Specifically, I think it would be beneficial to set up a tubes project on Github under the Twisted organization and try for a release in the very near future.
I think this has several advantages over the status quo:
+1 for splitting into a separate project.
even if Tubes would be considered experimental / mostly unusable right now.
the best thing that could happen: people actually start playing around with it. which provides real-world feedback.
"finishing" it in the backroom and only later release something "ready" without gathering real-world feedback: that's a risk IMO, rather than good practice.
I guess I am a proponent of the "release early release often" mantra.
better release something unfinished and gain feedback, than release something "finished" not meeting user expectations.
just my 2cts,