Hi Simo,
Since none of our community members have answered as of yet, I thought I'd add some initial thoughts.
There's no right answer here, but it may be helpful to know a couple things when deciding.
ISAD(G) is an international archival description standard maintained by the International Council on Archives (
ICA). The goal was to create a minimal description standard as a common basis for international exchange - different countries could then implement their own national standards that conform at a base level to ICA, but also added their own specific rules or extensions based on their juridical context, national conventions, etc.
The American DACS standard is a good example of this - it is very closely aligned to ISAD(G), but has a few minor variations based on U.S. description preferences and conventions.
The Canadian RAD standard has a much more complicated history. It was actually first created
before ISAD(G) existed, which is part of why it differs so much. Because there were no international archival standards at the time of RAD's creation, the library cataloguing standard AACR2 was used as the basis - and then everything archives related that AACR2 didn't cover was thrown into a "Notes" section in the standard. This is why RAD has a bunch of sections that are uncommon in archival description standards, like the Standard Number section (which typically means an ISBN or ISSN, something found on books and other mass produced items for circulation, but not usually on archival records). If you're interested in learning more about the long history of RAD, the best resource I know on this is Richard Dancy's 2012 article "
RAD Past, Present, and Future" in the Canadian journal
Archivaria, available as a PDF
here.
Dublin Core also does not originate in the archival profession. It was first developed in the library community as a way of being able to standardize and exchange minimal metadata about web resources. It also predates ISAD(G) - you can read a brief history on the DCMI website,
here. The prime advantage of using DC is that your data is more easily exchanged with other cultural heritage institutions including libraries, etc. This comes at the expense of not having more specific archival metadata fields.
Now, as to using the templates in AtoM:
One thing that may alleviate your concern is that all of our standards-based templates are crosswalked where possible. This means that you can flip the default template in Admin > Settings > Default templates to a different description standard, and whenever there is an equivalent field in the chosen standard, your data will display correctly in the new template. Title in ISAD(G) is Title in DC and DACS and RAD and MODS, for example. Scope and content in DACS/RAD/ISAD(G) is Description in MODS and DC.
This means you can change at any time - and while some fields with no equivalent in the new standard may be hidden from view, and other fields you couldn't previously access will now be available, none of your data will be lost.
Additionally, there's actually an easy way to preview your data in another standard without having to change the global setting for the default template! Simply add a semicolon followed by the short name for the standard after the URL:
- ;dc = Dublin Core
- ;dacs = DACS
- ;rad = RAD
- ;mods = MODS
- ;isad = ISAD(G)
This is described in slides 9 and 10 of the following slide deck:
You can also permanently change the display standard of a single description, or a hierarchy, without changing the global default setting. This can be done by entering edit mode, and using the dropdown in the Administration area near the bottom of the edit form. See:
Cheers,