The Fan Language of Splatoon

[PiyozR]

1. Derivative verbs are those that are nouns with {-kói}- at the end right?

Yes, they're already one section. Any verbs (if any) directly derived from a noun that don't really fit into a topic will go here.

2. I think the verbs should be grouped together with like themes (ex. "to bake" would go with all the food vocabulary). Thoughts?

They are. All these themes contain every relevant noun, verb and descriptor. That's the purpose of the new sections.

3. Should Descriptors be re-labeled as Adjectives to avoid confusion?

Seeing as there is no difference between adjectives and adverbs (in word form) in Inkling, the single word "descriptor" was chosen for simplicity.

Tonight I am going to begin reading through the Noun section and sorting every word into one of the above categories.

 

[SyMag]

Sounds good. I'll start looking for that textbook in the meantime in order to draft an outline for the new PDF

 

[Ikaheishi]

But would you still want the multisyllable glyphs, like {ayo} and {ise} separated out from the rest, or have their blocks interleaved with the others in a truly alphabetical fashion?

I put "alphabetical" in quotes because I couldn't think of a better way to say "collation by the order of Inkling letters". Back in February I asked PiyozR what the order is.

The alphabet shown in the PDF is the official order of characters in Inkling. That's how little children in school would learn and memorize it. Arranged by similar beginning phones sort of in the same order as the English alphabet but with the "e" and "Q" characters at the end with other miscellaneous letters.

I admit what I've done doesn't match up 100% with the guide. In the post for the 02/26/2016 version of the guide, PiyozR said about putting the letters in a specific order.

Now all the characters are arranged in order of vowel sound ("a" to "e" to "i" to "o" to "u") with the dipthong characters (ai, ie, oi) shown below their variant for clarity. The "e" and "q" characters are now arranged by the same vowel order also for clarity.

However, in the guide, {fu} is before {fe}, {eyu} is before {eyo}, and {uye} is before {uya}. I more-or-less corrected these in my charts to match up to the otherwise present {a}, {e}, {i}, {o}, {u}, {oi} order.

The pure-vowel set of glyphs {a}, {e}, {i}, {o}, {u}, and {oi} would mess up a pure wcscmp() without some simple logic around it anyway, but you are right, when existing code is available, make as much use of it as you can.

How would it be a problem? wcscmp() is used with strings of wchar_t… Each Inkling character will have its own codepoint in the BMP PUA, so regardless of whether a wchar_t is two or four bytes, each Inkling character will be only one wchar_t.

 

[Jonathx]

I put "alphabetical" in quotes because I couldn't think of a better way to say "collation by the order of Inkling letters". Back in February I asked PiyozR what the order is.

Okay, I see, so the doc's order is final. I agree with you though that for encoding purposes it makes sense to keep the {a} {e} {i} {o} {u} {ai} {ie} {oi} pattern strict, even if that means {fu}/{fe} and some of the dissyllable glyphs are out of canonical order.

Also, there is actually *barely* enough space to have all valid (poly)syllable onsets, including the irregulars, have their own set of 8 codepoints. It's a little bit space-wasteful but would allow greater preservation of the canonical ordering while keeping the established pattern. What do you think?

How would it be a problem? wcscmp() is used with strings of wchar_t… Each Inkling character will have its own codepoint in the BMP PUA, so regardless of whether a wchar_t is two or four bytes, each Inkling character will be only one wchar_t.

You're right, I was just being a derp. I haven't used wcscmp() much b/c we always used ASCII for projects in my CS courses, so I only really had to work with strings of chars or higher-level C++ strings, not wchar_ts.

 

[SyMag]

I've finished making those glyph associations in my software.

Spoiler: The finished chart of all inkling letters

View attachment 3235

Now, you'll probably notice the difference for some of the letter's names. I had to create my own romanization that fits Unicode's requirements for character names. One of my friends also created their own romanization that has some benefits for computer processing, and fits the Unicode requirements as well. I've attached a list in PDF format that shows the three romanizations for each letter, and the names and codepoints for each character.

It would be appreciated if someone made a vector font to this specification. I'm not familiar with them beyond some loose knowledge of how they work, so I can't really make one. (And believe me, I tried to make one… It didn't work out. :()

Okay, so I'm getting around to making a font file for the letters. Do you need me to make characters for punctuation as well? (they're on page 62 on the PDF for reference)

 

[Vylion]

Sorry, real life happened and wasn't able to come visit for a while.

I'm not very clear on how the noun, verb and descriptor material is less "key" than the prepositions, conjunctions and question words material.

I never intended to mean that. What I tried to say is not to do the whole "noun dump" or the whole "descriptor dump" together with their grammar sections, because it would push other grammatical explanations too low on the PDF; and I suggested to, instead, give first grammar explanations, and save all the vocabulary dump together to the end, divided into subsections for grammar categories and subsubsections in thematic categories. I actually tried to say that prepositions and conjuctions were more key.

But it isn't a separate guide. It's a sub-section of the Numbers section in the same PDF. The link to the Squidboards post is just for clarification on forming names for specifics shapes. I excluded that because it would take up space and nobody would miss it. Why would anyone want to download a separate PDF with one page of vocab for something as dull as geometry?

Honestly, I didn't really get it until I read the whole explanation on the original post. But anyway, I actually thought the secondary PDF more of a convenience for those who were not interested in it and would skip it every time they found it.

No, no, no, don't get me wrong. The geometry vocab is an excellent contribution by this community. It's a bit obtuse in comparison to everything else in the guide but it's neat and sensible and a great addition. I was just joking around in that post. I think it's just fine where it is. Using this geometry vocab for something else sounds fun but at this stage in the game I think adding more grammar elements is the last thing I have patience for. Instead we should be focusing on re-ordering the main vocab sections. I say we focus on that.

Ayy no problem, I was kinda using your joke as an excuse to mention that other grammar idea. Which is in itself a long-term idea, not for now, as you said.

I'm kinda in a rush so I couldn't catch up on the whole thread, but here are my opinions on what I saw:

• About the "letter" vs "syllable" nomenclature... I use "glyph".
• My font encodes the glyphs in the first positions of the Unicode Private Area; I read somewhere that it was the standard for conlangs. I don't know what you can do with that knowledge but there you have, @Jonathx and @Ikaheishi
• A bilingual dictionary would be great only if entries could be added way more easily into it than the PDF guide; this way, we could have an almost-always updated list, and PDF updates could be more sparse.
• My font is confirmed to work on the latest LibreOffice version, even in Windows. I understand that you aren't willing to download yet another Office if you already got one though, so I'll try to see if I can export the font to TTF format while keeping the ligatures so it works in Microsoft Office, @PiyozR. I got no way to test them though. Did you try the latest version of the OTF, anyway? I never got to ask you on that. @SyMag this is relevant to you too
• As for 'alphabetical' order, I've assumed it to be the glyph order appearing in the guide.

I'll now reread this while commuting but I won't have internet for a while to respond to anything new.

 

[Ikaheishi]

Also, there is actually *barely* enough space to have all valid (poly)syllable onsets, including the irregulars, have their own set of 8 codepoints. It's a little bit space-wasteful but would allow greater preservation of the canonical ordering while keeping the established pattern. What do you think?

Yeah, I agree it's a bit greedy, but I think it's best.

Of course wherever we put Octoling, we should try to tuck it as efficiently as possible in return for taking up space with Inkling. Looking at the UCSUR, the smallest open spot where Octoling would fit is in U+ED40 to U+EDFF. Think we switch to putting Inkling at U+ECxx or U+EExx then?

I also want to ask your opinion on whether the digits should come first (like my spec) or last (like yours)… I went with putting them first to have a sort-of uniformity between the PUA spec. and the 8-bit ASCII superset. The ASCII superset only has enough room for the letters, so with it, the ASCII digits have to be used. (I did ask about this part of collation, but I seemed to only have confused PiyozR. )

Okay, so I'm getting around to making a font file for the letters. Do you need me to make characters for punctuation as well? (they're on page 62 on the PDF for reference)

At this point the specification is going to change. The new one is going to look like what Jonathx posted, but with the letters in a different order, the digits possibly before the letters, and no punctuation. For punctuation we're just going to be using the codepoints for same-old period, exclamation mark, question mark, etc. as we do for English—so, yes, the Inkling fonts will need to specify glyphs for these characters.

So I don't know if you want to wait on the new specification or if you can somehow start working on the font and move things into the right places after the new specification is finished. I don't want you to have to go through much trouble… this is a fan project, and I think it should be fun to work on (for the most part). :)

(If/when you work on punctuation, please specify a glyph for the ellipsis itself. I know a lot of people just type three periods, but there is a separate ellipsis character…)

At everybody:
Actually, that gets me wondering about some of our inkling punctuation… Right now most of it is just the stuff we use in English, but used differently. Perhaps we should come up with designs more befitting the design of Inkling itself? I have an idea or two and could draw something up if others are interested.

 

[SyMag]

At this point the specification is going to change. The new one is going to look like what Jonathx posted, but with the letters in a different order, the digits possibly before the letters, and no punctuation. For punctuation we're just going to be using the codepoints for same-old period, exclamation mark, question mark, etc. as we do for English—so, yes, the Inkling fonts will need to specify glyphs for these characters.

So I don't know if you want to wait on the new specification or if you can somehow start working on the font and move things into the right places after the new specification is finished. I don't want you to have to go through much trouble… this is a fan project, and I think it should be fun to work on (for the most part). :)

(If/when you work on punctuation, please specify a glyph for the ellipsis itself. I know a lot of people just type three periods, but there is a separate ellipsis character…)

At everybody:
Actually, that gets me wondering about some of our inkling punctuation… Right now most of it is just the stuff we use in English, but used differently. Perhaps we should come up with designs more befitting the design of Inkling itself? I have an idea or two and could draw something up if others are interested.

I'm already working on them lol. I'm using the Inkscape/Fontastic.me method; right now I'm at the [p] consonants. I'm assuming by rearrange, you mean make a chart more akin to those used for Japanese Hiragana/Katakana charts? I don't know what I would have to do to rearrange them as such, but surely it wouldn't be too hard.

I'll make sure to include an ellipsis as well. I also added the question and exclamation marks, though I haven't done any of the other stuff yet.

What kinds of designs did you have in mind for the punctuation?

 

[Vylion]

I'm already working on them lol. I'm using the Inkscape/Fontastic.me method; right now I'm at the [p] consonants. I'm assuming by rearrange, you mean make a chart more akin to those used for Japanese Hiragana/Katakana charts? I don't know what I would have to do to rearrange them as such, but surely it wouldn't be too hard.

I'll make sure to include an ellipsis as well. I also added the question and exclamation marks, though I haven't done any of the other stuff yet.

What kinds of designs did you have in mind for the punctuation?

Hmm. I had planned to make the floating ellipsis from the square dot I had on my font whenever I had time, and add that as a ligature as wel; plus the Calamari N glyph. The problem is, I'm starting finals really soon so that won't happen until after a while.

Also I misunderstood the "alphabetical order" issue. If you're going to make yet a new font and write the characters in yet another different way I have nothing to add, but the ligature system I used in my font internally keeps track of the original Roman characters for stuff like alphabetical ordering and ctrl+f, so that would be a problem if you want to implement your own ordering.

 

[SyMag]

Okay, update. Fonastic.me won't let me make combined letters (such as {oi}); each character can only be assigned to one key on the keyboard. I'll keep looking for other sites that will let me assign characters to multiple keys, but right now it's not looking good.

 

[SyMag]

Actually @Vylion how did you get the Inkling font to work on that Tumblr page you created? Was there a specific site you used? Perhaps I can use that to make a font that could be used in other applications such as MS Word.

 

[Ikaheishi]

My font is confirmed to work on the latest LibreOffice version, even in Windows.

What version of LibreOffice? Don't just say "latest", give a version number. “Latest” could mean a bunch of different things, like a normal release version, a testing version, or even some specific revision of some specific branch of the source code.

And not just that, but what version of Windows?

A bilingual dictionary would be great only if entries could be added way more easily into it than the PDF guide; this way, we could have an almost-always updated list, and PDF updates could be more sparse.

What I was thinking certainly would be much easier to update than what I understand is how the current PDF is created. We could use wiki software such as MediaWiki (the software used for Wikipedia and its sister projects) to replace the entire guide as we know it, if desired. Of course, that means taking on all the problems of running a website.

I'm assuming by rearrange, you mean make a chart more akin to those used for Japanese Hiragana/Katakana charts? I don't know what I would have to do to rearrange them as such, but surely it wouldn't be too hard.

I wish I could better figure out a way to explain what I mean… I should get working on the new specification as quickly as possible.

I'll make sure to include an ellipsis as well.

Thank you very much. :) I tend to use the ellipsis a lot in writing.

What kinds of designs did you have in mind for the punctuation?

Here's some quick work from off the top of my head:

Spoiler: Ideas

Spoiler: In-text samples

(Sorry for the bit of sloppyness)

 

[SyMag]

Here's some quick work from off the top of my head:

Spoiler: Ideas

Spoiler: In-text samples

Not bad at all :) Personally I think the parentheses should be kept as is; the two examples in the last two sentences look similar to already existing Inkling characters (such as {ji}, {jie}, for instance). As for the quotation marks, I like the two examples in the middle. The last one isn't too bad either, but to me personally the middle two are the most visually appealing.

 

[Jonathx]

Yeah, I agree it's a bit greedy, but I think it's best.

Of course wherever we put Octoling, we should try to tuck it as efficiently as possible in return for taking up space with Inkling. Looking at the UCSUR, the smallest open spot where Octoling would fit is in U+ED40 to U+EDFF. Think we switch to putting Inkling at U+ECxx or U+EExx then?

For Octoling, I was thinking of having the alphabetical portion (27 glyphs including the gemination sign but not the vowels, so 2 columns) as compact as possible, but giving the syllabic portion of the script the same treatment as inkling: giving each coda its own set would take up 11 sets of 8 codepoints (including the vowels). But even with the bit of expansion in the syllable area, and including numerals and reserved space it should easily fit at U+ED40 to U+EDFF. So I agree, put it there and move Inking to the EC or EE block to make it adjacent. My personal preference would be for inkling to be at U+ECxx and maybe save U+EExx for splatoon-related dingbats.

I also want to ask your opinion on whether the digits should come first (like my spec) or last (like yours)… I went with putting them first to have a sort-of uniformity between the PUA spec. and the 8-bit ASCII superset. The ASCII superset only has enough room for the letters, so with it, the ASCII digits have to be used. (I did ask about this part of collation, but I seemed to only have confused PiyozR. )

I think I agree with your spec. Numerals first is more parallel to ASCII, and ASCII's always a good model to follow. Also that way converting codepoints of numerals to integer array is just subtracting ED00 from each code.

Also, for transcription in the spec should we stick with your {ž} = <J> or my {ž} = <ZH>? J makes the sets go in almost-English alphabetical order, and since the order doesn't highlight the parallel between it and either {š} = <SH> or {č} = CH I don't really care which gets used. Actually, here's the easy way to find out: @PiyozR is /d⁀ʒ/ or /ʒ/ the more common/"proper" pronunciation of {ž} among Inklings? Since Unicode character names must be specified in ASCII, we need to decide whether to transcribe it as and English <J> (better for /d⁀ʒ/) or <ZH> (better for /ʒ/), so while they're both allowed, it would be nice to know if there's a preference towards one of the alternate pronunciations among fresh linguists.

 

[MandL27]

I've finished making those glyph associations in my software.

Spoiler: The finished chart of all inkling letters

View attachment 3235

Now, you'll probably notice the difference for some of the letter's names. I had to create my own romanization that fits Unicode's requirements for character names. One of my friends also created their own romanization that has some benefits for computer processing, and fits the Unicode requirements as well. I've attached a list in PDF format that shows the three romanizations for each letter, and the names and codepoints for each character.

It would be appreciated if someone made a vector font to this specification. I'm not familiar with them beyond some loose knowledge of how they work, so I can't really make one. (And believe me, I tried to make one… It didn't work out. :()

Late reply, but personally I prefer these romanizations. They're more in line with modern romanizations for Asian fonts (especially Japanese), which could help people who already know those figure Inkling pronunciation out more easily.

 

[Jonathx]

Made this mock-up of where I think @Ikaheishi and I are currently with the encoding. Please note that this is not final!

I used <J> for {ž} in this version since I'm not sure which we're going with. Please tell me if you spot any mistakes!

 

[Ikaheishi]

@Jonathx
We're right on the same track. :)

I do see some errors in the chart though:

• The glyph for the digit 5 is horizontally flipped.
• {žie} is at U+EC45 instead of it its correct position of U+EC46.
• The K-syllables are partly out of order: the order of the glyphs shown is {ka} {ki} {ko} {ke} {ku} {kai} {koi}.

On a different note, I had side-thought/realization about the <J> vs <ZH> matter… Isn't J pronounced as /ʒ/ in French? (I'm thinking that influenced my selection of J on some level.)

 

[Jonathx]

@Jonathx

• The glyph for the digit 5 is horizontally flipped.
• {žie} is at U+EC45 instead of it its correct position of U+EC46.
• The K-syllables are partly out of order: the order of the glyphs shown is {ka} {ki} {ko} {ke} {ku} {kai} {koi}.

Yeah, the images are simple raster screencaps from the main language guide, and I've just been dragging them around to rearrange them, so there were bound to be some errors when I was trying to get it out so fast. Especially frustrating is that word sometimes interprets a click and drag over a short distance as a flip horizontal.

I've fixed the first two, but looking at the guide it looks to me like the K-stem syllables are correct...I might be blind tough, so anyone else on the thread want to double check?

On a different note, I had side-thought/realization about the <J> vs <ZH> matter… Isn't J pronounced as /ʒ/ in French? (I'm thinking that influenced my selection of J on some level.)

I took Latin and Spanish instead of French, so I wouldn't know. (makes sense that the original Latin /j/ would have to transition through a Norman French /ʒ/ before morphing into the English /dʒ/ tho. /j/ -> /dʒ/ directly is kind of a long step linguistically). But in any case, you can see that I've switched it to <J> for the time being, and I'm fine with keeping that way unless and until we get a verdict from "on high"

 

[Vylion]

Actually @Vylion how did you get the Inkling font to work on that Tumblr page you created? Was there a specific site you used? Perhaps I can use that to make a font that could be used in other applications such as MS Word.

You have to install the font on your computer for the Tumblr page to work. Also, MS Word is the only common relevant application where I know it not to work. I've used it many times in Photoshop already, apart from the mentioned LibreOffice Writer.

I used FontForge and an OFL licensed Sans font from which I could start with a Roman alphabet base, and then copied and pasted all the characters from non-bold Sunken Sans into the Private Area, and making a table of ligatures with all possible combinations of consonant romanization alternatives + vowel/accented vowel.

There was a discussion a while ago about alternative romanizations, and I tried to support them all. I forgot to include the "consonant + z" one though. In fact I am also missing the strikethrough T (written only by "ts") and the c Caron (which can be written both with "c" and "ch"). I did not find the "c Caron" but I wanted to look into the "strikethrough t" whenever I got around to add the "z romanization".

If you had to ask me, I'd say that my personal preference is the "consonant + h" romanization. It also reminds me of Chinese romanization, and I think Russian uses "zh" for that sound too.

What version of LibreOffice? Don't just say "latest", give a version number. “Latest” could mean a bunch of different things, like a normal release version, a testing version, or even some specific revision of some specific branch of the source code.

Woops, sorry. Iirc it was 5.1.something, maybe 5.1.2? It was the stable release.

And not just that, but what version of Windows?

I do have said this many times though, it's Windows 10.

 

[PiyozR]

I just finished organizing every vocab word from the Nouns, Verbs, and Descriptors sections into sixteen new topics. In the coming days, I may divide a few of these sections further and likely create more vocab to fill in the gaps. I will begin revamping the guide's sections from the ground up when you guys have crafted a font for us all to use.

In other news, the fourth video lesson for Inkopolis University should be up soon.