I strongly suggest to separate your DOI namespace/naming schema to be totally independent of your choice of repository/system. DOI is an infrastructure thing, and the main reason behind of assigning DOI is for persistency and permanency. At some point any repository system will go away and will be replaced by another one.
Secondly I do not think data is a good namespace. I suggest to have something persistent which can standalone even you do not have DOI prefix of 10.17348.
On 2/9/16, 11:56 AM, "Code for Libraries on behalf of Jason Best" <[log in to unmask] on behalf of [log in to unmask]> wrote:
>We recently started assigning DOIs to articles published in one of our journals using Open Journal System which generates the DOI and metadata within a namespace dedicated to that journal. We don’t yet have an institutional repository, but are moving in that direction and I hope we have one in a couple of years. But in the meantime, how could we go about issuing DOIs for items that aren’t related to the journal, but that we’d hope to eventually have handled by our IR? For example, we have a handful of datasets for which we’d like to issue DOIs so I planned on created a “data” namespace then just adding a serial number for each dataset (e.g. 10.17348/data.000001 ) which would resolve to a page (with metadata and download links) in our Drupal CMS until we get an IR. Will such an approach allow us to eventually use an IR to 1) become the repository for the items with DOIs previously issued in the “data” namespace and 2) continue issuing DOIs for new items within!
> the “data” namespace? I know the answer is going to depend on the IR platform we use, so I’m asking this in the broad sense to get your input about your experiences.
>But since DSpace is one of the likely candidates for our IR, I’ll use it as a more concrete example. From my limited understanding (just reading the documentation), items deposited in a DSpace instance will all share the same DOI namespace. The namespace and an internal identifier are then concatenated with the DOI prefix to create the DOI. If we’ve already issued a DOI outside of DSpace, would we have any control over the identifier that was assigned to a newly-deposited item allowing us to control the DOI that is generated?
>Any thoughts or suggestions?
>Botanical Research Institute of Texas
>1700 University Drive
>Fort Worth, Texas 76107
>817-332-4441 ext. 230