Identifying software in the archive- context and artifact identification

The archived software identifier can be represented with one of the options bellow:

1. the revision hash containing the metadata
2. the directory hash- direct link to the software artifact without the metadata which is already given on the client's page
3. full manifest with all the hashes for all the files and directories

If we use the double identification process with we see that the three solutions above are answering only the without the source and context information

With this approach the can be assimilated to the concept DOI (view on zenodo [2]) and the is the hash of a specific version (release, revision, directory)

DOIs:

1. https://blog.datacite.org/cool-dois/
2. http://help.zenodo.org/#versioning

---

Migrated from T736 (view on Phabricator)