[TYPO3-dam] Writing Metadata back into files - any experiences?

[Torsten Schrade]

Hi Lorenz,

many thanks for your quick reply.

> The cc_* extensions are in fact only doing meta extraction and
> unfortunately they are not very bug free. These days the extensions will
> be transferred to the DAM team and we will either do bugfix releases or
> (what we preferred in a team meeting back in January) will publish a new
> extension.

That's good news! On a more general perspective I believe that the
'historic' set of fields provided by DAM may need to be thought through
as well. At the time of the first release it was partly oriented on the
IPTC standard with some additional web and TYPO3 related fields. But
standards have moved on [1]... At least it would be really cool to have
an easy and extensible matching mechanism between embedded metadata and
DAM fields. But that's just my 2 cents...

> We are currently developing DAM 2.0. If you wish to share your insights
> about writing data back to the XMP/IPTC data, feel free to contact me as
> we then could add it to the feature set.

Thanks for the invitiation, I will report back during the next few
wheeks while I work on my project. There could indeed be several
scenarios where metadata is written back from DAM to files:

1. Any files generated for display on the website by TYPO3s imaging
mechanisms (typo3tmp files). At the moment, embedded metadata is
basically lost since it stays with the 'original' in DAM/fileadmin.

2. Files in DAM. At the moment, embedded metadata get's out of sync when
records are updated in DAM. It could be an interesting feature to have
things written back to the fileas as soon as a DAM record is saved.
The question then would be if the metadata of the file (and the DAM meta
field) should be fully replaced or just updated.

That's just a few raw thoughts on my part. Maybe someone else also has
some ideas or - even better - some experience on that topic ;)

Best regards, Torsten



[1]
http://www.iptc.org/std/photometadata/specification/IPTC-PhotoMetadata-201007_1.pdf






> Am 03.08.2011 07:10, schrieb Torsten Schrade:
>> Dear list,
>>
>> while working on a project that will employ the DAM in combination with
>> XMP/IPTC Core Data extraction I just wondered if anybody has ever tried
>> the reverse way - write metadata that was modified in DAM back into the
>> files?
>>
>> The cc_* metaExtract services have been in place for years and work
>> excellently - but to my knowledge (and as their name implies) they just
>> do the extraction...
>>
>> Generally, there seem to be some PHP based solutions out there [1], some
>> of which are already partly used in the cc_* extensions (like XMP PHP
>> Toolkit).
>>
>> So in conclusion, I would be very grateful if anybody has any
>> experiences to share.
>>
>> Cheers and have a nice day,
>> Torsten
>>
>> [1] http://wiki.creativecommons.org/XMP_Implementations
>>
>>
>