Let's add the major missing resources into "Community usage"

[matuskalas]

Message

Why this is so important

• This is extremely important for the maintenance of EDAM, to be able to see where and how a concept is used. In addition, it gives a head start in case curation is needed after a change in EDAM.

• This is also very helpful for fostering relationships with communities around resources that use EDAM.

• Last but not least, it allows end users to search over a number of diverse resources using EDAM.

A list of resources to add

Kind-of prioritised. Please feel free to add more or change order.
Perhaps also adding specific issues or even better draft PRs and linking them in the following list, when somebody starts looking into it.

• Galaxy (Please add Galaxy annotations to the "Community usage" #31)
• FAIRsharing
• Expasy
• Biosimulators
• WorkflowHub / CWL
• Debian Med (Please add Debian Med annotations to the "Community usage" #32)
• BIII (when EDAM Bioimaging works)
• EBI services
• ...

Open questions, risks, UX considerations

• Option to hide display of resources with 0 hits (resp. option to show also if 0 hits - sometimes needed to see explicitly). Either just one option for all (not sure what the default should be then), or options per all resources, with possibly different defaults (do we need that?)
• Configuration of 'community usage' might be needed in the future, if this should work across various ontologies. But for now I personally think it's ok to leave this feature EDAM-specific. An advantage of a "standalone" configuration might make it a bit easier to maintain the search queries (more explicit and less hidden).
• Perhaps the simplest solution for both of the above points: By default hide all with 0 hits, single option to list them all explicitly. If other ontologies need to add more resources to the code, we encourage them to do it (advantage: if such a resource starts using EDAM later, it's already included without extra work).
• Some resources are untrivial to query, of course and unfortunately. But it's important to do this, both for EDAM and for the resources in question. The best scenario would be working together with the resource communities on improving their APIs.

[hmenager]

My comments here:

List of resources

To be noted: accessing most of these requires a CORS-enabled API, which will probably turn out to be the most difficult here.
Each of these resources is going to require that we contact their maintainers (which is doable but won't be quick).

Open questions, risks, UX considerations

• Option to hide display of resources with 0 hits: I'm not sure whether we should do it or not. Knowing that a concept is not used is just as useful as knowing it is. Only justification for this is saving screen real estate, but I haven't seen it as a burning problem so far ;)
• Configuration of 'community usage' : I agree it's best to leave it as is for now.
• Some resources are untrivial to query: see my point above. It's going to be a time-consuming feature, but I basically agree!

[bryan-brancotte]

My comments here:

Why this is so important

Agreed !

A list of resources to add

Each resources will require different informations, and lead to distinct discussion. We can keep the current issue as a meta issue to gather discussion common to all community.

I created an issue template for requesting a new community usage. @matuskalas could you create an issue for each community you would want, and also indicate the API/WebPage where we could get information when available. You can also draft what you are aware of for this community, and how they should be helped to be able to expose their usage of edam.

Open questions, risks, UX considerations

• Option to hide : no hit is an information, I think we should not hide it for this reason but only if user want to hide a community
• Configuration of 'community usage': with a growing number of community, we will need it to spare both screen space, network, and endpoint cpu. Adding an option would be quite simple, it should be done only when new community are about to be added to the browser.
• Some resources are untrivial to query: I would even add that some cannot be queried for now. As @hmenager stated it will be time consuming, but if bringing the usage of a community in the browser is important, then allocating such time should be considered.