Hi,
at the Microsoft presentation at the summit of newthinking I took a note to check the conditions under which the OData standard (the OASIS open data standard proposal, heavily industry influence) is licensed. Turns out it is the "Microsoft open specification promise" found here: http://www.microsoft.com/openspecifications/en/us/programs/osp/default.aspx
Was there already an analysis of how these terms align with our understanding of what an open standard is? And if not, is this something where we should communicate actively?
All the best,
Mirko.
On Thu, Dec 06, 2012 at 02:04:04PM +0100, Mirko Boehm wrote:
Hi,
at the Microsoft presentation at the summit of newthinking I took a note to check the conditions under which the OData standard (the OASIS open data standard proposal, heavily industry influence) is licensed. Turns out it is the "Microsoft open specification promise" found here: http://www.microsoft.com/openspecifications/en/us/programs/osp/default.aspx
Was there already an analysis of how these terms align with our understanding of what an open standard is? And if not, is this something where we should communicate actively?
The following text discusses, that this "promise" is not sufficient to rely on for Free Software: https://www.softwarefreedom.org/resources/2008/osp-gpl.html
Hi Mirko,
* Mirko Boehm mirko@fsfe.org [2012-12-06 14:04:04 +0100]:
at the Microsoft presentation at the summit of newthinking I took a note to check the conditions under which the OData standard (the OASIS open data standard proposal, heavily industry influence) is licensed. Turns out it is the "Microsoft open specification promise" found here: http://www.microsoft.com/openspecifications/en/us/programs/osp/default.aspx
Was there already an analysis of how these terms align with our understanding of what an open standard is? And if not, is this something where we should communicate actively?
Please also have a look at:
http://fsfe.org/activities/os/msooxml-interoperability.en.html
Example #3: Microsoft's Open Specification Promise is not reliable legal coverage for complete interoperability
MS-OOXML files generated by MS Office 2007 contain content that is implementation defined. This is a cause for concern because content not described in the proposed specification has an unclear status regarding coverage under the Microsoft Open Specification Promised (OSP). OSP coverage is limited to patents "that are necessary to implement only the required portions of the Covered Specification that are described in detail and not merely referenced in such Specification." 8
The OSP states in the final sentence of paragraph two that "No other rights except those expressly stated in this promise shall be deemed granted, waived or received by implication, exhaustion, estoppel, or otherwise". 9 It appears reasonable to not rely on the OSP for content necessary to allow interoperability that is not described in detail or referenced in the proposed specification.
This concern becomes more acute if the document is saved in other variations of the proposed specification format. For example, XLSM documents contain unspecified content as well as binary content. XLSB documents contain content stored using a method apparently not described in the proposed specification. XSLX documents with a password are also stored using a document container apparently not covered by the proposed specification.
Also some information http://fsfe.org/activities/os/msooxml-idiosyncrasies.en.html
Regards, Matthias
Am 11.12.2012 11:26, schrieb Matthias Kirschner: ...
Please also have a look at:
http://fsfe.org/activities/os/msooxml-interoperability.en.html
...
Also some information http://fsfe.org/activities/os/msooxml-idiosyncrasies.en.html
These are rather dated and technical. Ordinary users, even politicians engaged in liberty, social or green issues, usually don't care about anything except that it works. Is there an FSFE page or elsewhere, which concisely explains to ordinary users the disadvanges of OOXML (and especially DOCX, XLSX, and PPTX) and encourages them use, if not ODF, at least RTF and PDF when possible?
Should people be encouraged to use the old DOC, XLS and PPT formats for interoperability in spite of their inferiority?
My own experience is now that I can mostly open DOCX, XLSX and PPTX files in Libre Office, but there are often things missing, e.g. lines of text, columns of data, or complex XLSX documents take ages to load. Is this the "fault" of OOXML, MS-Office, Libre Office or all three?
Cheers, Theo Schmidt
* theo.schmidt@wilhelmtux.ch theo.schmidt@wilhelmtux.ch [2012-12-11 12:40:21 +0100]:
Am 11.12.2012 11:26, schrieb Matthias Kirschner: ...
Please also have a look at:
http://fsfe.org/activities/os/msooxml-interoperability.en.html
...
Also some information http://fsfe.org/activities/os/msooxml-idiosyncrasies.en.html
These are rather dated and technical. Ordinary users, even politicians engaged in liberty, social or green issues, usually don't care about anything except that it works. Is there an FSFE page or elsewhere, which concisely explains to ordinary users the disadvanges of OOXML (and especially DOCX, XLSX, and PPTX) and encourages them use, if not ODF, at least RTF and PDF when possible?
We explained some of the issues in previous Document Freedom Days. Perhaps some of the links in http://blogs.fsfe.org/greve/?p=152 help you.
Regards, Matthias
On 11 December 2012 11:40, theo.schmidt@wilhelmtux.ch wrote:
My own experience is now that I can mostly open DOCX, XLSX and PPTX files in Libre Office, but there are often things missing, e.g. lines of text, columns of data, or complex XLSX documents take ages to load. Is this the "fault" of OOXML, MS-Office, Libre Office or all three?
Slightly off-topic: whenever you get a recent MS Office document like this that doesn't work, please file a bug with LO - they really do work hard on their format converters and love to know about this stuff, e.g.:
http://fridrich.blogspot.co.uk/2012/11/libreoffice-coreldraw-import-filter.h...
I don't know for sure about Apache OpenOffice, but I'd be surprised if they weren't similarly interested in their filters.
- d.
Am Dienstag, 11. Dezember 2012 12:40:21 schrieb theo.schmidt@wilhelmtux.ch:
My own experience is now that I can mostly open DOCX, XLSX and PPTX files in Libre Office, but there are often things missing, e.g. lines of text, columns of data, or complex XLSX documents take ages to load. Is this the "fault" of OOXML, MS-Office, Libre Office or all three?
At first it is Microsoft's "fault", as they have choosen a dataformat that is hard to implement from technical and legal aspects. There is reason to believe that they did this delibertately as it is the usual game.
I secondly believe that users (paying or non-paying) customers, especially organisations, are responsible for using the format and thus giving it some real world weight.
If you accept both as given, you could try to come up with enough helping hand or funding for someone to do a good Free Software implementation.
Hi,
On Dec 11, 2012, at 11:26 , Matthias Kirschner mk@fsfe.org wrote:
Please also have a look at:
http://fsfe.org/activities/os/msooxml-interoperability.en.html
Example #3: Microsoft's Open Specification Promise is not reliable legal coverage for complete interoperability
In this regard, I am trying to understand risks involved with the oData "standard". I think the OOXML case is different in that now with oData, competiting implementations are developing from the beginning (the spec is available while Microsoft develops their own product, not after).
Do you think that makes a difference?
Cheers,
Mirko.
* Mirko Boehm mirko@fsfe.org [2012-12-11 23:01:08 +0100]:
Please also have a look at:
http://fsfe.org/activities/os/msooxml-interoperability.en.html
Example #3: Microsoft's Open Specification Promise is not reliable legal coverage for complete interoperability
In this regard, I am trying to understand risks involved with the oData "standard". I think the OOXML case is different in that now with oData, competiting implementations are developing from the beginning (the spec is available while Microsoft develops their own product, not after).
Do you think that makes a difference?
I am not sure at the moment. I would need to look more into the details again. What do others here think?
Regards, Matthias
On Tue, Dec 11, 2012 at 11:01:08PM +0100, Mirko Boehm wrote:
In this regard, I am trying to understand risks involved with the oData "standard". I think the OOXML case is different in that now with oData, competiting implementations are developing from the beginning (the spec is available while Microsoft develops their own product, not after).
Do you think that makes a difference?
Sorry, I cannot tell you more. I don't know much about it.