Cistercian Numbers

86 points by debo_ a day ago on hackernews | 15 comments

tangus | a day ago

My minuscule pet peeve is that having only one source where the number 5 is depicted with a triangle (all others show it as a separated segment, like the number 6 but shorter), that's how every article or library draws it. It's all because the guy who wrote a book about them saw that source first so he based his figures on it.

Here's a small summary about the numbers with many examples: https://www.unicode.org/L2/L2020/20290-cistercian-digits.pdf

[OP] debo_ | a day ago

It would never have occurred to me that anyone would want to get these into a Unicode standard. This document you linked is excellent, thank you.

bobbiechen | a day ago

Being first matters :')

I wrote a font for these, which does use the triangle-5 and the vertical layout: https://bobbiec.github.io/cistercian-font.html (recent discussion here: https://news.ycombinator.com/item?id=46939312)

And my associated writeup: https://digitalseams.com/blog/making-a-font-with-9999-ligatu... .

As mentioned in the blog, I think the horizontal layout makes more sense too (in terms of writing order). But just like the triangle-5, the vertical layout is more commonly seen, so that's what I stuck with.

autoexec | a day ago

It might not be accurate but it does seem like it'd be easy to mistake a 5 and 6 without the triangle. Especially when the characters are being hurriedly written by hand. If I were going to use this system, I'd be sticking with the triangle.
I wish the 6 was a triangle in the other direction instead

jhncls | 23 hours ago

In a Numberphile video [0], Alex Bellos also uses a triangle for 5.

[0]: https://www.youtube.com/watch?v=9p55Qgt7Ciw

klondike_klive | a day ago

Wow, it's a while since I've seen one of those lists of hundreds of vampires that you have to deselect!

dcanelhas | a day ago

Shouldn't 523 in that list of "other numbers" actually be 522?

poulpy123 | a day ago

You're right

beratbozkurt0 | 21 hours ago

Actually, it would automatically translate when we scan it with the camera, like Google Translate.

onirom | 21 hours ago

did a small raw binary-like implementation of an extended (base-16) version : https://www.onirom.fr/wiki/snippets/#JavaScript_:_Extended_C...

somat | 20 hours ago

I am a little sad that bare zero is not represented. This is my first exposure to Cistercian numbers but it looks like a unadorned staff would fit for the bare zero.

The whole thing is a lot of fun, feels like a Myst puzzle. Or more accuratly, I don't think Myst had a number puzzle but Riven did and I recently picked up Obduction and it had one, So probably fairer to say a Cyan type puzzle as they appear to love creating wierd numeric representations.

iguana_shine | 19 hours ago

It's pretty cool but I'm surprised there wasn't confusion among the 2s (2 & 200, 20 & 2000) all the other symbols had enough to make it clear which side they're on, but the closer the 2-notch gets to the centre, the more ambiguous it gets. Could even be confused with a 1 if you're not careful

hackernj | 18 hours ago

Two OEIS sequences for Cistercian numeral system: https://oeis.org/A341737 and https://oeis.org/A381327

perilunar | 9 hours ago

Two odd things strike me:

1. 5 is the symbols for 4 and 1 combined, 6 is new, then 7=6+1, 8=6+2, and 9=6+3. It seems to me it would be more obvious to have 1 to 5 be unique symbols, then 6=5+1, 7=5+2, 8=5+3, and 9=5+4. Like how we do with tally marks.

2. The system is really just 4 digits combined in a square. I don't see the advantage over Arabic numerals.