Platform Idea: Retrieving website metadata

[raucao]

At the least via normal HTML meta attributes and Open Graph. This would be useful for clients to both not require a separate CORS proxy, as well as for not having to parse the original source code in the client.

We could also use the same platform to discover if the target is a Web Application, and retrieve meta data from the Web Application Manifest.

WDYT?

---

Also, a nice performance enhancement for a later stage would be to cache this data for a little while and deliver it to other clients requesting it from cache. For example when someone posts a link to a chat room, everyone's Hyperchannel would immediately request this data, and then Sockethub would only have to retrieve it once and then delete it from memory after a minute or so.

[silverbucket]

Yeah this makes sense, additionally. Might it be useful to try to find RDFa data [https://github.com//issues/162] (though I think Open Graph is much much more common)

[silverbucket]

Type can be page
https://www.w3.org/TR/activitystreams-vocabulary/#dfn-page