The reason for this is imo, that Dash promised a lot and a lot of projects started, but one has the impression that none of those really found it's end.
For instance darkmixing. For how long now is this feature available now? A year? Recently i used it for the first time and was SO dissapointed. It took me 2 (!) days to mix 1000 coins. Unbelievable! Who do expect to use that stuff when everything takes so long? I know its a solvable issue. But ffs its been a year now! Same goes for masternode blinding etc.
You got to finish things first before moving ahead and open new projects.
Justmy2satoshis.
It's a possible chicken-egg problem. Darksend speed will only increase with more active usage (thus more users). The question is if it will attract more usage being horrendously slow (I think 2 days counts as horrendously).
Any system like Darksend that depends on contemporaneous volume for mixing is hopelessly vulnerable to attackers flooding the mix with their own coins, in order to flush out targets of surveillance. The more fake usage (and compromised masternodes), the more actual usage is degraded in terms of privacy/security.
That's why ring sig approaches like Cryptonote are a quantum leap ahead, in that they depend on the total mass of (emitted) coins for camouflage, rather than just what's currently in the queue to be mixed.
The former is a kludge, vainly struggling to emulate the actual (rather than "sort of, hopefully) unlinkability of the latter's zero-knowledge based approach.
Dufffield used to say nice things about ring signatures and the desirability of using them to improve Dash's "bad crypto."
But protecting the instamine was apparently more important than fixing the broken "privacy theater" of Darksend (aka broken, snake oil rip-off cargo cult version of CoinJoin).