With regards to a comms infrastructure, it's actually remarkably easy. The currently confirmed plans extend to PM functionality in the forum, I personally feel this is lacking and yearn for more. There's multiple methods, ofc, but I lean towards XMPP. It's open source, and clients already exist on most platforms. It's well tested and industrially reliable. Mine survived most assaults launched by a well funded organisation, over a long operating window. Eventually they just gave up. The native TLS that wraps this website's use can be lent to the raw connection and additional techniques to ensure privacy and authenticity of the remote user can be applied via common technologies such as GPG/PGP and OTR. It's been used in the backend of things like gtalk and completely neutered of functionality and security and used in things like facebook chat and whatsapp. It's possible for any user to run a server on their home machine and connect to that, and then connect to another user on another server - so could serve as distributed network. It supports text, voice, vid, files - anything you can work out how to cram into an XML stanza. Connect it to SIP/VOIP server (say, Kamailio) and it can become a telephone, too. Does 1:1, 1:many, many:many type comms.
The existing infrastructure could cope with the load, it's quite lightweight. Heavy vid/voice use might pose bandwidth issues, however, but as a text only majority, and a negotiator in file transfers, it should cope admirably. Ideally, yes, this should require it's own machine - possibly several geographically distributed to spread the load, this is also trivial to achieve once the hardware is in place.
With the comments, even with subsections to the ratings, it's still basically popular opinion. What other people think shouldn't be ½ as much of a concern as much as what you think. That might sound a little arrogant, maybe even conceited - But the only measure you truely have in life is yourself. You might like marmite. You might think that's a good thing to put on toast - doesn't mean everyone else does. That level of information is only really suitable to folks who like marmite, to anyone else - no matter how much you like marmite - it's useless... How many that like it is irrelevant, what matters to them is they can stick it on toast...
I understand the "distraction", I truely do, but you have no actual requirement to acknowledge/respond. With better user interface and a few tools such work can progress - but I would not to think it to be any way exclusive. Another feature I've seen several times mentioned, and I feel is an excellent addition to the forum structure itself, is to be able to nest replies to individial posts within a thread presented to the user in a collapsable tree - so tangents can spawn and ratify without disrupting the flow of main discussions. This education of which I speak could take place there. Someone has a question about a principle, they go to reply to the post - see someone else has had the same question, and there's already a 30-page discussion... they might learn without having to even post... Maybe I'm just over-optimistic. There's no need to hide this - just give the user options not to see it, the critical difference being the optional component - something I feel essential wherever it's possible to sensibly place something. Such a structure worked well for newsgroups (NNTP) from the early 80's up until the web and forums appeared in the mid to late 90's. Some forums adopted this nested structure.
As to database... I'd describe as "admin aware" - with evidence of other problems in an effort to reduce their workload I was just going to give it to 'em fully working - just add data. The database itself should be nice and simple and some PHP form to search should be equally so, and integrating this should be dolly dimple. There's a couple shown interest in that topic so between us something, somehow, is assured to happen. Actually implimenting it, ofc, is entirely up to them - but I see no reason why that shouldn't happen.
I don't think there's anything "politically incorrect" about "scientists of Asgardia" - but do predict issues with retaining it to just "qualified scientist". I understand the arguments you make regarding, but am unable to resovle issues concerning "exclusivity". On this loose subject, someone somewhere else suggested an area where scientists/reserchers can submit papers/studies/data for peer review. Submissions to which I'm sure should entertain "scientists of Asgardia" - and could serve to set/support our position as a base of scientific understanding - and lend credible worth to our educational system once we actually have one.