It's over, TSB is finished

https://www.nakedcapitalism.com/2018/04/tsb-train-wreck-massive-bank-it-failure-going-into-fifth-day-customers-locked-out-of-accounts-getting-into-other-peoples-accounts-getting-bogus-data.html

https://twitter.com/thejackthomson_/status/988564354512687104?s=20

imagine being a software developer for banking software and then fucking up this badly

sometimes it's easy to forget that people depend on your code, and if you fuck up, there are serious repercussions

this is why Haskell is the future.

code will be responsible for more and more IRL effects like spending your money, driving your car, cooking your food, or detecting cancer in your anus. in these situations it's important to have a language that can safely manage side effects.

also Haskell's type system (and coming soon, dependent types) can heavily reduce the risk of logic errors like that which caused this banking catastrophe.

>>2
software already drive cars, cook food and detect cancer in anus. It's all written in C and C++ and it's not going to change in the next 50 years.

>>3

And that’s horrifying

>>3
GHC can produce machine code almost as fast as from C++.

I find it very hard to believe that in 2060 we'll still be using absurdly low level languages like C. all the legacy code will probably be autotranslated to better languages.

>>5
don't forget to read the small print:
"almost as fast as C++ for some specific operation, not taking into account the garbage collecting which add random and unpredictable time"

>>5
C was bad in the 1970s, but you are not allowed to know that. Even basic data structures like arrays, unions, structures, numbers, and strings have been degraded by C, with no intentions to ever make them better. Half a century of knowledge has been thrown out because it made C look bad and the new books have been worded to discourage you from ever knowing this knowledge existed. Groups like ISIS and C programmers do this because what has been done before can be done again. They are afraid of the mere existence of books and artifacts because it is proof that the past was better than they want you to believe it is. By eradicating knowledge of the previous existence of something, they can make you believe whatever fits their narrative, including that things that once physically existed are not possible. If it was up to these C programmers, we would still be using C in 2060 and everyone who made other languages in the past 110 years would not be remembered and the truth that better languages existed would not be permitted to be known.

C++ shows no signs of being replaced by anything.
Turns out "rewriting it in Rust" is harder than it looks.

>>8
What about D?

>>8
Rust isn't enough of an improvement on C++ to justify the effort tbh.

>>9
People didn't like it.
1.No preprocessor. It was replaced by some template string thingy no one bothered to use.
2.GC everywhere. Last time i heard they struggled to remove GC from the standard library.
3.Tons of bugs in "reference" compiler, that is slower than all other compilers.
Its still not as bad as Rust though and you can replace C++ with it in small projects.

bluepilled normie cuck: why my academic language is not used instead of the fastest language possible?