What a shame. All the credit should go where its due. But why did the Burst and Qora developers not try to develop it for NXT which was the most popular Java based platform at the time it was invented? I believe NEM is also written in Java.
As far as I remember, the AT mechanism (the base for the Burst/Qora atomic swap technique) was indeed discussed to be added to NXT, but the NXT team decided to not include it because of possible dangers of turing-complete smart contracts. I think they could have added atomic swaps in a "hard-coded" way, but - for whatever reason - they didn't. That's one of the problems with those "centrally-developed coins" - if the "official" development team doesn't like an improvement, it won't happen.
@687_2, DooMAD: I think the danger of an adversarial fork is mostly present when you trade small cryptocurrencies. In this case, a hostile attack is more likely because the adversary could control tools like blockchain explorers (e.g. bribing them) to support the "attack version" of the blockchain. This, however, even then would need a lot of preparation, and I don't think it would be profitable for <100USD trades. So I tend to agree with DooMAD here.
A problem could arise if people ignore the risk completely and rely on SPV/heavily pruned blockchains for big or multiple trades, because one adversary could for example offer an attractive price trying to lure many traders into his "trap". The risk could be similar to the acceptance of low-confirmation payments.