Val Lorentz

I don't think clients should strip it, because it makes it impossible for servers which want to show such prefixes to reliably show them. And servers that don't want to...

Well yes, I made it up because from what I noticed. Examples of numerics extended this way: RPL_TIME, RPL_LOCALUSERS, RPL_GLOBALUSERS, RPL_ADMINME, RPL_WHOISACTUALLY Also, without the human-readable trailing: RPL_TOPICWHOTIME, RPL_BANLIST

This is for future-proofing against servers who will add parameters in the future. eg. this issue with RPL_WHOISCOUNTRY in kiwiirc: https://github.com/kiwiirc/irc-framework/pull/318/commits/e6c761f9b32a66078c9df326736d4221346b8c3f

actually, the issue is caused by doi.org now serving its own JSON-LD documents (via a redirect to Datacite). https://doi.org/10.5063/schema/codemeta-2.0 is supposed to redirect to https://raw.githubusercontent.com/codemeta/codemeta/2.0/codemeta.jsonld , and it still does...

> SoftwareSourceCode does have `targetApplication` Do you mean `targetProduct`? > but this does not seem to refer to the application that the source-code is defining, but rather a runtime environment...

@dgarijo It's not obvious here because we always encode Codemeta as a tree rooted in a single SoftwareSourceCode object, but you need to think of JSON-LD as a graph. Objects...

> Probably best to just have a separate .csv file for each. As in, a .csv with two columns? Sounds good! > a JSON-LD context for each What do you...

Got it, thanks! Using a JSON-LD context here would have the extra advantage of working with nested structures; but as its expressivity would be very limited (eg. NPM's `author` field...

@mbjones I find the CSV table in the repository to be hard to use, because GitHub limits its width (even on large screen). Instead, it could be build from the...

Done (but not tested)