Hacker News Reader

Comment by 1_player | Hacker News Reader

Can I be extremely shallow? Because the language feels like it's from the 60s. Like writing Fortran or Ada or Algol in 2021. C is timeless, it doesn't count.

I love lisps, but Racket, for example, feels more modern than Common Lisp to me. And for shallow people like me, feel and ergonomics are very important when writing code.

View on HN

> C is timeless, it doesn't count.

Of course C counts. IMO it is a language that should have stayed in the last century.

> feel and ergonomics are very important when writing code

This is the reason why I can't stomach languages like Java, C++, etc.I feel like I'm fighting a very constrained interface.

C's succesor is either Go or Plan9/9front's C. Outside of "Unix 2.0", you have Go.

Please, do not compare Java and C++ against C, C++ is a monster it should never have born, and Java it's a shitty TCL "alternative" from its era.

If anything zig or rust are the successor, Go is a language from the 90s. We can agree on C++ and java, though.

Go is a rehash of Alef/Limbo. Is not from the 90's.

If you say so, we could say the same on Zig.

BTW, UTF-8 dates back to the 90's, good luck finding a better alternative. And TCP/IP it's even older, but it works.

Something being newer doesn't make it better.

When we're talking about strong type systems and classes of bugs eliminated just because the program compiled (a la OCaml, Rust, Haskell)... yes, newer is better.

Your argument isn't universal; something being older doesn't say much about its quality either.

Could you elaborate a few specific points on this?

Curious because I'm on the same boat.

I was annoyed for a while how ruby convention uses '?' characters at the end of method names that return a boolean.

I'm not even that old and I still want to yell `[a-zA-Z0-9_]+` identifier names or get off my lawn.

Why would you want to limit yourself so drastically in what you can express with function and variable names?

I don't think this is drastically limiting, it's a sane subset that gets you 99% of what you need from a function name.

We use punctuation and symbols in typography and writing to provide clarity. Why not use it in our programs?

Programs? Aren't we talking about variable and function names?

But for the general case, Lisp does use puncuation and symbols. It just tends to use less. The other side of the balance is operators like ~+#>. Most languages are in the middle ground. All of these options are fine and have their uses. I wouldn't like to see stuff like ~+#> every day, but I like being able to implement <.

Programs are composed of those things.

Sure, but you don't use a lot of punctuation in identifier names. For example, in JS:

    function sumOfArray(arr) {
        return arr.reduce((sum, element) => sum + element, 0)
    }

Now in Scheme:

    (define (sum-of-list lst)
        (reduce + 0 lst))

Both have punctuation, but usually not in the identifiers (except for - and _)

In fact we do, but as in typography you do not use punctuation to write words.

Then again, I really like '-' in symbol names.

I don't, I always follow whatever convention the langue/code base uses.

I'm just saying like GP I'm shallow and prefer C style identifiers where it makes sense to.

In Julia, the convention is that a function like "swap!" modifies its input. It's extremely handy. Also, "odd?" is just an elegant way to name a predicate.

I saw that convention in IBM/NAG's Scratchpad/Axiom computer algebra system which was designed in the 80s. It may go further back than that.

C was already pre-historic in the 1990's, of course it counts.

It's an IR for a never completed language.

Reply on news.ycombinator.com