Linking to extenal datasets?

[susan borda]

Hi-
We are currently using DSpace for our IR, and are considering using Dataverse for our data repository. How are large files handled in Dataverse? Can you link to datasets hosted elsewhere?

Thanks,
susan

[Philip Durbin]

As I mentioned at https://groups.google.com/forum/#!topic/dataverse-community/Pe6znGA000Y currently large files are not handled any differently than small files. We're working a new project to support large biomedical datasets: https://hms.harvard.edu/news/dropbox-structural-biologists

DVN 3.x supported a feature called "harvesting" wherein metadata for datasets hosted elsewhere could be made discoverable/searchable within Dataverse while the original files remain hosted on the remote site. When we upgraded the Harvard Dataverse from DVN 3.x to 4.0 we put all the datasets we had harvested in the DVN 3.x days at https://dataverse.harvard.edu/dataverse/harvested where you can see how the metadata is available but the files themselves are elsewhere. We haven't yet added the harvesting feature to Dataverse 4 but we are working in it at https://github.com/IQSS/dataverse/issues/813 . Comments are welcome on the Functional Requirements Document linked from that ticket!

I hope this helps,

Phil

--
You received this message because you are subscribed to the Google Groups "Dataverse Users Community" group.
To unsubscribe from this group and stop receiving emails from it, send an email to dataverse-commu...@googlegroups.com.
To post to this group, send email to dataverse...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/dataverse-community/1f0b9ed8-952b-4ec1-84b8-01d0128b7db0%40googlegroups.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