What happened to the idea of automatically kicking out bad delegates

[vegolino]

This may help
https://bitsharestalk.org/index.php?topic=5164.msg67657#msg67657

[santaclause102]

In the original post about DPOS and in the discussion following it there was the idea to automatically downvote delegates that harm the network. What happened to that concept?
Is it hard to identify in which case delegates are clearly malicious and a permanent threat? And/or is it hard to say which group (the 51% or the 49%) are the malicious actors?

From the DPOS whitepaper (http://bitshares.org/delegated-proof-of-stake/): 

Every client connected to the network can observe the behavior of all delegates and detect the following bad behaviors automatically:

Failure to include valid transactions that were broadcast more than 1 minute prior to the production of the block. One minute should be more than enough propagation time for all delegates to receive the transaction because their role requires them to be well connected to the network.
Failure to produce a block before the end of their window.
Failure to reference prior blocks that were produced and distributed in time.
Signing an invalid block.
When bad behavior is detected by any delegate all clients will automatically vote against that delegate the next time their user makes a transaction until that delegate is no longer in the top 200.