- Before creating CCMM profile in Dataspecer, it is necessary to make an analysis of CCMM and decide whether it is really necessary to create your own application profile. If CCMM does not meet the subject-specific requirements of the community, we also recommend exploring existing subject-specific metadata schemas and standards and trying to select a suitable existing schema. Only after this analysis and finding that there is no suitable existing schema, should you choose the way of your own CCMM application profile. Also consider that your own application profile will need to be maintained and updated in the long term according to the development of the standards used or according to the needs of the community.
- You can install Dataspecer locally or on a server using Docker, the procedure is described on GitHub. You can also use its public instances. They are managed by the authors of Dataspecer and regularly updated, but when using a public instance, anyone has full access to your work – your work needs to be backed up regularly, for example, on GitHub.
- Next, you need to create a new application profile according to the instructions in the Methodology for Metadata Profile Management, and select CCMM as its basis. Always import CCMM into Dataspecer from this URL address: https://model.ccmm.cz/research-data/en/.
- Add other schemas and vocabularies from which you would like to supplement your CCMM application profile according to the instructions. All these schemas must be according to the RDF Schema and must be publicly available on a server supporting CORS (Cross-Origin Resource Sharing). It is also a good idea to check whether the schema is published under a license and can be freely used. It is also possible to create your own classes, attributes and relations, in which case it is also necessary to create your own vocabulary for the application profile.
- When creating a profile, follow the instructions in the Methodology for Metadata Profile Management to create a vocabulary and your own application profile.
- Export your specification according to the instructions. You will get a compressed zip file, the contents of which, after unpacking, you can post to a publicly accessible URL (e.g. on GitHub Pages or on your own web server). This URL can then be imported into Dataspecer (see point 3.) and a derived application profile can be built again.
- The necessary terms are explained in the Methodology for Metadata Profile Management.