Advice on a quick & dirty migration of a small set of studies resident in DVN v3.6

[Joerg Messer]

Greetings,

We're in the process of trying to migrate a relatively small portion (less than 200 studies) of our DVN v3.6 data holdings to DVN v4.x (Scholars Portal).  I've had a look at the recommended migration procedure and it appears to require that we upgrade our current installation or, at least, a clone of our current installation to v3.6.2/v3.6.3.  The question I have, and I know this is probably a long shot, has anyone come up with a quick & dirty way to extract the DVN v3.6 metadata which might be quicker than simply carrying out a manual migration? 

//Joerg

[John Huck]

Hi Joerg,

We moved from Dataverse 3.x to 4.x two years ago, and if you run into questions, you are welcome to contact us. We went through the upgrade steps one at a time. I'm not an expert, but I think that's probably the simplest thing to do, actually. I haven't heard of another way. Most of the effort will be around the step up to 4.x anyway.

We discovered two metadata elements that weren't coming over properly. I see Phil referenced the one about Funding Agency already, so it may have been addressed.

https://github.com/ualbertalib/dataverse/issues/45

I guess I'm taking this opportunity to flag the other major one for everyone, which is Software name. I always meant to share this, sorry! I lost track of how we fixed it and so didn't have code to submit as a PR. Don't know if it's a known issue or not.

https://github.com/ualbertalib/dataverse/issues/50

I documented metadata differences between 3.x and 4.x on this gitHub epic (and related sub issues). You'll forgive me if my documentation is a little idiosyncratic and cryptic, but you may find some interest in it if you want to know about things like different note fields. My mapping is in a spreadsheet linked from the issue.

https://github.com/ualbertalib/dataverse/issues/44

Cheers!

John

--

John Huck

Metadata & Cataloguing Librarian

University of Alberta Libraries

5-25E Cameron Library

(780) 248-1337

--
You received this message because you are subscribed to the Google Groups "Dataverse Migration Working Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to dataverse-migrati...@googlegroups.com.
To post to this group, send email to dataverse-m...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/dataverse-migration-wg/98ee7e54-97aa-4064-9955-b926bf7d51a5%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Philip Durbin]

I'm sorry to be the bearer of bad news but the need to migrate Funding Agency (fundAg) from DVN 3.x still has not been addressed. Please see https://github.com/IQSS/dataverse/issues/3323 . On a related note, there are two ways to indicate funders in the Schema.org JSON-LD representation of datasets now that https://github.com/IQSS/dataverse/pull/5169 has been merged (but not yet released): under "Contributor" ("Funder") and "Grant Information" ("Grant Agency").

With regard to software name, John, it would be much appreciated if you create an issue for the need to migrate this field over at https://github.com/IQSS/dataverse/issues because I don't believe we're tracking this. We've been talking a fair amount about software lately in the context of code deposit and import from GitHub in https://github.com/IQSS/dataverse/issues/5209 . There's a software metadata block at https://github.com/jhu-sheridan-libraries/dataverse-ansible-role/blob/master/files/dv-mdblocks/jhuSoftware.tsv that's in production at https://archive.data.jhu.edu that we've looked at a bit. "swTitle" has a description of "Name of software package".

Joerg, I think this issue is related to your migration: https://github.com/IQSS/dataverse/issues/4996

Sorry I don't have more to offer. I hope this helps.

Phil

To view this discussion on the web visit https://groups.google.com/d/msgid/dataverse-migration-wg/CAPG0r%2BcDAywFr5fhD-XmNJDCEd%2Br7L5gH1%3DrnKSjT-m3SU8A0w%40mail.gmail.com.

For more options, visit https://groups.google.com/d/optout.

--

Philip Durbin
Software Developer for http://dataverse.org
http://www.iq.harvard.edu/people/philip-durbin

[Joerg Messer]

Hi John and Philip,

Thanks for you comments.  It seems that the complexity of a full migration may not be justified if we're just moving a few data sets.

//Joerg

To unsubscribe from this group and stop receiving emails from it, send an email to dataverse-migration-wg+unsub...@googlegroups.com.

To post to this group, send email to dataverse-m...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/dataverse-migration-wg/98ee7e54-97aa-4064-9955-b926bf7d51a5%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google Groups "Dataverse Migration Working Group" group.

To unsubscribe from this group and stop receiving emails from it, send an email to dataverse-migration-wg+unsub...@googlegroups.com.

To post to this group, send email to dataverse-m...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/dataverse-migration-wg/CAPG0r%2BcDAywFr5fhD-XmNJDCEd%2Br7L5gH1%3DrnKSjT-m3SU8A0w%40mail.gmail.com.
For more options, visit https://groups.google.com/d/optout.