One-letter programming languages | Hacker News Reader

One-letter programming languages

computerworld.com

first thing I noticed: missing E! https://en.wikipedia.org/wiki/E_(programming_language)

Looks like [1] the only letters not yet taken are: H, I, N, O, U, V, W, Y, and Z? Changes depending on whether you're allowing characters.

[1] https://en.wikipedia.org/wiki/List_of_programming_languages

Somebody once had a hobby project of a stack-based language called V. Can't find it now.

I don't think anybody could use Z, since the Z formal notation is too well known.

Zork games not use a virtual machine called Z ?

H http://dcoj.wmh3.com/cscos/h/

U https://github.com/upcrob/u-programming-language

W https://www.vttoth.com/CMS/projects/49-w-a-simple-programmin...

Y https://dl.acm.org/citation.cfm?id=954278 and https://esolangs.org/wiki/Y_(programming_language)

Z notation and also Z++ https://en.wikipedia.org/wiki/Z%2B%2B

A tiny language called Z: http://chrisdone.com/z/

I really wish modern language creators would take googleability into account when naming their languages. Even with rust, I occasionally have to use "rustlang" to avoid result pages with mostly plumbing sites, I imagine the situation is far worse for D, F and M.

Use "D language" and it googles no problem.

Google does good here, but I have to feel like there's a certain threshold of popularity for every name to be effectively googleable. Obscure names can get top google results easily.

Does anyone remember what that was like when the .NET framework first came out? It's literally just a TLD..

To this day I wonder why they chose .NET I can only think of Passport.NET which is likely pointless by now.

It was all about showing how network enabled and service oriented your were.

Back in those days people were still pushing XML web services, SOAP and UDDI like it was the goodest thing possible.

.NET was about showing how everything .net was all those things, kinda.

Basically marketing meets tech hype 101.

I'm pretty sure that at that time Google removed "#" characters from search queries, so a search for "C#" was just as fruitless...

And Google itself named a language "go".

I was expecting the "web naming industry" to go through a full cycle of trends to stay unique and googleable. Trademarks, open-source libs, product names, languages, I was expecting that more and more strange names would become acceptable:

- random strings (xkcd) - longer as all 4-, 5- and 6-letter domains are reserved

- names (Levenstein distance) - More and more strange names, perhaps invented

- patterns (.ly, .io) - But that gets crowded very fast

- composed names: "Diffie Hellman"

- 3-part composed names: Will that be seldom in 2020 when all 10-letter domains are taken?

- Even a small sentence, as long as it's unique enough for Google to identify it.

But it seems I'm not good at predictions: Companies keep recycling the same names, the domains expensive, and VCs don't like founders who didn't put enough thought in their brand.

Tangential, the band "Chvrches" chose their name specifically because it was easy to search for. And indeed, search engines never get confused.

It's not just languages. Apple named their word processor "Pages" and their spreadsheet "Numbers".

Surprisingly, these both give useful results in Google. The Pages app is at the top, and the Numbers app is only four down (after "Number" on Wikipedia and a couple of results for the Numbers/Numb3rs TV show).

I suspect some hand-tweaking on Google's part.

the worst language i've seen for that was "clean" - it's so generic a term that even other programming languages use it to describe themselves!

My personal favorite was trying to search for things related to the IO language, being a generic computing term as well and a fairly obscure language too.

For "D" you should search "Dlang"

Whenever I try to search for graphic programing related issues with Rust, there is a game with the same title that fills up the result pages.

As you say, use "rustlang". The same works for golang, dlang, nimlang, and probably a few others.

But I agree Googlability is a problem. C# or F# are really hard to google for example. It's hard with C too since clang refers to something else

csharp and fsharp are just as effective as *lang in my experience

Rust also clashes with a video game named Rust. I tried to check the Rust Reddit yesterday but ended up at the video game Reddit. Fortunately, the game will soon fall into obscurity while Rust lang grows in popularity.

"a manager counted the lines of software coming out of the cubicle farm and determined that programmers wrote N lines of code a day. It didn't make a difference what language was used -- the company would get only N lines out of them. The manager promptly embraced APL, the tersest, most powerful language around"

They can only do N lines, so the natural decision was to use terser languages, I love it.

I mean, yes there are reasons to use a special language for a problem (for example Elixir instead of Ruby, for distributed systems). Also sometimes some problems get solved more elegant in newer languages (Rust instead of C++ for system programming).

But if the only metric is lines of code, every language solves this with enough abstraction layers...

Just the opposite. If you figure out the number of lines of code is constant, like the alleged manager, you can start taking it as a constraint and optimizing for other things.

It's odd that F# is covered, but not C#, Objective-C, or especially C++. It was especially odd to read the description of D without C++ being named. The inclusion of RAII in D is a dead giveaway of its inspiration.

The article mentioned C++ and Objective-C under the C language, and C# under F#. They may have done well to mention C++ and maybe Java's influence on D. They completely neglected E too.

C# = 2 (letter + symbol), Objective-C = 11 (10 letters + 1 symbol), C++ = 3 (1 letter + 2 symbols) so its not odd :-)

It does say 1 letter languages not 1 character languages. But then if they were included, there'd be a comment on how C# is two letters, a discussion whether # should be counted as a letter or not, which really is one of these "look, we're arguing axioms now" discussions.

Then why was F# included?

I think they wanted to post "F" but posted "F#" :-)

https://en.wikipedia.org/wiki/F_(programming_language)

Not to be confused with F Sharp (programming language).

https://en.wikipedia.org/wiki/F_Sharp_(programming_language)

The author also didn't get the history of C right. 1. it wasn't Dennis and Brian who wrote UNIX but Ken and Dennis. 2. B wasn't an internal language of AT&T but a language designed by Ken. 3. 16-bit words weren't immense at the time...yes, the PDP-11 had addressable bytes (unlike any other PDP) but I don't understand what this had to do with 16-bit words being "immense". 4. It was only Dennis not Dennis and Brian who developed B into C.

To add to that point. The following quote is from an old version of dlang.org[1]

> Can the power and capability of C++ be extracted, redesigned, and recast into a language that is simple, orthogonal, and practical? Can it all be put into a package that is easy for compiler writers to correctly implement, and which enables compilers to efficiently generate aggressively optimized code?

Although this has slightly differed now, one of D's core objectives was to make a better C++ from the ground up.

[1]: http://web.archive.org/web/20120111102019/http://dlang.org/o...

Missed out q!

https://en.wikipedia.org/wiki/Q_(programming_language_from_K...

I am sorry, those are programming languages whose name has one letter.

For a true one-letter programming language please check SuperPython http://search.cpan.org/~mjd/SuperPython-0.91/SuperPython.pm.

Btw. does anybody know what happened to kOS or z? I can't find anything on kparc.com anymore.

I think they regretted saying anything about kOS early on, I think it won't be mentioned until release.

KxCon2016 is coming up May 19-22. Arthur Whitney is scheduled to give a presentation on K6. You might try asking on the Kdb+ Personal Developers Google group afterwards.

Thanks!

"Long ago, when Brian Kernighan and Dennis Ritchie, aka K&R, set out to write Unix".

Unix was created by Ken Thompson and Dennis Richie.

A little unfair that they write the K program out when finding primes smaller than a number is a J primitive.

Reply on news.ycombinator.com