sitelen pona is a logography, pretty much as basic as they come. Each word has a gliph and you put them in the order you say them, left to right and top to bottom because of course we're only using the dominant one even though li and e (gliphs that appear in nearly every non-simple sentence) can literally be used to point the direction. Well, sucks for us free directionality fans.

There are only two noteworthy features of this logography: Its (bad) approach to loanwords and its (worse)compound gliphs.

The loanwords are dealt with a (non-standardized) system where you put inside of a box gliphs of words that each represent their first letter. Example given in pu: Kanata (Canada) can be written as [kasi alasa nasin awen telo a]. Yes it takes six times the space of a regular word and each of the three As is a different letter. This is not a bug, it's a feature and shouldn't. This system isn't even followed by everyone, with some really weird implementations by some people... (looking at you, jan mi-sona-ante-li.)

Meanwhile, the compounds are made shoving a gliph inside another gliph to write two words in a single slot! Despite the fact some regular words' gliphs are formed like that... Yes that's all there is to say about the compounds, which I am pretty sure only exist to make toki pona's symbol one thing instead of two.

One of the bigger problems you will encounter trying to figure out sitelen pona is telling gliphs apart, doubly so if you're dyslexic.

This section will focus on jan Sonja's bad habit of making distinctions solely by rotation and mirroring, and the next on other difficult distinctions.

awen, kama and tawa

mi, sina and ona

mi is the number 9, sina is the number 6 and ona is σ but upside down. They're supposed to be pointing hands, pointing at the relevant person: mi is the writer, sina is facing them, and ona is to the side. Of course, this isn't good for dyslexics. Also, this analogy isn't very good, since it breaks whenever you write text that isn't flat in front of you.

The personal pronouns are notably difficult for logographies, so this is close to excusable. My solution was to start with the jan (person) gliph for mi and sina, but swap out the head for uta (mouth, for the speaker) and kute (ear, for the listener) respectively.

monsi and sinpin and the other direction gliphs

All the direction gliphs (except that one) are built with a box and a dot. All the spinning of the box and relative positions of the dot makes them annoyingly difficult to tell apart, and a big dyslexic hazard.

To fix this, I used a cross (X) to mark location gliphs, and combined them with a relevant word (usually a body part):

• sewi (up) uses lawa (head);
• poka (side) uses luka (arm);
• anpa (down) uses noka (foot);
• insa (inside) uses poki (container);
• sinpin (front) uses uta (mouth);
• monsi (back) uses ike (bad, because there's no gliph for butt. I originally used jaki (nasty), but they got similar (see Part 4));

lon (exist, at, be) and meso (middle) stay the same except for changing their dot to the cross, since they're pretty self explanatory.

ni and epiku

ni, meaning "this", is an arrow pointing downward. epiku, meaning "epic", is an arrow pointing upward.

This one in particular really annoys me because epiku is a ku suli, a word that got added after the original book and somehow is recognized by enough people to become part of the must-know vocabulary. A word that literally just means "very good", in the not many words language.

To fix this epiku is now the pog face. It is a joke of a word so it gets a joke of a gliph.

Exception: ike and pona, lupa and nena

a and o

Both of them are exclamation points with fat dots. And only one of those has a tail. Excluding that, they're the exact same.

They appear again in Part 3 with some related gliphs, so I'd rather wait until we get there.

ala, ante and weka

en, sin and namako

ike and lupa, pona and nena

All of these are literally just arches except lupa/nena are vaguely longer. The second pair means hole and bump, so I stole sitelen pona pona's fix and added a ground line to them, making them similar to Ω.

lape, lawa, mi/sina/ona and misikeke

len, lipu, selo and sijelo

lili and suli

"Let's differentiate two gliphs on size and literally nothing else", said a linguist that does not know jaki about scripts.

nasin and epiku

Both are upward arrows, but epiku's is normal while nasin's has its tip in the middle. What were the people making this sitelen sin thinking?!

nimi, lipu and palisa

While I am aware that toki pona is not remotely attempting to represent every language in the world, it gets lazy with eurocentric gliphs and (nearly) only eurocentric ones, with languages like Mandarin getting no such treatment.

a, o, kin, n and too many nimi sin

ala

ma

moli

nanpa

seme

sewi

ken and wile

jan Sonja, my sister in Ra, why are these gliphs literally just the latin letters k and w? This is unexcusable laziness! At least you tried with the exclamation marks... It doesn't even make sense to have specifically these two be like this, they have nothing in common except being preverbs and the other preverbs are fine! (well, they're not, but they have actual pictograms at least...)

To fix wile, I turned it into an upside-down pilin with a dash through the middle, resembling an arrow or the spades suit. (♠️)

Exception: math inspired gliphs (wan, ali, en, kipisi, sama)

Other exception: pilin (but not olin)

An absurd amount of gliphs but especially lili, pona and ike: not space filling

Most toki pona fonts have the gliphs take up a squarish space. Despite that, many gliphs don't take up the entire square, making writing by hand more unintuitive than it needs to. This happens very often in the original book's chart, but fonts like linja pona fix most of those.

Except lili, which is literally designed to not fill its square. Why.

lili's new groundline is wide enough to fill up the square, and pona and ike can be written wide enough for that now that differentiating from lupa and nena isn't an issue, so this problem got fixed while I was on the other problems! Neat.

ko and jaki: not standardized

meli and mije: gender presentation (featuring tonsi)

These are a big problem. Gender categories are a fundamentally cultural thing, and meli and mije are not only biased towards the western view, they're incorrect: you can find cis women with broad shoulders and cis men with long hair, and don't even get me started on trans people.

These gliphs are an interesting problem, as when divorced of culture all that remains is the reproductive meaning, yet I think it would be crude to use genital pictographs for this purpose.

kala, sewi, tonsi and kin: breaks off from a pre-established pattern

akesi and waso: pictogram too abstracted

kalama and toki: unintuitive origins

One of these is a mouth with emitters and the other is a thing with emitters. Guess which got assigned to mean talk.

...

Of course, thing-emitter is talk, and mouth-emitter is noise! What else could it be!?

I don't even know how to quip about this one, it's such an obvious oversight it hurts my head! I undid this obvious swap for sitelen Kaku.