<orcmid comments="below" /> -----Original Message----- From: Pedro [mailto:[email protected]] Sent: Monday, October 03, 2011 20:00 To: [email protected] Subject: [libreoffice-users] Re: Assessing ODF Conformance (Re: OASIS Standard ODF 1.2 Approved)
Hi Dennis It's interesting that you find out about this now... When I did some tests back in May 2011 and I questioned if TDF shouldn't worry about this and have its own tool, nobody found it important... http://nabble.documentfoundation.org/Odt-size-difference-between-MS-Office-11-and-LO-3-3-2-tt2978634.html#a2985444 I guess too much information is lost in these mailing lists... <orcmid> hi Pedro, Concerning the weird ODF Validator results you noticed in May, I can add a little of what I had encountered independently. I agree that it is difficult to find like-minded people who think these are important to reconcile. - Dennis BACKGROUND I was a newcomer to this list in June, I think. Mining older list posts doesn't seem to be in my DNA. I think I knew about the <manifest:manifest> manifest:version disconnect before that though. I shall have to look. I know it came up in discussions on some list that I was on before here. I don't think OASIS needs to do an ODF validator, there is at least one already and it is more about making a concerted effort to verify what they do and don't detect and keep improving them. It needs to be a publicly built and supported tool, that is exercised more by folks looking for ODF discrepancies in files from products or in the validator itself. It takes a community of practical people, not a standards body, in my experience. I think you were seeing a situation where the validator was updated to final ODF 1.2 and so was OOo-3.4 and no one was worrying about the down-level compatibility issues -- not the ODF TC either. My approach to this is to propose an OIC TC Advisory that recommends ignoring the rigidity of the manifest:version requirement, advise validators to be forgiving (maybe warnings instead of errors about it), and advise consumers to work with and without it in incoming documents. I think it is a practical matter and a pragmatic solution matters. I stumbled on this in an unexpected way on a document forensic analysis on a completely different issue when I noticed that the addition of <manifest:manifest> version broke the strict conformance to syntax that is implemented in what Microsoft Office 2007 and 2010 accept as well-formed ODF 1.0/1.1. (Office will "repair" the file to ODF 1.1 compliance and continue, but it scares users no end to be told the input file is corrupt.) Apparently older versions of OpenOffice.org and LibreOffice ignore manifest attributes they weren't built to support, but the ODF Validator is more strict about it. There is no guidance of that sort in the ODF specification itself. </orcmid> -- View this message in context: http://nabble.documentfoundation.org/Assessing-ODF-Conformance-Re-OASIS-Standard-ODF-1-2-Approved-tp3388789p3392038.html Sent from the Users mailing list archive at Nabble.com. -- For unsubscribe instructions e-mail to: [email protected] Problems? http://www.libreoffice.org/get-help/mailing-lists/how-to-unsubscribe/ Posting guidelines + more: http://wiki.documentfoundation.org/Netiquette List archive: http://listarchives.libreoffice.org/global/users/ All messages sent to this list will be publicly archived and cannot be deleted -- For unsubscribe instructions e-mail to: [email protected] Problems? http://www.libreoffice.org/get-help/mailing-lists/how-to-unsubscribe/ Posting guidelines + more: http://wiki.documentfoundation.org/Netiquette List archive: http://listarchives.libreoffice.org/global/users/ All messages sent to this list will be publicly archived and cannot be deleted
