[beambot]

I find it ironic that your examples were single-character roman & greek letters. Why not Cyrillic? Phoenician?

I wouldn't use pictograms everywhere, but they could be elegant in certain contexts. E.g. having a dedicated pictogram for various base physics quantities (e.g. for mass, length, time, etc).

[qsort]

Because the latin and greek alphabets are widely known and understood, at least an order of magnitude more than Cyrillic and at least seven orders of magnitude more than Phoenician.

This is like asking why I'm writing in English even though it's not my native language, I do that because almost every educated person understands English, while almost nobody would understand if I wrote in Italian.

Pictograms would be difficult to standardize and hard to reproduce, unless you commit to a small set of them, which would just become an alphabet in disguise.

[beambot]

I appreciate the Western convention, but a significant fraction of the world uses pictogram-like written languages (e.g. Chinese). Using pictograms to represent words or concepts isn't that far-fetched.

[alexvoda]

Precisely. As an analogy, the Julia programming language allows Unicode variable names.

https://docs.julialang.org/en/v1/manual/variables/#man-varia...

[codethief]

Yes, and there is a reason why Chinese is so hard to learn: There are so many characters that learning them takes years.

[alexvoda]

Chinese being difficult to learn has nothing to do with the current discussion. And the wealth of symbols of chinese script is beneficial to mathematical notation where names are forced to be single character. They are descriptive and diverse which cannot be said about latin or greek letters. But they are foreign to westerners.

The real issue is forcing names to be of single character lenght.

[codethief]

> Chinese being difficult to learn has nothing to do with the current discussion.

It has everything to do with the current discussion. GP was suggesting that a "pictogram-like written language […] isn't that far-fetched". And I was countering that argument by saying that a large alphabet (a large set of pictograms) is hard to learn or at least takes a lot time, as can be seen from trying to learn Chinese.

That time is much better spent wrapping one's head around the math itself.

[alexvoda]

> And I was countering that argument by saying that a large alphabet (a large set of pictograms) is hard to learn or at least takes a lot time, as can be seen from trying to learn Chinese or oher asian scripts.

And my counter to that is that the same way you do not learn greek in order to use some greek characters in math, the same way you do not learn chinese in order to use some chinese characters in math.

There are cases where "Δ" read as "delta" makes perfect sense for the delta between two values. Whereas "η" read as "eta" is not a obvious synonym for efficiency, or for the coefficient of viscosity, or for the metric tensor in Quantum Field Theory. And "c" read as "c" is not an obvious synonym for speed of light, or for 100, or for the space of convergent sequences. But "光" read as "Guāng"(chinese) or "hikari"(japanese) or "light"(english) might be obvious for something related to light. You do not need to learn the original pronunciation of the character to use it. Just have a 1 to 1 mapping between notation and notion. Same way Chinese and Japanese have completely incompatible pronunciations for the same characters.

And even better, just use words instead of characters.

[CJefferson]

I'll willing to bet a significant chunk of the "emoji" alphabet is better known and understood than the Greek alphabet nowadays.

It is true you would have to stay in some reasonable-sized subset, and I can't begin to imagine the arguments if someone tried to define "common emoji" or something..

[codethief]

> E.g. having a dedicated pictogram for various base physics quantities (e.g. for mass, length, time, etc).

What if you have two different lengths, one being the distance between two objects and the other being the position of the center of mass over time? What about their two masses? And, in relativity, their two proper times?

Now you're back to using subscripts and the Latin alphabet because what better way to describe the precise object you're talking about in a given language (here: English) than the language itself?