How long do you think it could take us to build something like this with a nano kontrol and tidal 
https://www.torsoelectronics.com/product/t-1
I am being mean, since actually this review gave me new ideas on how to use tidal :
But at the same time, having a limited tidal for 500 $ feels a bit frustrating.
Tidal is the ultimate sequencer. I'm extremely biased and you'll never convince me otherwise! Semi-inspired by some of what I saw in the video (untested):
d1
$ degradeBy 0.5
$ gain "1(<10 7 13 15, 16, <0 6 3>)"
# s "arpy"
# note (scale "ritusen" "<0 1 2 3>")
And I haven't even added conditional logic, any pattern transformations, or "Turing Machine" style randomness with sample-and-hold. With such a small amount of code you can do so much.
Nano Kontrol MIDI input is 100% doable. You could easily map a knob to something like the degradeBy amount in the above pattern. It could also control the Euclid shift, conditional triggers, the sample being played, etc.
I agree 100%, I think Tidal is able to do much more than what this little hardware does. I was just both a bit sarcastic and at the same time glad those concepts are going mainstream. This and the NDLR that is mentioned in the review are great controllers but still limited, and I got really interested in Tidal to do exactly this kind of control.
Back to your example, last sentence, I actually tried to control euclid with MIDI last night but completely failed. I referred to it in another thread : Using variables within patterns - #9 by artheist
I also tried stuff like:
d1 $ euclid (cI 3 "21") (cI 8 "22") $ sound "bd"
d1 $ euclid "^21" "^22" $ sound "bd"
But both these attempts remained silent. My feeling is that taking the pattern as an input is the root of the problem, as opposed to effects applied to it, but my tidal konwledge is still limited yet to find the right answer.