{"id":26756,"date":"2024-06-12T20:56:04","date_gmt":"2024-06-12T20:56:04","guid":{"rendered":"https:\/\/davidgerard.co.uk\/blockchain\/?p=26756"},"modified":"2024-06-13T16:51:42","modified_gmt":"2024-06-13T16:51:42","slug":"the-australian-securities-exchange-chess-blockchain-seven-years-of-sunk-cost-fallacy","status":"publish","type":"post","link":"https:\/\/davidgerard.co.uk\/blockchain\/2024\/06\/12\/the-australian-securities-exchange-chess-blockchain-seven-years-of-sunk-cost-fallacy\/","title":{"rendered":"The Australian Securities Exchange CHESS blockchain: seven years of sunk cost fallacy"},"content":{"rendered":"

Enterprise blockchain never made any sense. A slow data store was never going to magically make your business or bureaucracy more efficient. Because magic doesn\u2019t happen.<\/p>\n

That didn\u2019t stop the fierce and dedicated pursuit of the dream \u2014 nor setting millions of dollars on fire in the process.<\/p>\n

In 2015, the Australian Securities Exchange’s CEO-before-last, Elmer Funke Kupper, got a blockchain bee in his bonnet. “Every now and then, something comes along that might just change everything. And this is one of those moments,\u201d he told the Sydney Morning Herald. [SMH<\/i><\/a>, 2016<\/i>]<\/p>\n

In 2022, ASX finally shut down its blockchain trading system project \u2014 after seven years of no results and AUD$250 million up in smoke.<\/p>\n

\"\"<\/a><\/p>\n

The magic of blockchain<\/h3>\n

When I was writing Attack of the 50 Foot Blockchain<\/i><\/a> in 2017, Blythe Masters\u2019 company Digital Asset Holdings was in talks with the ASX to replace their ageing Clearing House Electronic Sub-register System (CHESS) with … something descended from blockchains.<\/p>\n

CHESS was launched in 1994. It works and it\u2019s reliable enough. But it\u2019s thirty years of legacy COBOL code, and maintenance has long been trepidatious.<\/p>\n

Unfortunately, Digital Asset sold the ASX on the idea of a \u201cdigital ledger.\u201d<\/p>\n

The CHESS replacement system \u2014 it never had a real name, just \u201cthe CHESS replacement system\u201d \u2014 was based on Hyperledger. ASX would administer it centrally. The functionality would be smart contracts written in DAML \u2014 a functional, formally-provable and non-Turing-complete language \u2014 though DAML hadn\u2019t actually been finished by the time ASX accepted it.<\/p>\n

In enterprise computing, \u201csmart contracts\u201d are called \u201cdatabase triggers\u201d or \u201cstored procedures.\u201d They\u2019re a nightmare, because they\u2019re very hard to reason about or maintain, and they\u2019re prone to unexpected and spooky effects. You only use them if you absolutely have to for performance.<\/p>\n

Even cryptocurrency exchanges<\/i> don\u2019t use blockchains for their internal systems. They run their trading engines on a conventional database system.<\/p>\n

The CHESS replacement system used the wrong approach, with the wrong technology, from a company who’d never done a real system at all<\/i> \u2014 let alone a huge one.<\/p>\n

Circling the drain<\/h3>\n

Digital Asset spent January to June 2016 pitching a prototype system. ASX invested US$17.4 million into Digital Asset in 2016 for 8.5% of the company and engaged them to build the replacement for CHESS in December 2017. [FT<\/i><\/a>, 2016, paywalled; <\/i>press release<\/i><\/a>, 2017, PDF<\/i>]<\/p>\n

Funke Kupper quit ASX in 2016, but his replacement as CEO, Dominic Stevens, stuck with the blockchain plan. Stevens said the blockchain system would “put Australia at the forefront of innovation in financial markets.” [BBC<\/i><\/a>, 2017<\/i>]<\/p>\n

Digital Asset\u2019s prototype blockchain system was built and rebuilt, with increasing delays. Perfectly sensible and capable IT people were working on the project \u2014 I saw a presentation in early 2018 that impressed me with how sane and reasonable it was \u2014 but it could barely deliver even contemporary CHESS speeds, let alone the promised improvements.<\/p>\n

Other market participants hated the new system. It didn\u2019t work properly, they were spending a fortune building and rebuilding their interfaces to it, and they complained repeatedly to the regulator that the ASX was ignoring their problems.<\/p>\n

The other participants were also unhappy that the project would make the ASX a new centralised controlling octopus, to a degree it wasn\u2019t before. This is a blockchain project failure mode I detailed in chapter 11<\/a> of the book.<\/a> The users asked that the project be delayed further, as it just wasn\u2019t ready to go live. [FT<\/i><\/a>, 2020, paywalled<\/i>]<\/p>\n

The system moved from Hyperledger to VMWare Blockchain in 2019. This didn\u2019t help matters. But ASX launched Synfini, giving customers DAML on VMWare Blockchain as a service! ASX got crypto firms DigitalX and Zerocap running pilot systems on Synfini. [VMWare<\/i><\/a>, 2022; <\/i>ASX<\/i><\/a>; <\/i>AFR<\/i><\/a>, 2022; <\/i>Cointelegraph<\/i><\/a>, 2022<\/i>]<\/p>\n

And flush<\/h3>\n

Finally in 2022, the ASX got a new CEO, Helen Lofthouse. She looked at this shuddering zombie and killed it in the most effective possible manner \u2014 she got Accenture to write a report on it.<\/p>\n

Even Accenture couldn\u2019t see a way to cash in on this thing, and said it needed killing. [ASX CHESS Replacement Application Delivery Review<\/i><\/a>, PDF, 2022<\/i>]<\/p>\n

Accenture found four key issues:<\/p>\n