Hacker Newsnew | past | comments | ask | show | jobs | submit | gottheUIblues's commentslogin

Rather COBOL is a living fossil? And today's Fortran is the FORTRAN family with horizontal gene transfer from the Algol lineage of programming languages.

Both languages have their standards updated still, latest year in both cases was 2023.

Fortran is one of the reasons OpenCL lost to CUDA, and now even AMD and Intel have finally Fortran support on their own stacks, not Khronos based.

https://developer.nvidia.com/cuda-fortran

Whereas Cobol, even has cloud and microservices.

https://www.rocketsoftware.com/en-us/products/cobol/visual-c...

https://aws.amazon.com/mainframe/

Incredible how being monetary relevant keeps some languages going.

Also note how the beloved UNIX and C are from 1971 - 73, only about 10 younger than COBOL.


> Fortran is one of the reasons OpenCL lost to CUDA, and now even AMD and Intel have finally Fortran support on their own stacks, not Khronos based.

FWIW, I loved using CUDA-Fortran. I think the ease of use of array variables maps very well with the way CUDA kernels work. It feels much more natural than in C++.


Can COBOL be called a living fossil?

I mean, programming languages do not live; and they do not "die", per se, either. Just the usage may go down towards 0.

COBOL would then be close to extinction. I think it only has a few niche places in the USA and perhaps a very few more areas, but I don't think it will survive for many more decades to come, whereas I think C or python will be around in, say, three decades still.

> family with horizontal gene transfer

Well, you refer here to biology; viruses are the most famous for horizontal gene transfer, transposons and plasmids too. But I don't think these terms apply to software that well. Code does not magically "transfer" and work, often you have to adjust to a particular architecture - that was one key reason why C became so dynamic. In biology you basically just have DNA, if we ignore RNA viruses (but they all need a cell for their own propagation) 4 states per slot in dsDNA (A, T, C, G; here I exclude RNA, but RNA is in many ways just like DNA, see reverse transcriptase, also found in viruses). So you don't have to translate much at all; some organisms use different codons (mitochondrial DNA has a few different codon tables) but by and large what works in organism A, works in organism B too, if you just look to, say, wish to create a protein. That's why "genetic engineering" is so simple, in principle: it just works if you put genes into different organisms (again, some details may be different but e. g. UUU would could for phenylalanine in most organisms; UUU is the mRNA variant of course, in dsDNA it would be TTT). Also, there is little to no "planning" when horizontal gene transfer happens, whereas porting requires thinking by a human. I don't feel that analogy works well at all.


I don't know; COBOL pops up in banking all the time and Fortran lives on in a lot of specialised engineering applications.

Is there anything to be gained from rewriting it in Rust? Five years ago would there have been anything to be gained from rewriting it in Haskell? Ten years ago would there have been anything to be gained from rewriting it in Ruby? Fifteen years ago would there have been anything to gain from rewriting it in Clojure? Twenty years ago would there have been anything to gain from rewriting it in Java?

And so on, all the way back.


Analogies are inherently reductive; by deeply scrutinizing the differences you do not prove the analogy is useless.

Err? Peano Arithmetic is provably consistent in ZFC, but it is not in itself (if PA is consistent). Therefore if PA is consistent it is not equivalent to ZFC (regardless of whether ZFC is consistent or not)


I am referring to this slide : https://youtu.be/EVwQsvof7Hw?t=1646


"Embarrassingly" considered harmful?


"Embarrassingly" considered harmful is all you need.


Programming Introduction to "Embarrasingly" considered harmful is all you need in 21 hours.


Cringeworthily parallel, not even serial


I don't think of 9 as really being a signal to the process at all, more of an instruction to the OS kernel to terminate the process


If that theory holds - have to ensure that the models have not been trained on any code that is licensed incompatibly with the GPL, in which case the models could not be distributed at all


So specified .. that it can actually prove it can't be completely specified by any single specification


All mathematical statements we care about fall out of the purview of incompleteness


To the contrary (as summarised by Gemini):

Gödel showed that arithmetic cannot prove everything about itself.

Turing showed that computers cannot predict everything about themselves.

Rice showed that we cannot automatically verify what programs will do.

Chaitin showed that mathematics is full of random, unprovable facts.

Lawvere showed that they are all failing for the exact same structural reason!

These are not fringe issues. They define the absolute boundaries of human and machine intelligence.


Right so strictly speaking C++ could do anything here when passed a null pointer, because even though assert terminates the program, the C++ compiler cannot see that, and there is then undefined behaviour in that case


> because even though assert terminates the program, the C++ compiler cannot see that

I think it should be able to. I'm pretty sure assert is defined to call abort when triggered and abort is tagged with [[noreturn]], so the compiler knows control flow isn't coming back.


Shouldn't that be the "dam spelling" then?


Film score composers are quite famous for borrowing from 12 tone serialism - quite a bit of discussion on it available by Googling or using your favourite chatbot


Money is a social construct, not some kind of physical quantity subject to conservation laws, and can be and is introduced into the economic system all the time. The real question is really would introducing more money or a UBI cause social disruption by e.g. disrupting price signalling by high inflation or changing incentives to work so less goods and services that people actually value are produced.


One important function of money is allocation of resources, and if a society is bad at allocating resources, it declines.


It is a social construct but if you just print money you get ... inflation. You can't just increase money supply to redistribute wealth without consequences.


Consider applying for YC's Summer 2026 batch! Applications are open till May 4

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: