For the sake of correctness and completeness, Twitter doesn't use multi-word tags because it's a mixed system (i.e., both tags and regular text are input in the same shared, entry, space). Thus, while the example given to illustrate solution 1 was correct and valid, the qualification used (namely, "strangely") was not.
Plus. That same Twitter scenario is the one where solution 3 makes (perfect)  sense (in order to allow for the use of multi-word tags, that is).

BTW:

I've also neglected to mention I see great potential in this Application. I would have not ventured making this (or any other) suggestion otherwise.

Thank you for the kind answers.
Will follow the new developments with added interest.

After using CN for a while, this is my number 1 request (even if I have others, not so pressing) for continuing to use it on a regular basis.
I've read the rationale for the initial decision and while I understand it I don't share it (and less as time and more time goes by: applications should strive, IMO, to approach the user POV and become more 'organic' and user-friendly, as time goes by). Workarounds create exceptions in the 'normal' behavior and are just one more hurdle to overcome (and not really solutions).
The "problem" seems based around the input: how to differentiate between different tags? 
There are, as I see it 3 possible solutions:
1 - Initial delimiter (like Twitter uses it -- even f they don't allow spaces either, strangely -- eg. #work #2014)
2 - Final delimiter (as YouTube uses it, allowing the uploader to tag videos with multi-word terms -- by using ,)
3 - A composite of 1 and 2: initial and final delimiter (for instance, "my tags")

My favorite(s): 1 or 2 (3 is overkill and not really needed).