FIAT/IFTA Television-Archives FORUM

[Gerhard Stanz]

In the past we have elaborated a set of metadata items to describe tapes sufficiently for archive purposes.
What are the archive-experiences with technical metatadata for files so far?
The problem seems trivial as long as you have only one file format, so I would especially be interested in comments of broadcasters who have different file-formats in their archives.

[Kesavan_kk]

I work with Tangerine Digital Entertainment, offering metadata capture services to customers from offshore locations in India. We work with two large broadcasters one in UK and another in the US capturing technical metadata for their thousands of hours of archive content. In our experience I have seen companies approach this problem in the following way:

1. Kick-off an archive assessment project upfront and identify archives of similar format, set priorities for digitisation and also highlight their re-purposing needs.

2. Digitise the content to a standard archive format, hence remove the source format differences.

3. Have a well defined taxonomy for capturing metdata.

4. If there are more than one digital distribution platform then there is some amount of metadata transformation that is required. For very large archives, we have developed automated tools to transform metadata once captured in a format to other formats. For smaller archives, typically its a manual process.

5. Invest in some amount of training upfront and outsource the work to people at offshore locations who can focus on the work specifically and get it done at cheaper costs.


Essentially, there is no one product or automated tool at this point to do this, though the industry is moving fast. Hope this partially addresses your question.

[Rafeeque]

Hi Kesavan,

Very interesting to see what you do. I wish to find out if you employ standards for applying metadata element sets, if yes, what it is. I would also like to know what are key elements that you focus on...In an already digital environemnt, where metadata to content are missing, how would go about it? In other words, if content is already digitized on server, but without metadata, what techonology can be applied to capture all relevant metadata?

Thanks.