Yingzi Modern

[Emily]

heyyyyyyyy

so i was re-reading zompist's yingzi page again, and i got to the part at the end where it says "Four milennia have reduced the pictorial content of the hanzi primitives almost to nil. What the 'pictograms' are pictures of is often evident only to the scholar." presumably if yingzi were adopted we would expect a similar thing to happen over time as well. so i took the time to take some reasonable guesses as to what this would look like. below are "modern" versions of all the words/signs used in the yingzi article:

simple characters:

these are arranged in stroke order (strokes counted in the "modern" version of the glyph, not in the original versions on the yingzi page). "one" through "bean" have one stroke each, "vertical" through "see" are two, "work" through "sun" are three, "tree" through "toad" are four (this is a reduction of sixteen strokes from the original version of "toad"!), "east" through "mouth" have five, "bill" through "gang" have six, and "guilt" through "king" have seven

compound characters:

these are the signs composed of radicals and phonetics, arranged in the radical's stroke order

multi-character words:

and these are the words listed in the article that consist of more than one character

i had some trouble with relative line thicknesses (they're supposed to all be the same thickness) and getting some of the ends and corners in the angled characters right (trigonometry is hard) so let's just call this a draft. still, it was fun. i wish there were more characters!

[Emily]

"One man fell in Peking"


"Toads fight bugs in gangs"

[So Haleza Grise]

To quote someone, pretty scripty!

[zompist]

Very nice indeed!

Which reminds me, I should really upgrade the graphics on that page into at least the 21st century.

[Moose-tache]

This is really cool!

But I'm wondering, how would glyphs mutate over time? In the case of Hanzi, and every other writing system, the simplifications were informed by medium: brush strokes get connected or twisted or omitted to make writing quicker and easier; carvings get straighter; calculator numbers are designed to use as few "pixels" as possible. So if Yingzi were adopted and we run the clock for a couple thousand years, what is the medium used during that time? If it's exclusively digital ("Those dang-nab Millennials!"), then ease of writing is no concern at all. The main concern would be keeping the glyphs distinct at a small display size. The easiest way to do that I can think of that ignores human drawing ability is to maximize the number of basic shapes. Instead of square-or-circle-or-triangle, you could have glyphs that are pentagons, diamond shape, half-circle-half-triangle. These would be a nightmare to keep distinct if you were writing by hand, but in a digital format it's pretty easy to tell them apart.

[bradrn]

This is really cool!

But I'm wondering, how would glyphs mutate over time? In the case of Hanzi, and every other writing system, the simplifications were informed by medium: brush strokes get connected or twisted or omitted to make writing quicker and easier; carvings get straighter; calculator numbers are designed to use as few "pixels" as possible. So if Yingzi were adopted and we run the clock for a couple thousand years, what is the medium used during that time? If it's exclusively digital ("Those dang-nab Millennials!"), then ease of writing is no concern at all. The main concern would be keeping the glyphs distinct at a small display size. The easiest way to do that I can think of that ignores human drawing ability is to maximize the number of basic shapes. Instead of square-or-circle-or-triangle, you could have glyphs that are pentagons, diamond shape, half-circle-half-triangle. These would be a nightmare to keep distinct if you were writing by hand, but in a digital format it's pretty easy to tell them apart.

I would say that some of those shapes you listed would also be a nightmare to tell apart at small display sizes — more so than normal human scripts! For instance, I did a couple of experiments just now, and I can keep Latin letters readable within a maximum bounding box of at most 4×6 pixels per letter, whereas a pentagon already starts looking a bit too un-pentagony for my liking at 9×9 px, and becomes impossible to draw below about 7×6. The other shapes you mention share the same problems, albeit to a lesser extent — it simply becomes impossible to easily draw these sorts of convex shapes in a distinguishable way below a certain size.

(Admittedly, I don’t have any better ideas. I think it depends on whether you optimise for larger or smaller display sizes. As I already said, shapes such as you suggested would be a nightmare if you want to ‘keep the glyphs distinct at a small display size’, as you originally aimed to do, but they would be perfect for a larger display size. On the other hand, if you want to optimise for a smaller size, you would want a script — preferably alphabetic, to minimise the number of letters you need to keep distinct — which works off pixels, a bit like Dotsies or Pixel Script.)

[zompist]

which works off pixels, a bit like Dotsies or Pixel Script.)

OK, interesting idea, but I don't like either realization. I see that pixelscript uses a 3x3 grid (M doesn't fit), and it bugs me that they didn't make it work instead of inventing completely new signs for many letter. It's not that hard.

(BTW, there's no reason a pixel font has to be 1-bit. However, I tried anti-aliasing the font, and it really doesn't help with the more difficult letters.)

[bradrn]

which works off pixels, a bit like Dotsies or Pixel Script.)

OK, interesting idea, but I don't like either realization. I see that pixelscript uses a 3x3 grid (M doesn't fit), and it bugs me that they didn't make it work instead of inventing completely new signs for many letter. It's not that hard.

It says in the description that Pixel Script ‘is primarily … for decoration and signatures. Messages can be hidden within the pixel art …’. So presumably the creator deliberately avoided Latin letters to achieve this aesthetic.

(BTW, there's no reason a pixel font has to be 1-bit. However, I tried anti-aliasing the font, and it really doesn't help with the more difficult letters.)

I know. But 1-bit gives the highest amount of contrast with smaller sizes, always a good thing in a writing script.


(By the way, I didn’t get a notification when you quoted me. Any idea what might have gone wrong? Usually I find phpBB to be very good with notifications. EDIT Never mind, notifications are working again)

[Raphael]

Arguably, a form of Yingzi already exists - emoji. (Yes, of course they are not specific to English.)

[Emily]

This is really cool!

But I'm wondering, how would glyphs mutate over time? In the case of Hanzi, and every other writing system, the simplifications were informed by medium: brush strokes get connected or twisted or omitted to make writing quicker and easier; carvings get straighter; calculator numbers are designed to use as few "pixels" as possible. So if Yingzi were adopted and we run the clock for a couple thousand years, what is the medium used during that time? If it's exclusively digital ("Those dang-nab Millennials!"), then ease of writing is no concern at all. The main concern would be keeping the glyphs distinct at a small display size. The easiest way to do that I can think of that ignores human drawing ability is to maximize the number of basic shapes. Instead of square-or-circle-or-triangle, you could have glyphs that are pentagons, diamond shape, half-circle-half-triangle. These would be a nightmare to keep distinct if you were writing by hand, but in a digital format it's pretty easy to tell them apart.

good question! the mutations i made were developed based on handwriting: i would write a symbol over and over at higher and higher speeds and see what came of it, then regularize it, and only after i had done that with all of the symbols on that page did i turn them into something in a font program. but this (and to a certain extent the original yingzi page itself) sort of ignores the change in any given glyph's phonology and meaning over time. if we were to take yingzi as having been introduced in the present day, rather than several millennia ago, who knows what the primary medium for writing will be and how it would change them?

[Moose-tache]

The lack of spaces in both those scripts has grown on me since I first saw them. But I still doubt that they would be very readable at small scales. But I have to give credit to the Dotsies teaching sample: that is some brilliant presentation!

[Curlyjimsam]

Cool idea, and interesting results!

[Emily]

question for zompist: would the plural -es glyph also be used for the 3sg -es morpheme, or would there be a different glyph for that?

[zompist]

question for zompist: would the plural -es glyph also be used for the 3sg -es morpheme, or would there be a different glyph for that?

There's not enough Chinese equivalents to say 'correctly' (it's based on the -r morpheme). On the other hand, the glyph is obviously phonetic rather than semantic, so I think it could be used for both.