "criticism" seems strange to me, especially as the person who instigated the pledge in the first place. The pledge can only apply to things Sun has been involved in reviewing, so it is no surprise that Sun should state that Sun would not extend the pledge beyond the boundary of its involvement. Since future versions could be expected to build on earlier versions, and since earlier versions would continue to be covered by the pledge, this essentially means that Sun doesn't agree to hold blameless inclusion of new stuff in versions it's not involved in. It does not mean that in the case of Sun's non-involvement all protection is withdrawn, as the anonymous author seems to imply. Does anyone have any evidence to support the assertion this is a widely-held criticism - there is no attribution in the text. Webmink 12:54, 3 March 2007 (UTC)

No, it's not a troll. Although, if I were a troll, I'd say that, so I'm not sure that adds anything. I put it in becasue I had seen the criticism here: http://ooxmlhoaxes.blogspot.com/ - see the section titled "OOXML hoax 2: The standard is not really open". The issues raised there may be open to criticism, but I thought it worthwhile including. What's the rebuttal? WLD^talk|edits 13:07, 3 March 2007 (UTC)

The rebuttal is as above. The covenant is not terminated if Sun leaves the OASIS committee, but the covenant stops being extended to new additions. I'd be happy to re-word it but I gather Wikipedia doesn't like input from primary sources. Webmink 22:31, 3 March 2007 (UTC)

The key point you make is "The covenant is not terminated if Sun leaves the OASIS committee, but the covenant stops being extended to new additions." - perhaps I'm just being dense, but I don't see that in Sun's Patent Statement. It says that Sun "will not seek to enforce any of its enforceable U.S. or foreign patents against any implementation of the Open Document Format for Office Applications (OpenDocument) v1.0 Specification". That says nothing about later specs. Perhaps I'm just being over-suspicious, but it seems to me that Sun could enforce patents that it doesn't enforce against v1.0 against a later spec - e.g v3.17 or whatever. I'd prefer your interpretation, but my natural suspicion makes me think otherwise. I'm sure the wording has been subject to very careful legal approval. WLD^talk|edits 22:59, 3 March 2007 (UTC)

That's the problem with legal documents - when discussed by non-specialists (I am bravely assuming that category includes you) it's easy to make assumptions that are wrong. We have each made interpretations that lead to different conclusions, and prima facie they are equally likely. The only thing that makes mine more likely than yours is that I was involved in the publication of the covenant and I asked the lawyers who wrote it exactly this question at the time of publication. Whatever you feel about this, however, I assert that the existing text is not NPOV and needs modification. Webmink 01:46, 4 March 2007 (UTC)

Well, to use that well-worn acronym, IANAL, so you are safe there. How best could the text be improved to reflect our two prima facie conclusions? Note that I'd be happy to link to something that explicitly supports your view, just as a link to "The Wraith"'s web log is at present. I'm not particularly happy at linking to a 'blog, but I've not found a better exposition of the position. As far as I'm concerned, I think as much valid criticism of ODF as possible should be included in the article as that can only help the cause of clarity. It's a bit of a Nietzschien approach - that which does not destroy ODF makes it stronger! WLD^talk|edits 13:45, 4 March 2007 (UTC)

This issue was already worded in the last sentence of the licensing section. I'm not sure it warrants a seperate place in the critisisms. I shows that ODF future version are as much Sun dependant as futere OOXML version are MS dependant. It just shows that openness always has it's limits. hAl 08:50, 5 March 2007 (UTC)

(Unindenting) The more I think about it, the more I think it should be there. OOXML is criticised by many for not being 'open' and 'freely implementable' when the same criticisms can (and should) be levied against ODF. If Sun gets into severe business difficulties (and there are many precedents for large technology companies getting into business difficulties), then non-participation in producing a new version of ODF standards allows them to start using their intellectual property portfolio as a revenue generator - much like Unisys did with GIF and the Lempel-Ziv-Welch compression algorithm. A clear expression from Sun saying that they would not seek to enforce intellectual property rights on any 'technology' used in any ODF version (developed with their participation) in any future version (not developed with their participation) would help. For example, if ODF 1.0 used method 'foobar' patented by Sun, the current wording means that Sun can enforce rights in relation to 'foobar' in ODF 3.7 if Sun have not participated in 3.7's development. All they simply need to say is that any technology used in a version developed with the aid of Sun can be freely used in a later version not developed with the aid of Sun, but any new technologies in later versions not developed by Sun are susceptible to litigation. I think open, freely implementable standards are a good thing, but it looks like ODF may, possibly, not be as open as some people believe. I could be being overly cautious here. I wish it were possible to uninvent software patents and business method patents. WLD^talk|edits 10:23, 5 March 2007 (UTC)

And I suppose that the Wikipedia article on the U.S. Constitution should have a criticism stating that there is nothing in the text of the Constitution that prevents a future Constitutional Convention from reintroducing slavery? This seems to be the logic here. There is a standard called ODF now and today, and this criticism does not seem to be about anything that is actually in ODF. You might as well complain that there is nothing that prevents a future version of ODF from forbidding the letter 'e' in text or font point sizes greater than 20, or dotted lines in presentations. In fact, maybe a future version of ODF will be turned into a recipe for cherry cheesecake. There are an infinite number of possible criticisms of unwritten future versions of ODF, and none of these criticisms can be refuted. But such unfounded speculations seem to belong more to a horoscope than an encyclopedia. I'd like to think that criticisms are based on some minimal threshold of plausibility, based on more than an unsubstantiated anonymous blog postings. RCWeir 04:14, 21 March 2007 (UTC)

I think your view of the logic is skewed. It is quite simple: all Sun needs to say is that any of Sun's technology used in a version developed with the aid of Sun can be freely used in any later version not developed with the aid of Sun. That is not what Sun have said. It is quite interesting from a legal point of view, as the statement, in its current form, seems to work against a defence of laches being used by someone should Sun prosecute. Note: I am not a lawyer, and this is not legal advice - it's just an area of interest for me. It is anough of an issue for Webmink to have taken this up with Sun's lawyers. Webmink was happy with the reply, but is unable to add it (it's probably not verifiable in the Wikipedia sense, anyway) - so I would content that it is a substantive issue. You have to look at what Sun have officially said, and a statement written with the explicit involvement of lawyers will have been very carefully crafted, as lawyers are experts in precise language. What commentators say they believe about Sun's intentions when the statement is made doesn't really help - you have to look at the actual text and see what meanings are compatible with what is written. Of course any future version of ODF could be turned into anything, but that is not the point, and irrellevant to the point I am making - that (a) Sun are unlikely to participate in all future versions of ODF and (b) any future version definied without the help of Sun cannot use Sun's technology, even if it has already been used in a previous version (dependant on lifetime limits on patents/copyright or other intellectual property used). It's the even if it has already been used in a previous version that is the kicker. I hope that is clear. Regards, WLD^talk|edits 08:29, 21 March 2007 (UTC)

I think it is relevant because ODF is mentioned as being an alternative for OOXML for even Microsoft. How could Microsoft commit to using a format where possible control of the future versions development is in the hands of a competitor with it's own Office suite. The same of course applies to OOXML where MS controls future development using it's intellectual property. This argument show that there can be very valid reasons for more than one format. hAl 12:40, 21 March 2007 (UTC)

That sounds a lot like original research. That doesn't belong here. Analysis in an anonymous blog posting of unknown expertise can hardly be considered reliable, especially when we have on-the-record public expert statements that say the ODF license is fine, for example: this review by the SFLC. So, on one hand, you have the General Counsel of the Free Software Foundation, and Professor of Law at Columbia, saying that the license is fine (interesting that the Wikipedia article doesn't mention this), and on the other hand you have an anonymous blogger called The Wraith who thinks there are problems. And which statement do you go with? The anonymous blogger, of course. But my understanding of Wikipedia policy is that this criticism should attributable to a reliable source, and I don't see that here. RCWeir 15:57, 21 March 2007 (UTC)

Look, I'm sorry, and perhaps I'm being particularly dense but having read the SFLC review, I don't believe it supports you to the extent that you say. The Licensing section of the article says "Key contributor Sun Microsystems made an irrevocable intellectual property covenant, providing all implementers with the guarantee that SUN will not seek to enforce any of its enforceable U.S. or foreign patents against any implementation of the OpenDocument specification." (My emphasis), but I can't see that being said in either

• the Patent Statement [1], or
• the SFLC review [2], or
• the OASIS intellectual property rights policy [3] referenced in the SFLC review.

Sun's patent statement specifically references version 1.0 of the specification, "or [...] any subsequent version thereof [...] in which development Sun participates to the point of incurring an obligation [...]", and the OASIS intellectual property rights policy does not force licensing of subsequent versions - the "Continuing Licensing Obligation" applies only to the specification version developed with the aid of the contributing member. That is, if Sun withdraw from OASIS now, Version 1.0 of ODF will continue to be covered by Sun's Patent Statement. Nowhere is there stated to be an obligation to license patents used in v1.0 in subsequent versions. Read it yourself. I would love to be wrong. The SFLC review references back to Sun releasing the StarOffice source under the LGPL and says "The LGPL required Sun to license any patents covering ODF in a manner “consistent with the full freedom of use specified in [the LGPL].”". I'm not conversant enough with the LGPL to say whether or not patent licensing can be version dependant - i.e. Allowing Version 1.0 of foobar licensed under the LGPL to use patented software means that all subsequent versions of foobar licensed under the GPL can also use the patented software. It is by no means clear. And if you point out that OASIS requires a perpetual licence, that's actually not relevant - it simply means that V1.0 of ODF is licensed in perpetuity, not that all subsequent versions of ODF are. Why did Sun specifically reference only ODF 1.0 when it would have been easy (and very clear) to add "and all subsequent versions."? I'm not a lawyer, and I can't afford to get one to conduct a review on this specific topic. It is, however, crucially important to businesses looking to use ODF in the long term. Why move from Microsoft controlled formats to what may potentially be Sun controlled formats? ODF 1.0 is safe, that much is clear, which is far, far better than nothing. I've said my piece, and if you have done, thank-you for reading. I really don't want to be labelled as a nut-job who is capable only of playing a one-note samba on this topic - I've got other things to do, and as RCWeir points out, this may well be construable as Original Research, so I'll just toddle off into the sunset. Cheerio. WLD^talk|edits 20:17, 21 March 2007 (UTC)

It is odd since I also haven't seen this anywhere else except on Wikipedia (paging Reliable Sources needed on this). Why would Sun need to donate unknown patents to an unknown process if they're not involved in it; is that unreasonable? Should their patents on unrelated things be a free for all? The wording of "If Sun does not participate, then the assurance not to seek to enforce any of its enforceable U.S. or foreign patents against any implementation will not apply" makes it seem like a very valid threat, which contradicts the article itself! E.g., earlier, "Key contributor Sun Microsystems made an irrevocable intellectual property covenant, providing all implementers with the guarantee that Sun will not seek to enforce any of its enforceable U.S. or foreign patents against any implementation of the OpenDocument specification. This Statement is not an assurance that an OpenDocument Implementation would not infringe patents or other intellectual property rights of any third party." I don't see how this is malicious as the criticism seems to be.AnyPhish02 08:13, 4 July 2007 (UTC)

Inconsistent spelling of formulæ

yup —The preceding unsigned comment was added by 69.248.209.139 (talk) 03:25, 13 March 2007 (UTC).

On Portal:Free software, OpenDocument is currently the selected article

(2007-05-20) Just to let you know. The purpose of selecting an article is both to point readers to the article and to highlight it to potential contributors. It will remain on the portal for a week or so. The previous selected article was Free Java implementations. Gronky 12:29, 20 May 2007 (UTC)

The selected article box has been updated again, OpenDocuments has been superceded by ZFS. Gronky 12:23, 28 May 2007 (UTC)

Standardization, again

Hello,

it has been questioned whether OpenDocument Standardization is notable. I think it undoubtably is, and have removed the "importance" tag. However, it might be good to once again verify OpenDocument Standardization against OpenDocument and see whether they are in sync, and whether some content is currently duplicated. The OpenDocument article is so long that it is certainly OK to retain a separate article on standardization; but maybe the content can be condensed somewhat and then integrated into OpenDocument. This should be decided by the experts. --B. Wolterding 08:31, 22 May 2007 (UTC)

This article isn't really that long, it just appears that way because of the inevitable hijacking which occurs whenever an article is permitted to have a Criticism section. There's no reason the two can't be folded into each other, neither is particularly high-quality right now. For now I've collapsed the subheaders on the offending section, which makes this article a little more hospitable to a merge. Chris Cunningham 09:31, 22 May 2007 (UTC)

Removing the subheaders probably was a good idea allthough the readability has suffered a lot now. Mayby bulleting simular to the OOXML article critisism section would be better. The whole standardization section is overly long anyways without adding much information that is really usefull. Spending a whole article on that seems just plain boring but I can't be bothered to look up that article to check hAl 11:04, 22 May 2007 (UTC)

Roadmap

Allthough the article suggest that ODF completing the standard in version 1.2 in an accelerated schedule there seems little evidence of that actually happening. At the current rate a final 1.2 version more likely at the end of the year or start of next year than in an accelerated schedule which would put it wel before oktober. So it even seems unsure whether version 1.2 will in the ISO version during 2008 even, as atm the 1.1 version isn't even an ISO version yet and the 1.2 version lagging more than a year behind that one. Is there any more up to date roadmap to put in the article ? hAl 08:54, 26 May 2007 (UTC)

Good point. I'd noticed as well - I've not found an updated scheduled anywhere. 195.10.54.12 13:09, 8 June 2007 (UTC)

Why aren't Abiword, Gnumeric, Google Docs & Spreadsheets, etc mentioned?

All of them support ODF, yet only KOffice, OO.org/StarOffice are mentioned. According to this [[4]], several other word processors also support ODF. GNOME office is a particularly prominent app.AnyPhish02 08:13, 4 July 2007 (UTC)

Lists of applications which use OpenDocument Format have been hived off to a separate article: OpenDocument software. WLD^talk|edits 13:40, 28 July 2007 (UTC)

Abiword and Gnumeric only implement import- and export-filters while OpenDocument is not there native format. Also abiword, gnumeric and google didn't participate last few years in the Oasis TC like KOffice did to create and extend the specs. —Preceding unsigned comment added by 212.202.0.85 (talk) 00:28, 4 September 2007 (UTC)

New sections

Could we get a section on the history of the format. It seems similar to OpenOffice.org 1.0 files and it'd be good to see whether those in turn were the file format from StarDivision? —The preceding unsigned comment was added by 202.160.118.227 (talk) 11:53, August 22, 2007 (UTC)

Math

This criticism is listed in the article:

• Some mathematicians do not think that the choice of the MathML W3C standard for use in OpenDocument is a good choice. MathML is a W3C recommendation for the "inclusion of mathematical expressions in Web pages" and "machine to machine communication" that has been around since about 1999. However, most mathematicians continue to use the much older TeX format as their main method for typesetting complex mathematical formulae. TeX is not an ISO standard, but is fully documented and is the de facto standard for typesetting mathematical expressions. OpenDocument is also criticized for not using the ISO 12083:1994 standard for mathematical formulae, which is not used within MathML either. MathML has a few issues with representing mathematical formulae correctly compared to other methods like TeX.

I don't understand this. MathML and TeX serve different purposes as far as I understand (I'm a long-time LaTeX user, and fairly green on ODF or MathML). TeX is a convenient input format and an excellent typesetting algorithm, but it can be a bit ambiguous if you'd ever want to use a computer to automatically evaluate an expression. For example, does the letter i represent sqrt(-1) or just any integer number? The MathML standard allows a human-writable "annotation" element, e.g. this is how OpenOffice stores a simple equation in an ODF document:

<math>...</math>
  <math:semantics> 
     ... lots of MathML gibberish ...
     <math:annotation  math:encoding="StarMath 5.0">( a + b ) over (c^2 + d)</math:annotation>
  </math:semantics>
</math>...<//math>

Indeed, the MathML spec gives an example[5] with

<math:annotation math:encoding="TeX">...</math>

So the real question is whether ODF allows TeX encoding rather than StarMath-5.0 encoding. If the answer is yes, then this criticism should be deleted (especially since it is unreferenced). If the answer is no, then this criticism should be rewritten. According to the odf spec[6], ODF simply incorporates MathML 2.0 without further restrictions, so I think the answer is yes. Han-Kwang (t) 09:21, 1 September 2007 (UTC)

I was going to say kind of the same. It's pretty clear to me that MathML and TeX serve pretty different purposes. TeX was created to be typed directly, and only to typeset math. There is no way to know if some text was typeset in roman because it's an operator, or because it's text. MathML allows different elements for this different purposes: <mo> and <mtext>. Also, in the MathML faq, it's pretty clear that MathML and TeX serve different purposes (see syntax and technical issues). 80.32.129.34 22:10, 3 September 2007 (UTC)

www.opendocumentfellowship.org ?

I am unable to access opendocumentfellowship.org. Is this just me/now, or is the site no longer accessible? (If the latter, then the links should be removed/updated) --RealGrouchy 04:01, 2 September 2007 (UTC)

JUst checked - it is accessible now. I had problems accessing it yesterday too. WLD^talk|edits 22:17, 3 September 2007 (UTC)

XML created and developed by Open Office?

The second paragraph of the article makes a claim that OpenOffice.org created and implemented the XML file format, yet I have found no evidence to support this claim. —Preceding unsigned comment added by 74.74.239.101 (talk) 06:08, 6 September 2007 (UTC)

historical names

Before the OpenDocument format was called "OpenDocument," it was called (formally or informally):

• "StarOffice XML [File Format]" or "Star Office XML" (October 14, 2000) [7]
• "Open Office XML" (till January 30, 2005, this article was named as such; see Open_Office_XML history)
• "OpenOffice.org XML [File Format]" [8] (December 2002)

There is presently quite a bit of confusion about the names "Open Office XML" and "Office Open XML" in the mainstream press (Google News: "open office xml") and blogosphere (Google Blog Search), substantially diluting the "Open Office" brand. I think it would be instructive for this article to treat this history, and for Office Open XML to note in its "distinguish" tag that "Open Office XML" is not the same as "Office Open XML." I'm only proposing this change here, instead of "being bold" about it, because User:HAl feels strongly (see Talk:Office Open XML#"distinguish" template, redux) that such a contribution to this article would be unwelcome — he believes the OpenDocument community is intentionally closeting its heritage (correct me if I'm wrong, Hal). —Fleminra 22:29, 9 September 2007 (UTC)

Support from Microsoft Office 2007

This is a quote from the <http://en.wikipedia.org/wiki/Microsoft_Office_2007> article, which claims that:

"Microsoft backs an open-source effort to support OpenDocument in Office 2007, as well as earlier versions, and also through a converter add-in for other programs, which works by having third-party programs call a command-line utility. As of yet, the project only supports conversions between OpenDocument Text and Office Open XML Word documents."

This is cited to [9] —Preceding unsigned comment added by 74.94.157.145 (talk) 17:04, 2 October 2007 (UTC)

Analysis of OpenDocument article

hAl reports the following text that comes from a dense document. I believe it was taken out of context and needs more work. The reporting by hAl is evidently biased.

The president of the OpenDocument foundation stated that Opendocument does not adress the three 'big problems' facing any transition of documents applications and processes in Microsoft formats to XML whereas OOXML, being created by Microsoft, is inteded to do just that..

Those problems are:

• Compatibility with existing documents-file formats: including the volumes of MS binary documents.

• Interoperability with existing applications: including the over 500 million MSOffice-bound workgroups.

• Convergence of desktop, server, device, and web systems as fluid and highly interoperable routers of documents, data, and media. Also know as "Grand Convergence.

He blames it on the “two ODF groups”: one pro-OOXML and the other pro-ODF.

Simosx 18:22, 6 October 2007 (UTC)

I have reverted the readdition of this piece by hAl, because I thought it was not less, but maybe even more biased than it'd been before.

Then it was re-added by a newly-registered user Jimb1234, who might be a sockpuppet: while the first edit is correct, this other one is suspicious (and the comment doesn't look like a new user's comment). I've reverted it, but I have to admit that the user has made it a little less biased. Only a little, though, so I believe it should still be discussed first.

--AVRS 11:14, 7 October 2007 (UTC)

I do not understand. This is very newsworthy article with info coming from the Opendocument foundation and the text is a near copy from a relvant part of the article. How is a text copied from an article biased? Or is the Opendocument foundation an unreliable source? Jimb 1234 12:20, 7 October 2007 (UTC)

It's not about reliability of the source. The info is notable, but it has to be worded in a way that does not change the meaning. Let's see what Simosx says. --AVRS 12:36, 7 October 2007 (UTC)

As I mention above, the referenced text is dense, and not similar to other text available. Taking a segment out of it and pasting it here is a poor choice. Not only that, I also see bias in the addition. The original text says that OpenDocument was not designed to address those issues. The addition here changes the text to OpenDocument does not address. Simosx 12:46, 7 October 2007 (UTC)

@Jimb 1234: You are a new user to Wikipedia, you dive in a complex issue, you focus only on document formats, and you tend to make a familiar type of typos (relvant instead of relevant, nname instead of name). Have you been here before? Simosx 12:46, 7 October 2007 (UTC)