So, is this already implemented on bitcoin's code?
no because anyone smart enough to warrant trying it will already understand why its not worth it. but *just in case* there are stupid people in the world with more money than sense (i know). that thread gives some reassurance that its not the issue people think it is.
whats more telling is the way people seem to think this is a black an white issue.
Setting aside transaction manipulation, which is largely a non-issue given the nature and transparancy of the bitcoin protocol.
In terms of block reward hash rate doesn't guarantee anything. hash rate as the fuel to the fire that is block generation. block generation is what gives you the power to manipulate the chain.
Having 51% doesn't magically give you guaranteed control, just as 49% doesn't mean you can't possibly manipulate it.
The hash rate is a reflection of the probability you can solve the next block before anyone else. 51% is marginal, its basically still a coin flip, just like 49% is.
With 51% The chance you can solve two consecutive blocks is 0.51^2 = 26%
The chance you can get 6 blocks is 0.51^6 = 1.7%
Anyone with the most base understanding of how things work can see that with 51% of the hashrate you are going to average 75BTC per 6 blocks.
Over time you are averaging 12.5 BTC per block.
Or you can get 150BTC 1.7% of the time by trying to build your own 6 block longer chain before anyone else.
Over time you would average 0.43BTC per block.
To make it worthwhile, you would need to get your 6 block chain in 50% of the time or more. To guarantee that you need 89% of the hashing power (6th root of 0.5)
If you have 89% of hashing power though, you'd still be better of just hashing. because you would average 22BTC per block.
In a way its good that idiots with a lot of hashing power might try this, because in the long term it means their hashing power wouldn't count, and any remaing honest miners get a bigger piece of the pie