Trying to request a prefix for SPASE identifiers

[BaptisteCecconi]

I would like to register a prefix for SPASE resource identifiers.

A SPASE resource Identifier is a name for a resource within the SPASE registry. This name can be used to identify a resource from the SPASE registry (for reference in metadata), to cross-reference entities in the SPASE registry, and to retrieve the metadata of an entity present in the SPASE Registry like a dataset, an observatory or a repository.

The requested prefix in spase.

The identifiers have the following URI form:

spase://NASA/DisplayData/ACE/MAG/Daily

The regex pattern to validate the URI is:

^spase://([A-Z]*)/([A-Za-z]*)/([A-Za-z0-9_/\-\.]*)$

Related information and references can be found in:

• https://spase-group.org/docs/conventions/Resource-ID-Formation-Rule-v5.pdf
• https://spase-group.org/services/naming-authority.html
• https://doi.org/10.48322/E72C-5Y75
• https://doi.org/10.1016/j.asr.2023.09.066

The NASA GSFC team (Greenbelt, MD, USA, ROR-ID: https://ror.org/0171mag52) is a main maintainer of the SPASE registry ecosystem, and also maintains the SPASE data model.

SPASE identifiers are URIs, but are not HTTP URLs. However, it is possible to access a web view of the SPASE registry by replacing spase:// by https://hpde.io/ in the identifier, so that we get a URL to the resource landing page.

For the last part (transforming the identifier into URL), I could not describe it on the web form.

Any advice on how to proceed in such a case?

[renatocjn]

Hello @BaptisteCecconi, we mainly don't support translation into non-HTTP(s) URIs as identifiers.org is mostly used as a redirection service for HTTP(S) users and we don't have a way to verify on the long run if the translated URIs are still valid.

If you have a strong use case, perhaps we could compromise by having two resources for spase, one for the hpde.io URLs and one with the spase:// URIs. Then we mark the hpde resource as the primary one for translation, making the redirection service use it every time to redirect http access. I will need an explanation on your use case first. Feel free to use our contact email (identifiers-org@ebi.ac.uk) if you prefer to not describe this publicly.

[renatocjn]

Hello @BaptisteCecconi,

I had another look and it would be possible to have a spase prefix in identifiers.org, but it would not be possible to support curies like the one you put as an example (spase://NASA/DisplayData/ACE/MAG/Daily) because of the double slashes after the colon. This is because the resolver of identifiers.org directly replaces what is after the colon into the URL pattern making it not compatible with the URL for webviews that you provide.

So what we could support would be URLs like http://identifiers.org/spase:NASA/DisplayData/ACE/MAG/Daily that would redirect to https://hpde.io/NASA/DisplayData/ACE/MAG/Daily. What do you think?

[renatocjn]

Will be closing this as stale. Feel free to reopen if you wish to move forward with the spase prefix.

[BaptisteCecconi]

Hi @renatocjn, thanks for the heads up. I didn’t see the previous answer. Since the last post, the SPASE landing pages are being moved to another domain (since the SPASE group lost the hpde.io domain…).

Our SPASE identifiers were created more than 20 years ago and have been curated and the related schema has been developed and maintained since then. Although not http URIs, they are used as defined in my first post. Changing the identifier syntax is not really an option, as it would imply a huge work on the SPASE registry records, as well as within the metadata of tens of thousands of data files, in many data repositories, that have been published in the last 20 years by the heliophysics community.

Retrospectively, the spase:// URI scheme was probably not the best choice, now that many application and identifiers rely on the http protocol to resolve the point to the record landing page. However, having an identifier being used for more than 20 years in a community is a nice achievement. That’s too bad that identifiers.org can’t support then (although I understand it would require a special processing).

Note: there is another set of identifiers, also defined about 20 years ago, by the astrophysics community, and that are used by all astrophysics archives in the world. They are named IVOA (International Virtual Observatory Alliance) identifiers, which are URIs, with scheme ivo://. So it is the same situation.

[renatocjn]

Hi @BaptisteCecconi,

I found this documentation on the SPASE Registry Server at https://spase-group.org/tools/registry/index.html. Is this something hosted publicly or is this meant to be something people host "indoors"?

If this is available publicly, it might be the solution to have SPASE on identifiers.org by redirecting to its "render" endpoint. For example: .../render?id={$id}. I think this way the double slashes could be supported.