Hacker News Reader

Comment by djjaxe | Hacker News Reader

I read the article is does not mention speed, performance once, in which it had nothing to do with what I stated. I was simply stating that higher level languages will cause the library to be slower also less easy to be used by other high level languages like python.

View on HN

The author has several articles about ATS (http://bluishcoder.co.nz/tags/ats/) and from what I've read, it outputs C code that is proven (for the parts that are in ATS) to be correct. There's a bit more detail in "Safer handling of C memory in ATS" (http://bluishcoder.co.nz/2012/08/30/safer-handling-of-c-memo...) and the end of the article contains some generated C code.

I do not understand why people keep pushing unsafe code when computers keep getting faster and we have more and more headroom (cpu, memory, bandwidth). There is no excuse to keep running unprovable crypto.

Maybe because most of the code currently out there being use by the biggest companies in the world still use these "unsafe" languages. & tons of the job market still is in these "unsafe" languages.

I think you should read the article again. The language in question isn't higher level really it just has compile time type checking, which has no overhead.

That's still not my point. At the time of starting openssl I don't believe that ATS was around. In any case my point is that back then C lang was the best choice for performance and still is revered as the "fastest" as `nearly` all other languages are written on top of it either directly or indirectly. In any case I would love to see someone tell all of the openssl community to just drop C and switch to a different language.

C being loved in the UNIX community is one of the primary reasons that these libraries are in C.

Ada has been around since the beginning of the eighties, has and had performance that is near that of C, does not use a garbage collector, provides C linkage, and is far more safe than C.

If you do allow garbage collection, there were many performant and safe alternatives in the 90ies, such as ML.

It's culture as much as performance.

Ada is a higher language weather or not it has linkage to C or not. The UNIX community cares about performance, performance, performance.

'Higher level' doesn't necessarily say anything. Rust and ATS are both higher level than C, but they can both do everything that C does.

Is Ada less performant than C? I know it has bounds checking, but that can be turned off for "shipped" software. Does it have some features that incur a runtime cost and that can't be disabled?

Ada with all the runtime features left on is slower than comparable quality C. It's faster than most languages though.

With all the runtime turned GNAT can/should produce code with in a percent or two as fast as GCC (they share the same backend).

And Ada has a thing called SPARK which is a set of compiler checks to formally verify your code so you can provably turn off those runtime features safely. https://en.wikipedia.org/wiki/RavenSPARK

So rust & ATS & ADA & higher level languages can modify memory space? That I am aware of most higher level languages stray from being able to modify memory space on purpose as it's dangerous but, someone has to do it for the operating system is all I am saying about low level now that we are completely off topic here.

Don't know about Ada, but Rust and ATS can. The Rust code would need to be written in an unsafe block in order to as freely modify memory as C, but regular Rust can still do a lot without requiring automatic memory management, safely.

As you can see from the article (not that you seem to have read any of it), ATS can express C, and optionally prove low-level stuff about it.

Reply on news.ycombinator.com