Financial exchange's efforts to replace core systems with blockchain founder – again One of the world's most prominent and mission–critical uses of the blockchain - The Australian Securities Exchange's (ASX) project to replace its CHESS core trading systems with a distributed ledger – has struck further trouble. As The Register has previously noted, the project commenced in 2017 but saw its go-live date pushed …
So the ASX stock exchange isn't trustable to record transactions on ASX?
Do they plan on having voting? So if enough entities agrees a consensus blockchain record that is different from ASX's record, will those be able to overrule ASX?
But then what if you have trading, but don't participate directly in the blockchain? Doesn't that mean, you cannot rely on ASX's transaction record because it could be overturned by blockchain consensus?
And perhaps some of those blockchain entities would be foreign, Chinese even. So you're telling me it could be overturned and outside Aussie legal domain?
And the privacy problem, you do a trade, and it's between the buyer, seller and the market. But stick it in a big shared blockchain and you're violating a mass of privacy laws around the world. Plus of course participating in the block chain lets you see everyone else hand, so it favors the people in the blockchain.
Just cancel it already, dumb ass idea. Take away somebodies red stapler.
Fork is not a valid thing here.
Two forks cannot co-exist, one has to be the real transaction. If ASX will not yield to the majority then what is the point of the blockchain? It still comes down to "ASX's record is the definite record"!
At the end of the day I, (not in either group) has to pick sides on which record is correct! Which price was real, which transactions were real!
It's really dumber than that: THE BLOCKCHAIN IS NOT THE TRANSACTION. It's merely a RECORD of a REFERENCE to the transaction.
ASX has to be trusted to map those blockchains REFERENCES to the actual mechanics performing the transaction! Those actual database updates, the actual external messaging, the last trade price updates, the day volumes, the share transfer orders, etc., those things don't happen by magic when the blockchain is updated.
So ASX not trusted, yet has to still be trusted, because the transaction is not the blockchain record.
The blockchain adding nothing but problems.
There are plenty of good applications of Merkle trees already, like zfs and git. We really don't need good applications of blockchain, which is just a toy Merkle tree.
As Bruce Schneier has pointed out, we already have plenty of algorithms for append-only ledgers. Blockchain isn't novel and isn't particularly interesting.
All these people trying to cram a round peg in a cthulhu shaped hole because it's the hot new thing and the market is flooded with scammy consultants.
Like NFTs being a terrible solution for in-game problems that have been long solved better with other means, there are very few things a blockchain is actually needed for other than getting money thrown at you during funding rounds (this might not be the case any more).
Combining an ancient language, a decrepit processor and a useless blockchain
That's not what ASX are trying to do. They're taking a COBOL-on-Itanium application, which does not use blockchain, and trying to replace it with a new application that uses neither COBOL nor Itanium but does use a blockchain.
It may be a daft plan,1 but it's not what you described.
1Frankly, it's hard to see how a half-assed toy implementation of an append-only ledger is an improvement on the techniques that exchanges have been using for decades. Migrating the existing CHESS application off Itanium would be a good idea. Porting it to another language is a questionable one. Reimplementing it entirely using blockchain and DAML seems ludicrous.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2025
