They did this in the hope they could get their (useless) block size increase after activating segwit. However there is absolutely nothing that guarantees the lock in of the hard fork 3 months from now based on bit 4 being active at the moment
The *only* useful thing that could come out of this is that bitcoin could finally jailbreak out of the 1 MB artificial lock-up, and segwit is in fact making this worse by trying to increase a very last time the amount of transactions per block, by delegating part of the block data outside of the "official" block size, but this cannot, as far as I understand, be extended yet another time. In other words, segwit is the "end of the road" for more transactions if the jail of 1 MB blocks remains. Yes, it will get us some temporary relief, but on the other hand, it has graved even more into stone the 1 MB limit and the "danger of hard forks" (which is a non-issue in most other crypto coins which do this regularly).
Once you break the jail once, you're free for ever. However, if bitcoin doesn't succeed in overcoming this silly 1 MB limit now, it may even be much, much harder in the future to do so ; this is why it is of utmost importance that 2 MB blocks (with or without segwit) are adopted, to kill this 1 MB hard wall once and for all.
That doesn't solve fundamentally the essential design flaws in bitcoin, but at least, it doesn't turn a silly error into yet another extra fundamental design flaw for ever (I'm talking about the 1 MB limit).
You misunderstand me, probably because of my flamebait "useless" comment. I'm for a 2MB hard fork and said so earlier in this thread. What I'm against is a rushed hard fork done by morons who don't understand the code they're modifying earlier than is needed. Once segwit locks in it will ease the pressure on transactions for the time being - and there actually ISN'T any pressure on transactions now that the artificial spamming is over. I'd be for a hard fork in 12 months from now, done with the cooperation of core instead of rushed by outsiders forking the code and missing out on other developments in the code. The bulk of the core group has said they'd push for 2MB in the past, BUT the pressure in all directions is making all this information obscured.