The ASCII character set standardization and subsequent keyboards has had a huge impact on programming languages. Characters not found on a standard keyboard are hardly ever used. And if thy are, it cause lots of issues, because they are so hard to type.
I hope we get cheap reconfigurable keyboards with oled key caps in the near future. It could revitalize programming language design using a more powerful syntax with symbols etc.
People will use something like |> and make a ligature for it in the font, that looks like this ▷. and that's it. However, have you seen this: https://fluxkeyboard.com
In some equipment, like the Teletype Model 33[1], you had left arrow and up arrow instead of _ and ^. This was part of the 1963 draft of ASCII, but was changed in the final version. In the early 1980s there was still a lot of equipment by DEC and Xerox stuck with the original version, which was what the creators of Smalltalk-80 were familiar with.
Having lived part of the early history, I'd counter that this was a very good thing indeed. ASCII provided some sanity and allowed ideas to spread and proliferate.
We can revisit the character sets now that Unicode is ubiquitous. As for typing non-ASCII characters, that has been daily practices for decades _outside_ of north America.
Eh, wait. Maybe for comments and strings, but we, the non-ASCII language users such as Spanish, avoided to put ñ's and accented chars as _code_ almost as a religious dogma.
Tildes in Spanish are just used to mark the stressed syllabe when it's outside the stressing rules, and the diaeresis it's to make the 'u' non silent in gue/gui/que/qui. So if we read "funcion" without being written "función", don't worry, we aren't writting a literary test, it's code.
I'd hate to debug code written in Chinese instead of English, even if English it's a language I just use in academical and technical environments (and some nerdy games translated from Japanese or classical retro games from Unix workstations, playing either Trek or Slashem it's not reading Oscar Wilde or Shakespeare). And Golang allows that I think, and lots of languages too.
Cool stuff.
What's funny is that when learning Smalltalk, I got used to proportional fonts for programming, but only in a Smalltalk environment. I still find it odd to see that in another editor, and I have no idea why.
I actually replaced my programming fonts with serif fonts in all languages after a few years in Smalltalk. It’s not quite as good in a brace language like Java, but for Python it works really well, I find.
The smalltalk interface is really one of the cleanest interfaces ever created. Too bad the software stack hasn't been put together in a complete desktop even like a simple one such as TempleOS
There is the more recent CogNos/NopSys [1,2] "an evolution of the SqueakNOS project" "a Smalltalk virtual machine running on bare x86 hardware", but that project has also been inactive for several years.
Readit News
Let's not forget that in the early 80's, more than 8 bits per character was something that almost never happened.
It was a real PITA because it was underbar (_, shift-dash) on the keyboard iirc.
Any assignment made me want to drink, thus the character name.
[1] https://en.wikipedia.org/wiki/Bit-paired_keyboard
We can revisit the character sets now that Unicode is ubiquitous. As for typing non-ASCII characters, that has been daily practices for decades _outside_ of north America.
Tildes in Spanish are just used to mark the stressed syllabe when it's outside the stressing rules, and the diaeresis it's to make the 'u' non silent in gue/gui/que/qui. So if we read "funcion" without being written "función", don't worry, we aren't writting a literary test, it's code.
I'd hate to debug code written in Chinese instead of English, even if English it's a language I just use in academical and technical environments (and some nerdy games translated from Japanese or classical retro games from Unix workstations, playing either Trek or Slashem it's not reading Oscar Wilde or Shakespeare). And Golang allows that I think, and lots of languages too.
[1] https://www.lg.com/us/monitors/lg-28mq780-b
1. https://github.com/nopsys/CogNOS
2. https://charig.github.io/assets/papers/SCDE-DLS.pdf