I find it interesting that the two articles you cite concerning XMP are both from 2005 - a lifetime ago in the computer industry. XMP itself, as well the Adobe XMP Toolkits have advanced significantly in that time frame and has been adopted as the standard metadata format for various other international standards including PDF/A.
However, you are correct that at this time, Adobe is still the "maintainer" of the XMP specification. We've been in talks with a variety of standards bodies concerning trying to make it "fully open", but it's difficult to find one that has expertise in both metadata and XML, so that it would thrive there and not "die on the vine". We're open for any suggestions - especially from those that would participate in a open standardization process. We are working on actual schemas for all of the standard "namespaces" that are documented and in common use today, and will be including them with the next update to the documentation and toolkits. I believe this also address the RDF update issue, but I'll need to double check that. As above, we're always open for feedback on our technologies! Leonard Rosenthol PDF Standards Evangelist Adobe Systems On Nov 25, 2007, at 4:10 PM, Cyrille Berger wrote: > On Saturday 24 November 2007, Hubert Figuiere wrote: >> Could you enlighten us about the bad things? > > Good point: > * backward compatible with Exif/IPTC > * good documentation > * based on Open Standard and extensible > * can be embedded in a lot of file formats (and describes how to > do it) > > Bad point: > * controlled by Adobe (and only by them, I have nothing against > Adobe, just > that the working group of a spec should be opened) > * lacks schema (documentation isn't enough for me) > * incompatible with latest revision of RDF (that's annoying > because most of > the semantics web, and metadata outside the graphics world use > latest RDF) > see [1] for more details > * their dublin core shema is too limited, for instance the Creator > field only > allows the name of the author, while an author is more than that > (think of > address, contact, etc...), I know that because it's extensible I > can create a > new "xmp schema" that fixes that issue, but then my application > would be the > only one in the world to read/write them > > In the end, I do believe that all the issues could be easily > solved, but > because of the first point it's really depending on the willingness > of Adobe > to solve them, and I am not convince they will to do it, they > didn't took > those issues into account for OpenDocument (see [2]). > > [1] http://www.ldodds.com/blog/archives/000261.html > [2] http://lists.oasis-open.org/archives/office/200512/msg00009.html > -- > Cyrille Berger > _______________________________________________ > CREATE mailing list > [email protected] > http://lists.freedesktop.org/mailman/listinfo/create > _______________________________________________ CREATE mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/create
