You’re right about the limits but I’m not sure I agree in this case - in many communities email is still the primary way to engage, and discourse is so powerful in large part because it supports email so well.
The digest is already customized for each subscriber and the translation is already stored in the database, so why not also send them the translation that they can read?
Let me explain my use case. Our network has a strong contingent in Myanmar, many of whom speak Burmese and little or no English. They make strong use of Facebook and email for networking and information sharing already, and we’d like to encourage them to use our discourse platform instead. Unfortunately web apps like Discourse are slow or hard to access in Myanmar. So there has been a request for a regular email digest (say twice a month) of especially valuable topics, translated manually into Burmese. This is a time intensive project and not scalable to other language communities also of strategic importance to us, so we’d prefer not to have to put this digest together manually.
Instead I’d like to explore using discourse and the discourse-translator++ plugin to encourage engagement by Burmese speakers. I think we’re pretty close - we need to get discourse localized into Burmese (I’ve gotten this started already), switch to Google for providing automatic translations which I think supports Burmese (but costs money so I haven’t tested it yet), and add functionality for manually adding/improving translations (which @tgxworld has indicated should be doable).
In this context, unless I am missing something about the technical aspects which is quite possible, I think localized digests make plenty of sense and are scalable. I can see this being handy for many communities.