Hi @krisk 

I hope I’m not misunderstanding your question.

Dublin Core is flexible and does not prescribe a particular way (syntax or formatting) of populating the fields, e.g. any of these is valid:

<DC.Title>Dublin Core Basics: the beginners guide</DC.Title> 

<DC.Title>Dublin Core basics: the beginners guide</DC.Title>

<DC.Title>Dublin Core Basics: The Beginners Guide</DC.Title>

<DC.Title>Dublin Core basics : the beginner’s guide</DC.Title>

As are these:

<DC.Creator>Fernando Paladini</DC.Creator>

<DC.Creator>Paladini, Fernando</DC.Creator>

What you choose will usually depend on your institutional context and policies. For example, it might be determined by your general cataloguing practice. We have a library catalogue and, when creating DC metadata for Preservica, we would copy what is there, including the syntax/formatting. When our catalogue used to be based on AACR2, that’s what our DC metadata looked like. Nowadays it is based on RDA.

If you don’t have any policies to go by, I think a good practice is to copy whatever is on the resource. My advice would be -  don’t overthink it.

I find most of the DC resources quite onerous and confusing. If you haven’t seen these yet, here are some of the better ones (IMO):

https://paladini.github.io/dublin-core-basics/

https://www.dublincore.org/specifications/dublin-core/usageguide/

Here is an example:

And in the NLA catalogue https://catalogue.nla.gov.au/catalog/8540289

Thanks @lcoufal, I have also replied to @krisk via email regarding the Starter training video that is also applicable.

Thanks Libor for that example.

Yeah, DC is a simple key/value pair schema, and most of the fields are just text, so you can use them however you like. Your institution probably has some formatting guidelines for how you want to name authors and things like that, but they don’t need to conform with any particular format from a data perspective.

A couple of points on this for more advanced DC use cases:

• Fields in DC allow multiple values. This doesn’t make sense for some of them, but fields like creator, subject and publisher have good use cases for that
• If you want to index the date field as an actual date in Preservica (for date range filtering/faceting in search) you need to enter values as an ISO-8601 timestamp

For both of the above, the default search indexer that comes with Preservica would need modifying to index them for facets and filters. (It will still be searchable in a keyword search in any case.)

You’re on Starter, so this isn’t relevant for you at the moment (you can’t modify indexer documents, or set up faceted search), but I’m mentioning it here for context and for others who might find the thread. Your metadata can still be like that in case you want to do that in future.

I think we have quite a lot of other users using DC so maybe some of them in your sector can offer some guidance as well.

Interesting! Perhaps this is why librarians generally don’t care for DC and why LOC is now involved in developing sets of “qualifiers” for DC. Those qualifier sets would be like the guidelines that I am seeking, so maybe I will look into using those. I do think that I have decided not to use DC, however. I plan to put all of my metadata into the Description field, and will do so in such a way that the data will be moveable later once a good standard becomes available.

Information science folks have always known that it is necessary to adhere to standards, syntax, naming conventions and authority files when describing objects. Otherwise it is a case of “garbage in, garbage out” with everyone making it up as they go along. Perhaps computers have finally become so sophisticated that wild west style description won’t pose an issue for discoverability? Who knows?

Thank you all so much!