[odf-discuss] Gnome, Ecma, and what governments (and FOSS?) should have done

Alberto Barrionuevo abarrio at ffii.org
Sun Nov 4 14:18:10 EST 2007

On Saturday 03 November 2007 23:18:07 Jody Goldberg wrote:
> That is only partially true.  From my perspective ODF as a spec is
> not sufficient to implement OO.o compatible content.  ODF has
> several advantages over MOOX, but for me the most important
> differentiator is that ODF comes with OO.o.  The existence of a true
> free software implementation of ODF is the differentiator.

Jody is right on this.

I think that was Bill Joy who requested that one standard to be considered 
an "open standard" must have a BSD licensed implementation (or at least a 
FLOSS one).

Unfortunatelly I haven't found the reference to his actual quote...  :o(

Anyway, usually and officially is not requested that condition to consider an 
standard "open". A well developed open specification in a really open process 
should be enough. [1] But of course, when you have a free reference 
implementation (as in TCP/IP), there is no need to document as in detail the 
specifications. You can refer to the *free* implementation.


[1] And this is one of the main differences between ODF and OOXML. ODF has a 
free reference implementation, so when you have any doubt, you can check it. 
OOXML hasn't and besides it's specification is quite poor. No matter of this 
the OOXML specification is continuously refering to the MS-Office as 
reference implementation when you cannot check its code... It is a trick that 
Microsoft is using in the NBS when comparing the ODF specification to their 
OOXML one.

More information about the odf-discuss mailing list