GS Unicode 2.0 (Plane 0)

Share:
by Greenstar967

Download disabled

The designer of this FontStruction has chosen not to make it available for download from this website by choosing an “All Rights Reserved" license.

Please respect their decision and desist from requesting license changes in the comments.

If you would like to use the FontStruction for a specific project, you may be able to contact the designer directly about obtaining a license.

What if GS Unicode wasn't just made of pixels?  What if it had a better, more appealing width/height ratio?  What if it utilized the new FS inventory?  And what if I wasn't way too ambitious/crazy in making this?  No idea, but I'm eager to find out!

I've split this font series up by plane since font files have a technical limit of 65,535 glyphs.

Things that are planned but not fully implemented yet are italicized.

This font (Plane 0) should also serve as the hub for most major updates (like those relating to the series as a whole).

Links to planes:

Plane 0: [THIS FONT]
Plane 1: here
Plane 2
Plane 3
(No Unicode characters exist in Planes 4 through D as of U14.0)
Plane E: here
Plane F: here
Plane 10: here

PUA assignments for Plane 0:

E000~E07F - Tengwar
E080~E0FF - Cirth
E100~E14F - Engsvanyali
E150~E1AF - Kinya
E1B0~E1CF - Ilianore
E1D0~E1FF - Syai
E200~E26F - Verdurian
E270~E28F - aUI
E290~E2BF - Amman-Iar
E2C0~E2CF - Streich

E2D0~E2FF - Xaini
E300~E33F - Mizarian
E340~E35F - Zirinka
E360~E37F - Sarkai
E380~E3AF - Thelwik
E3B0~E3FF - Olaetyan
E400~E42F - Niskloz
E430~E44F - Kazat Akkorou
E450~E46F - Kazvarad
E470~E48F - Zarkhand
E490~E4BF - Rozhxh
E4C0~E4EF - Serivelna
E4F0~E4FF - Kelwathi
E500~E51F - Saklor
E520~E54F - Rynnan
E550~E57F - Alzetjan
E580~E59F - Telarasso
E5A0~E5BF - Ssuraki
E5C0~E5DF - Gargoyle
E5E0~E5FF - Ophidian
E600~E62F - Ferengi
E630~E64F - Seussian Latin Extensions
E650~E67F - Sylabica
E680~E6CF - Ewellic
E6D0~E6EF - Amlin
E6F0~E73F - Unifon Extended-1
E740~E76F - Unifon
E770~E77F - Solresol
E780~E7FF - Visible Speech
E800~E82F - Monofon
E830~E88F - Dni
E890~E8DF - Aurebesh
E8E0~E8FF - Tonal
E900~E97F - Glaitha-A
E980~E9FF - Glaitha-B
EA00~EA9F - Lhenazi
EAA0~EAFF - Wanya
EB00~EB3F - Orokin
EB40~EB5F - Standard Galactic

EB60~EB9F - Braille Extended-1
EBA0~EBDF - Cistercian Numerals
EBE0~EBFF - Kaktovik Numerals
EC00~EC2F - Cylenian
EC30~EC6F - Syrrin

EC70~ECFF - <reserved for future (U)CSUR allocations>
ED00~ED3F - Deini
ED40~ED7F - Niji

ED80~EF7F - <reserved for future (U)CSUR allocations>
EF80~F2FF - Latin Extended-1
F300~F3FF - Combining Diacritical Marks Extended-1
F400~F4FF - Symbols and Punctuation Extended-1
F500~F53F - Greek Extended-1
F540~F57F - Cyrillic Extended-1
F580~F7BF - <unassigned>
F7C0~F7FF - Kana Extended-1
F800~F81F - Font Specifics
F820~F89F - <unassigned>
F8A0~F8CF - Aiha
F8D0~F8FF - Klingon

Additional PUA assignments can be found in the Plane F and 10 fonts.

Feel free to recommend PUA assignments!

Also, it's more than likely I won't know every script well enough that there won't be any mistakes.  If/when I make a mistake, please tell me and I'll do my best to fix it!

[Description last updated 5 June 2022]

257 Comments

Created 26 June 2021, published 16 July 2021 with 935 glyphs.

So far supports:

All blocks from U+0000 to U+036F and most of the Greek and Coptic block

Comment by Greenstar967 16th july 2021

Also: if anyone has any suggestions for me to put in the newly added PUA, let me know!  It doesn't have to be a formal Unicode proposal or anything, but do please tell me approximately how much space it'll take and where you'd want it in either PUA.

Comment by Greenstar967 16th july 2021

Klingon. It can be in the last three rows of the BMP PUA. THey can be encoded according to this CSUR standard

Comment by Bryndan W Meyerholt (BWM) 16th july 2021

(one row = 16 codepoints)

Comment by Bryndan W Meyerholt (BWM) 16th july 2021

Oh you, lovely and blessed crazy people! 💕 (... And I'm happy if one of my fonts reaches 200+ characters...)

Comment by elmoyenique 16th july 2021

Added: some basic Cyrillic stuff, as well as modifying lowercase Greek Kappa

Current character count: 1031 

@BWM, if I understand correctly, I should add Klingon to U+F8D0~U+F8FF?

@elmoyenique Thank you very much <3, though a large inventory isn't the only thing that makes a font great; you've got a whole bunch of stuff that looks really cool!

Comment by Greenstar967 16th july 2021

@GS987 Yep, those should be the correct codepoints…

Comment by Bryndan W Meyerholt (BWM) 17th july 2021

Added: First half(ish) of Latin Extended Additional

Current character count: 1191

@BWM Perfect, I will do that!

Comment by Greenstar967 17th july 2021

I should also probably point out that as .ttf fonts only allow for a maximum of 65,535 characters, I will only be encoding the BMP here.  Supplemental planes will be added in separate fonts.  BMP PUA proposals will still be added here, while the F and 10 planes' PUA proposals will be added in those separate fonts.

I also need to figure out what I'll be doing about the CJK characters, since they'll probably have a different width/height ratio....

Comment by Greenstar967 17th july 2021

Added: Klingon (in the same places as designated by CSUR) - @BMW and anyone else interested, please let me know if I made any mistakes; I'm more of a Star Wars person lol

Current character count: 1230

Comment by Greenstar967 17th july 2021

Good luck! You'll need it if you want to beat me! In the meantime, see you in Most Glyphs!

By the way, the Unicode 14.0 beta review charts are here and the roadmaps are here (but you'll need some hacking to access new blocks).

Comment by erictom333 17th july 2021

Ahaha thanks for the wishes - good luck to you as well! - and thanks for the charts.  U14 should come out in September though, and while it might be nice to be done by then, I don't particularly see it happening; I'm not that optimistic, especially since I'm not doing a pixel font.  I'll use hacks if I need it, but for now I'll probably just get everything up to U13 done, and by that point U14 will already be out, and hopefully Mr. Meek will have that encoded as well.

And we'll probably both be sitting at the top with 65,535 glyphs eventually....

Comment by Greenstar967 17th july 2021

Added: the rest of Latin Ext. Additional

Changed: how double Latin diacritics look

CCC: 1326

Comment by Greenstar967 17th july 2021

Added: Latin C

Changed: a few hooks and stuff, including on eng and emg (m-hook for IPA)

CCC: 1358

Comment by Greenstar967 17th july 2021

Added: (almost) all remaining U13 Latin characters and a lot of punctuation

Changed: the way a few Latin letters and combining characters look

CCC: 2012

Comment by Greenstar967 18th july 2021

Took me way longer than I'd like to admit to draw the star.  Suprisingly hard in a 9x9 square....

Comment by Greenstar967 19th july 2021

I wonder how you'll do the accents that are typed after a letter, but placed before it, and two-part vowels when you get to those languages eventually, especially abugidas…

Comment by Bryndan W Meyerholt (BWM) 19th july 2021

Added: the rest of the non-combining Latin and Greek letters and a lot more Cyrillic, punctuation, and symbols

Changed: a few Latin aesthetics

CCC: 3010

As for the two-part vowels and other combining character weirdness, I had a way of getting around it in the original GS Unicode (successfully?), so I'll probably do something similar here.

Comment by Greenstar967 19th july 2021

This looks really good so far, in both quality and quantity. I hope my browser can load the font up when you get to about 18k letters (I have 12 GB of RAM installed, my laptop can have 16GB max).

It even looks good at pixel size and double pixel size…

Comment by Bryndan W Meyerholt (BWM) 19th july 2021

Oh thank you!

Comment by Greenstar967 19th july 2021
Comment by Greenstar967 20th july 2021
Comment by Greenstar967 20th july 2021

3172 Fonts that's too many! Now it's too hard to get too many glyphs! So I need to make this font with 300 glyphs!

Comment by VTOTFS2021 20th july 2021

Coptic is now supported!

Comment by Greenstar967 21st july 2021

Armenian is now supported!

Comment by Greenstar967 22nd july 2021

Hebrew has been added, as well as some basic Arabic and some more symbols.  This is Genesis 1:1.  Hebrew vowel points seem a little skewed - I don't think it's a problem with combining characters generally though, since they work well with Latin, and I've done the same thing for Hebrew.

The box drawings seem to be better, but still broken.  I believe it's a problem with font height, since this problem exists with some combining characters that extend very high or low as well.  I'll fix that.

CCC: 3819

Comment by Greenstar967 25th july 2021

I really love how this is going so far.

For Arabic and other connected scripts, could the letters connect to one another so it looks nicer?

The block elements could be extended vertically to match with the box drawings, and also connect to one another horizontally…

Comment by Bryndan W Meyerholt (BWM) 25th july 2021

I will definitely do that with the box drawings.

I could try for Arabic but, unless this has been fixed, letters default to their isolated forms anyway, even when they're supposed to be linked to other letters.  I'm not sure what connecting isolated forms would do if this is still the case.

Comment by Greenstar967 27th july 2021

The isolated forms don't need to be connected. Just the initials to the medials to the finals

Comment by Bryndan W Meyerholt (BWM) 27th july 2021

Ah so like for the Presentation Forms blocks?

Comment by Greenstar967 27th july 2021

Yep…

Comment by Bryndan W Meyerholt (BWM) 27th july 2021

Ah yes I'm planning on doing that anyway, thanks!

Comment by Greenstar967 27th july 2021

The rest of Arabic and APF B has been added (still need to add most of APF A).  Block elements have been fixed to connect, and the glyph height has been fixed so that combining symbols don't cut off at the top....which means I have to re-fix the box drawings and other top-bottom connecting symbols.

Comment by Greenstar967 29th july 2021

This is (I think) the first verse of Muhammad's first revelation.  The first is with vowel markings and such (that don't seem to combine properly at the moment), with all letters from the basic Arabic block at U+0600.  They don't naturally connect based on context, though I'm not sure if this is due to FS limitations or my own.  The second line is the same without vowel markings, and with the (I hope) correct contextual forms from APF B.  Additionally, the first line naturally moves RTL while the second does not.

Comment by Greenstar967 29th july 2021

Also I may have needed to have an inital qaf in the second line, rather than medial.

Comment by Greenstar967 29th july 2021

CCC: 4234

Comment by Greenstar967 29th july 2021

Added: Basic APF A, Braille, Yijing, some arrows, and more symbols

Fixed: Some Arabic glyphs

CCC: 4896

Comment by Greenstar967 30th july 2021

5K!

Comment by Greenstar967 30th july 2021

Added: Lisu, more dingbats, more symbols, more arrows, more shapes, the rest of the box drawings, and some fullwidth compatibility

Fixed: box drawings (hopefully), other top-bottom connecting characters, some Arabic linking characters, and a Latin ligature

Changed: a few Latin characters

CCC: 5705

Remember to send PUA suggestions if you have any!

Comment by Greenstar967 1st august 2021

Some suggestions for the PUA:

-The extended IPA characters in Latin Ext-G

- Hiragana and Katakana yi, ye, wu (link)

- Cirth and Tengwar, as in the CSUR

- The Metelko alphabet for Slovenian

Latin letters used in Soviet minority languages

Comment by erictom333 2nd august 2021

@Eric any suggestions for character placement?  For the Cirth and Tengwar I'll assign them to the same codepoints as in CSUR standard, as you suggested.

I suppose if you don't have a preference I could put all the Latin and IPA extensions from EF00~EFFF (would be analogous to the U14 proposal of 1DF00~1DFFF), and reallocate them to Latin xG etc. as they come out in Unicode.  Perhaps the same with the kana, in F100~F16F or so, as E100~E16F would conflict with what might become a future full implementation of the CSUR...although that's tentative.

Comment by Greenstar967 2nd august 2021

I've put my Latin extensions in F0C0-F2FF (Latin Ext-F in F0C0-F0FF, Latin Ext-G and phonetic extensions in F100-F1FF, and orthographic extensions in F200-F2FF (including the aformentioned Soviet minority in F200-F251). Metelko is small enough to fit snugly in the unallocated part of Latin Ext-D (A7DA-A7EF). Kana is up to you, as I put them over Japanese squared symbols that I figured I'd never be able to do.

Comment by erictom333 3rd august 2021

That all sounds good, except I may put Metelko in the PUA next to xG, or maybe what you have as Orthographic Extensions, since I'd rather everything not presently in Uniciode be in one spot.

Comment by Greenstar967 4th august 2021

Character suggestions (Within the PUA):

-Cistercian Numerals - Range: U+EBA0 to U+EBDF; Mapping: Image is below (Codepoints are below the numerical values of each symbol). (Note that the tens, hundreds, and thousands combine with the units)

U+EBA0 is the stave without anything at the ends

Comment by Bryndan W Meyerholt (BWM) 6th august 2021

The tens, hundreds, and thousands of the Cistercian Numerals could be moved a bit more to the left to perfectly align with the units…

I also like how Tengwar looks…

Comment by Bryndan W Meyerholt (BWM) 6th august 2021

Added and fixed a lot of symbols, arrows, and CJK stuff.  Also added Tengwar, the Cistercian numerals, and a few Latin xF and xG characters.

For Tengwar, I've used the 2001 CSUR proposal (https://www.evertype.com/standards/iso10646/pdf/tengwar.pdf) and a few characters from the Free Tengwar Project (http://freetengwar.sourceforge.net/mapping.html), as opposed to the 1998 CSUR draft (https://www.evertype.com/standards/csur/tengwar.html).  It has all the same characters (and more) as the original, and still fits into the E000~E07F block, so it should be fine.

I've fixed the Cistercian tens to be aligned with the units, and by extension fixed (I think) every other combining character to be properly aligned with their respective base characters.  For some reason, though, the hundreds and thousands of the Cistercian numerals seem to not be aligned, even though they're structured in exactly the same way as the tens.  I could probably shift the hundreds to be one pixel left of the tens, and the thousands to be one left of the hundreds, but this doesn't seem to be a problem with any other combining characters when placed in a stack like this.  Maybe because it's in the PUA....

CCC: 6601

Also: When Latin xF, xG, and the new kana come out in U14, I'll most likely delete them here and move them over to the Plane 1 version of this font (which doesn't exist right now but will shortly).

Comment by Greenstar967 6th august 2021

As of now, I have this to show for the PUA glyphs so far.  I'm currently working on Latin xF and xG, as well as a few other things.  Cirth seems to be heavily based off Runic, so I want to finish Runic (which will be soon) before going into Cirth, and the same goes for Kana xB (the archaic yi, ye, and wu) which is obviously based off Hiragana and Katakana.

*Yes I am aware that Kana xB will be coming out in U14 as one row of tone marks, and my block name will change to Kana xC when it needs to.

Comment by Greenstar967 6th august 2021

"For some reason, though, the hundreds and thousands of the Cistercian numerals seem to not be aligned, even though they're structured in exactly the same way as the tens.  I could probably shift the hundreds to be one pixel left of the tens, and the thousands to be one left of the hundreds, but this doesn't seem to be a problem with any other combining characters when placed in a stack like this.  Maybe because it's in the PUA"

I found why they aren't being aligned: The widths of the combining tens, hundreds, and thousands are non-zero. (I opened up the font in FontForge and noticed that the width of them was 70 em units, which I assume is one brick in this font)

Comment by Bryndan W Meyerholt (BWM) 6th august 2021

Weird, I thought I set them to zero.  They are now, though, so it should work.  Thanks!

Comment by Greenstar967 6th august 2021

cool. but i'm downloading this when this has about the same character count as GS unicode, the original.

Comment by erikkoev 7th august 2021

Thank you! - and I'd recommend that as well, since this is still a WIP

Comment by Greenstar967 8th august 2021

A few points of criticism.

- Lowercase iota (both Latin and Greek) should have its stem centered; the right hook shouldn't be that big, and the left serif should be larger to compensate.

-  Two-part Arabic ligatures (possibly except lam-alef) should be double width. Word ligatures can be narrower than all their parts combined but should be wide enough to read.

- Lowercase half H should be a small capital.

Other than that, you're doing a good job.

Comment by erictom333 8th august 2021

Ha, it's a good thing I waited to do the rest of APF-A then....I've deleted the word ligatures in APF-A (except the ones for Allah and Rial) and the combining words in Arabic xA, which will be reinstated soon.  Not sure if those two should be widened as well; I like how they look for now, but maybe they should be to fit with the rest of the characters there.  I'm not too much of an expert on Arabic unfortunately.

Should the two- and three-part ligatures in APF-A be double width as well?

Latin, Greek, Cyrillic, and IPA iota have been fixed, as well as the half H.

Thanks for the input!

Comment by Greenstar967 8th august 2021

I mean, the rial is a currency symbol, so it should be one wide. The Allah ligature is up to you.

And yes, I'm meaning the two-letter ligatures in APF-A, they should be two wide. Similarly, the 3-letter ligatures should be three wide.

Comment by erictom333 8th august 2021

Ohh I see; I'll make sure to do that

Comment by Greenstar967 8th august 2021

The word Arabic ligatures I had before have been widened and reinstated.  Allah has been made double-width, while rial has been kept the same.

Comment by Greenstar967 8th august 2021

Many, many more symbols have been added (the entirety of U+2000 to U+2BFF), as well as Ogham and Runic.  A few things have been changed (notably the ASCII asterisk) and fixed (notably Arabic ligatures, a few Latin glyphs, and some combining characters).

Any suggestions, corrections, and/or PUA proposals still welcome.

CCC: 7470

Next up is Cirth, then some more CJK compatibility and kana (probably).

Soon we will reach the heights of the original GS Unicode....

Comment by Greenstar967 10th august 2021

Cirth has been encoded, using a self-made combination of the 1997 (https://www.evertype.com/standards/csur/cirth.html) and 2000 (https://www.evertype.com/standards/iso10646/pdf/cirth.pdf) proposals.  Essentially, the block - still E080~E0FF - is the 2000 proposal with the extra characters from the 1997 proposal stacked on the end.

CCC: 7582

Comment by Greenstar967 10th august 2021

Added more CJK compatibility, Metelko, and some other Soviet minority extensions.

After doing many of the other scripts and learning how to do angles that aren't necessarily 45 degrees (kind of), I realize some of the Latin, Greek, and Cyrillic characters could benefit from something like that...maybe.

I've changed the A, M, N, V, v, W, X, Y, and Z to reflect this change, if I make it.  I haven't changed any of those with diacritics or modifications yet, as I want to get your opinion before I do.  Let me know what you think or if you have any suggestions.

Comment by Greenstar967 11th august 2021

I prefer the older, octagonal versions of the letters.

Comment by erictom333 12th august 2021

I'll make it the octagonal version then, and maybe dedicate some of the PUA to the more slanted/"modern" versions.

Comment by Greenstar967 12th august 2021

Proposed PUA allocation for Latin characters.  The changed Latin letters have been reverted to their original state, while the newer forms have been put in a block from F060~F0BF (basically replicating ASCII, with holes in case of potential future glyph variants.

Latin Extended-G (F100~F1FF) has been split into three parts (with more following if necessary).  The first section will be moved to the appropriate U14 block when that comes out, but the block's range will remain consistent.

Comment by Greenstar967 12th august 2021

I've moved the Kana block as well, to accomodate for broader script type organization, as Unicode does (roughly).  Perhaps F000~F05F could be used for extra Latin or something Latin-adjacent (e.g. Unifon).

All that remains as of now is the rest of the Soviet minority extensions and the Kana forms.

Comment by Greenstar967 12th august 2021

Here's my roadmap for Latin extensions, used in Pixel Sans. I'd recommend you to follow along for compatibility.

White: Characters in Unicode 14.

Khaki: Metelko. I've put it in Latin Ext-D, but since you don't want any non-Unicode glyphs in existing blocks, you could put it in U+F000-F015 (-F013 if you want to keep the Metelko schwa unified with the Zhuang tone two).

Grey: Zhuang tones 3 and 4 disunified from Cyrillic, which you probably won't need to add; and I with bowl, used in Soviet minority languages, disunified from Zhuang tone six, which I would recommend you add after Metelko as it differs very slightly.

Orange: The Initial Teaching Alphabet.

Yellow: Benjamin Franklin's phonetic alphabet.

Red: Soviet minority letters. Light red denotes characters not in the proposal, but attested in sources; leave gaps if you don't want them in.

Pink: Unifon.

Purple: Pitman's English Phonotypic Alphabet.

Green: Additions for African and Native American languages.

Not highlited: Just ignore them (for now). Leave gaps.

P.S. Sorry for the long post.

Comment by erictom333 13th august 2021

Don't apologize for the long post; I quite like this compatibility idea.  It seems like, from this, Latin/Latin-adjacent characters would take up F000~F2FF.

A few questions:

What, if anything, do you have between F000~F0BF?  If nothing, are you reserving it for future Latin/Latin-adjacent?  I may make F000~F2FF one block if this is the case, unless you want to make the block names compatible as well, which I'm fine with.

What do you plan to do with the U14 characters once they're out?  I've been debating back and forth with myself on whether to delete and move them to the SMP version of this font.  Another option may be to keep them in for compatibility, and for people who may not want to download a separate font just for some extended IPA.  I'll probably do whatever you decide, if you have strong opinions on this.

Is there a difference between F20F and F210 and Latin gha, and/or between the Soviet minority ze/che and the Metelko ze/che?  For the latter, the Soviet minority proposal references 19th century Slovene as a language that uses those characters, which I assume is Metelko (and is the reason I didn't encode them twice).  Seeing as all of these are light red anyway, I'll probably just encode the characters for which there's a need to distinguish between one and another.

Thank you for the suggestion!

Comment by Greenstar967 13th august 2021

Welp, never expected such a response! To your questions:

- In Pixel Sans, F000-F0BF were formerly used for scripts that are now in the SMP, like Old Italic; I moved them when FS added support for the SMP. I have my own set of Latin (and Cyrillic) presentation forms in F800-F87F.

- Once Unicode 14 comes out (or when I manage to hack them in earlier), I'll move Latin Extended-F and -G over, though I'll probably keep them in for compatibility and ease of use; I'd want to be able to easily type Small Capital L with Belt or Turned R with Long Leg and Retroflex Hook from Character Map (which doesn't support the SMP).

- Yes. F20F and F210 denote the short gha, which can be found in some documents, though I'm unsure if it is simply a glyph variant of the long gha or is distinct. As for F241 and F24E, they are small capital forms of ze and che attested in some orthographies, and also formerly used as tone letters in Zhuang.

Comment by erictom333 13th august 2021

Ok perfect!

I'll also be keeping Latin xF and xG here then, as well as implementing them in the SMP.  F020~F05F will be reserved for future allocation, and F060~F0BF will stay as my Basic Latin Glyph Variants block, unless there's any future compatibility that needs to be done.

F241 and F24E will be encoded as the tone letters, which means I'll probably disunify all the other tone letters as well.

I may use F20F and F210 as a glyph variant of gha, without the descender, as opposed to 01A2 and 01A3, which have one.

Comment by Greenstar967 13th august 2021

I've mapped out the F000~F2FF range as per (mostly) your compatibility proposal.  The only differences here are in Metelko and my own alternate Latin glyphs range.  I added a lowercase che and disunified tone 6, leaving a space for lowercase tse (currently mapped to 019E), and tones 3 and 4, if I ever decide to disunify them as well.  I'll also remove lowercase che if it's unified with turned h

Perhaps my range from F060~F0BF could be moved to where you have your compatibility forms, but that may be thinking too far in the future.

A few other questions I thought of that may be relevant:

What will you do if/when Unicode adds to the Latin blocks in future versions?  I imagine, especially in the case of xG, that moving every other glyph a few spaces forward would be kind of annoying.  This is part of why I'm wary of allocating PU characters anywhere outside the PUA, but I'll put the U14 Latin xD where it should go, since the characters are frozen for publication anyway.

What are Unifon book and Pitman's dh unified with, if anything?  I assume Frankin's uh is unified with turned h (or che maybe), and Pitman's u with notch is unified with either turned m or the ui ligature, though I could be wrong.

Finally, where are you getting the Africanist/Americanist letters, as well as F295 and F296?  I was able to find a list of Latin-script letters (in Russian) here https://ru.wikipedia.org/wiki/%D0%A1%D0%BF%D0%B8%D1%81%D0%BE%D0%BA_%D0%BB%D0%B0%D1%82%D0%B8%D0%BD%D1%81%D0%BA%D0%B8%D1%85_%D0%B1%D1%83%D0%BA%D0%B2 , and I've gathered that some are from the African Reference Alphabet, or something similar.

Comment by Greenstar967 13th august 2021
Comment by Greenstar967 13th august 2021
Comment by Greenstar967 13th august 2021

Lowercase che and Franklin's uh are both unified with turned H. Unifon book is unified with Metelko O with vertical stroke. Pitman dh was an oversight and can be placed in F2A4. Pitman U with notch is unified with F1C3, a character used in Germanic and Nordic dialectology (F14D-F1D2, which I categorized as "just leave a gap"). And yes, I did get a lot of letters from that Russian wikipedia page (more specifically the infobox at the bottom).

Comment by erictom333 14th august 2021

Perfect - I'll get this done in due time then - thank you so much!

Comment by Greenstar967 15th august 2021

@Eric Is S with descender unified anywhere?  It seems to be in the Soviet Minority Unicode proposal but not in the broader PUA assignment.  If it was an oversight, maybe it'd fit in F1FE and F1FF, or F01E and F01F, so as to not move everything else.

Comment by Greenstar967 18th august 2021

Yes, this was an oversight. It can be placed in F2EE and F2EF.

Comment by erictom333 19th august 2021

Got it, thanks!

Comment by Greenstar967 19th august 2021

Assuming F293 is Latin capital theta, and F294 is modifier letter theta, should the former be unified with F260?  I'm not sure of a difference, and there seems to be evidence that they're the same glyph (see https://www.unicode.org/L2/L2012/12138-n4262-unifon.pdf and http://std.dkuug.dk/JTC1/SC2/WG2/docs/n4195.pdf) - unless it's used in Africanist orthographies (as opposed to Americanist), but even then I think they could be unified unless there are significant glyph differences I'm not aware of.

What are F295 and F296 meant to be?

Also, is the ue letter in the ITA unified with the one at 1D6B?

Sorry for the numerous questions; I just want to get everything right.

Comment by Greenstar967 20th august 2021

- F293 is actually lowercase theta; the capital is unified with Unifon F260.

- F295 is a small capital F214, a variant of ɦ used in Kabardian. F296 is a mirrored D used in Komi and Udi.

- Yes, the ITA ue is unified with the existing ᵫ.

Comment by erictom333 20th august 2021

Ah got it, thanks

Comment by Greenstar967 20th august 2021

I wanted to tell you a little detail: the signs "¡" and "¿" go down in Spanish, at the level of the descenders of letters as "y" or "p". Thanks.

Comment by elmoyenique 20th august 2021
Comment by elmoyenique 20th august 2021

Ah yes, I'm not sure how I missed that.  Fixed!

Comment by Greenstar967 20th august 2021

And why the glyphs "¿" and "?" are differents?

Comment by elmoyenique 20th august 2021

They shouldn't be anymore - I fixed that along with the shift.

Comment by Greenstar967 21st august 2021

new inventory?? what else have i missed

Comment by merrybot_v2 21st august 2021

In the past, FS has only supported certain characters. This is why, even though I encoded every U13 character possible in FS in my original GS Unicode, I only reached about ~14.5k characters.  Now, FS supports all of Unicode, including supplementary planes and private use areas!

Comment by Greenstar967 21st august 2021

very awesome

Comment by merrybot_v2 21st august 2021

Added more symbols, Cirth, and more CJK compatibility, as well as a few ideographs.  Also finished the PUA outline listed above, as well as S with descender, Pitman Dh, and a few more glyph variants for Latin.

Fixed turned punctuarion for Spanish, etc. and a few other Latin glyphs.

Changed: Latin Z with descender and K with descender

CCC: 8234

Comment by Greenstar967 21st august 2021

Very cool!

My request is that you add the Seussian Latin Extensions (U+E630 to U+E64F), as well as Aurebesh (U+E890 to U+E8DF) and throw Muric in somewhere. (https://omniglot.com/conscripts/muric.htm)

Also, ignore U+E643 (Seussian Letter ABCDEFGHIJKLMNOPQRSTUVWXYZ). It just looks weird.

10/10

Comment by Fluxxor392 27th august 2021

That will be done, although I'll probably add E643 as well just to complete the set, if I'm able.  Thanks!

Comment by Greenstar967 28th august 2021

i'd request the canIPA except that (a) it's huge (b) you'd have to come up with your own PUA mapping. that said, if you do end up adding it, the important stuff is on pages 9 and 70-105 of this pdf 

Comment by merrybot_v2 28th august 2021

I'll see what I can do for this - if it ends up being too large (more than 512 characters), I may put it in one of the supplemental PUA planes.

Comment by Greenstar967 28th august 2021

there are approximately 1200 unfortunately (source: very approximate math)

Comment by merrybot_v2 28th august 2021

My thoughts on the above proposals:

- As Muric is not a notable script, it should go in one of the supplementary PUA planes.

- I propose for the canIPA to be encoded in E0000-E05FF (depends on how many characters there are). Character charts are on pages 36, 66-68, and 140-141.

- The Seussian letter ABCDEFGHIJKLMNOPQRSTUVWXYZ can take up multiple spaces, as it is a word ligature, like the Basmala.

- Can I make a reservation for part of the supplementary PUA? I want to reserve all of EE000-EEFFF for my myriad of personal scripts, as per Pixel Sans's PUA (E000-EFFF). These include: Minecraftian (E420-E6FF, a Korean-like syllabary with 676 syllables), Hylian (E700-E77F), Old Gerudo (E780-E7DF), Latin extensions for phonetic use (EA00-EBFF), Mitrata Lipi (EE00-EEFF), various symbols, and a litany of smaller scripts from years ago.

Comment by erictom333 28th august 2021

My thoughts on your thoughts:

- Agreed, Muric will go in one of the supplementary planes

- I've been reviewing the character charts roughly.  I'm not finished yet, but from what I see, a large number of characters (mostly vowels) can be unified with existing characters).  Do you mean F0000~F05FF?

- I'd imagine the alphabet letter would take up multiple spaces.  Agreed.

- If you mean FE000~FEFFF, then sure.  The E plane isn't PUA, but if you want to put stuff there as well that's fine also.

I also propose a joint working proposal for compatibility, similar to the way we did/are doing the extended Latin - just to keep things organized and compatible.

Comment by Greenstar967 28th august 2021

"a joint working proposal" meaning what exactly?

Comment by merrybot_v2 28th august 2021

Muric proposal below.  This is subject to suggestions/improvements and won't go in F300~F34F, but instead one of the supplemental planes as per the relegation of uncommon scripts.

Notes:

xxx4A and xxx4B are combining characters.

Non-Mouse [ĩː] and other additional letters can be made with combining characters.

I'm not sure how the [b] letter works to be honest.  It seems to be unicase(?)

Comment by Greenstar967 28th august 2021

some of the canIPA symbols for 'stop-strictives' (affricates) might be hard to fit into monospace so they can take up multiple character widths if need be. also, there are some consonants that aren't included in the charts on pp66-68. pages 70-105 list all of the consonants.

some of the symbols on pp140-141 are diacritics, some are not, and some can be both.

Comment by merrybot_v2 28th august 2021

@merrybot_v2 what is canIPA?

Comment by Hensley Dodson (Hensilly) 28th august 2021

@hensilly it's an extension of the int'l phonetic alphabet with a bunch more symbols

Comment by merrybot_v2 28th august 2021

what symbols does it include? also, I like how you abbreviate (or abb't) with the apostrophe.

Comment by Hensley Dodson (Hensilly) 28th august 2021

a whole bunch of things like uh "lowercase m but the middle part has a hook on it"

also thanks lol

Comment by merrybot_v2 29th august 2021

@merry "joint proposal" meaning a collaboration between Eric's and my fonts.  Many of the codepoints in our individual PUAs map to the same glyph in the other font.  For example, U+F234 is the same character in both of our fonts.  "working" meaning it isn't finalized yet...at least I don't think so.

Comment by Greenstar967 29th august 2021

ah cool

Comment by merrybot_v2 29th august 2021
Comment by Greenstar967 29th august 2021
Comment by Greenstar967 29th august 2021
Comment by Greenstar967 29th august 2021
Comment by Greenstar967 29th august 2021
Comment by Greenstar967 29th august 2021
Comment by Greenstar967 29th august 2021
Comment by Greenstar967 29th august 2021
Comment by Greenstar967 29th august 2021
Comment by Greenstar967 29th august 2021
Comment by Greenstar967 29th august 2021
Comment by Greenstar967 29th august 2021
Comment by Greenstar967 29th august 2021
Comment by Greenstar967 29th august 2021

Apologies for the many individual comments; unfortunately, FS only allows for a file size of <1MB.  In any case, I've pulled 853 characters (source: counting, which I still may have done wrong) from the PDF, with page numbers labelled where appropriate.  I wouldn't be surprised if I missed any, so if you decide to check my work and I did, please let me know.  --I wouldn't necessarily advise it though; this took me upwards of 10 hours....--

This is a very rough draft, and a full proposal with a code chart has yet to be made.  @Eric perhaps I should consult you on this if you want the canIPA to be in your font as well - you mentioned the block placement at F0000~F05FF (maybe now F0000~F03FF) - although I also understand if you don't want to include this.

Comment by Greenstar967 29th august 2021

so wow if you put this at the top of plane F it'd stretch from F0000 to like F035F. which is just insane.

if you want, i can start working on an arrangement of all this sometime tomorrow :)

also i never thought i'd see canIPA handwriting 

Comment by merrybot_v2 29th august 2021

(which yeah you're probably not going to put it at the top of plane F but wow 853 is a big number)

Comment by merrybot_v2 29th august 2021

No, I'm perfectly happy to include the canIPA in my font(s). I could find a few characters that could be unified: S with short stroke with F22C, L with reversed fishhook with F124, C with tilde with F201, and small capital L with bar with F2F7 (added to my roadmap and Pixel Sans since I posted it here; Unifon small capital). While more can be unified with characters of my own personal use (EA00-EBFF in Pixel Sans), I'd prefer if you didn't do so and kept the canIPA characters in that block.

Comment by erictom333 29th august 2021

Merry: If you'd like to start an arrangement that's fine, but you don't have to.  It might be useful though, as I assume you know the canIPA better than me/us.

Eric: That sounds good with me.

Comment by Greenstar967 29th august 2021

I propose to encode the canIPA in F2000-F235F, so as to avoid conflict with the CSUR. Furthermore, since merrybot is considering joining the collaborative space and has considered adding sitelen pona to their font, I propose to leave a space, either in the PUA (E830-E8BF) or the supplementary PUA (F1700-F178F).

Comment by erictom333 30th august 2021

Sounds good!

Comment by Greenstar967 30th august 2021

i vote for sitelen pona to go in the SPUA-A (F1700~8F)

Comment by merrybot_v2 30th august 2021

After consulting the UCSUR (successor to the CSUR), I have decided to move and enlarge sitelen pona to F1900-F1AFF. canIPA remains in F2000-F235F.

Comment by erictom333 30th august 2021

that USCUR page links to https://www.kreativekorp.com/ucsur/charts/sitelen.html which

(a) is based off an old version of a sitelen pona font (the current version is 4.9 and version 5 will be released in two days), and

(b) has only 130 out of the 137 canonized words. 

there are also theoretically 137x137 = 18,769 combined glyphs. if we were to include those as well, along with the combing pi and cartouche glyphs, and still start at Eric's proposal of F2000, sitelen pona would stretch to F69EF for a total of 18,922 characters.

Comment by merrybot_v2 30th august 2021

which is … too many.

Comment by merrybot_v2 30th august 2021

(edit: oops Eric's F2000 was for canIPA)

Comment by merrybot_v2 30th august 2021

For new versions of sitelen pona, you could add new glyphs at the end of the glyph group…

Comment by Bryndan W Meyerholt (BWM) 30th august 2021

@BWM That would work. We should first encode all elementary and compound glyphs in the proposal (and cartouche and pi pieces), then add the new glyphs at the end.

Comment by erictom333 31st august 2021

I have two ideas regarding the canIPA. Either it could be encoded by letter order (all the As first, then the Bs, through the Zs and glottal stops), by phonetic order (as per @Greenstar967's charts), or by a mixture (first all vowels by alphbetical order, then all nasals, et cetera). Once we reach a consensus, I'll draft up the charts.

Comment by erictom333 1st september 2021

i say we do:

- the vowels, first those on p9 then p33 then p36, in the order of the tables, but sorted by roundedness (so after A4 is B0 and after F4 is A5 etc.)

- the other vowels on p37-38, sorted by Greenstar's chart V

- the consonants, sorted by order in the p77-105 lists

- the p106 semi-grooved consonants

- optionally the four consonants in parentheticals beginning with "formerly" on p113

also important are these few symbols in a slightly different pdf.

Comment by merrybot_v2 1st september 2021

edit: on the fourth bullet point above i meant to say all the consonants on p106-119

Comment by merrybot_v2 1st september 2021

I honestly have no preference as to the order of the glyphs in the canIPA block, though I see utility in doing things roughly in order of the pdf (i.e. vowels, then consonants, then symbols), with maybe the stylized numbers at the end or something if we want to include them.  This may also allow for less oversight or forgetting of glyphs (hopefully).

Comment by Greenstar967 1st september 2021

and there is still the issue of diacritics: since my font isn't monospaced, the diacritic positioning will be inconsistent across characters. what i/we could do is have a separate range for characters in the, say, p70-105 lists that my font would include but yours wouldn't because it'd be unnecessary for them.

Comment by merrybot_v2 1st september 2021

Added: Some hangul and CJK stuff and not much else.  Apologies for the slower progress and response time lately.  I've been adjusting to a new class schedule, but I should be moving a bit faster now.

Changed: A few Latin things

CCC: 8310

Comment by Greenstar967 2nd september 2021

Alright. I'll go with the order in Greenstar's charts then. Expect a list by tomorrow or so.

Comment by erictom333 2nd september 2021

I love how this is going, but 丨 (CJK Ideograph) looks like 一 (Again, CJK Ideograph), but rotated 90 degrees to the left. I wonder if the tick on 丨 could go rightwards from the top to make it look a bit nicer…

Comment by Bryndan W Meyerholt (BWM) 2nd september 2021

Thank you both Eric and Merrybot for the charts - no pressure though; I don't want you to feel like it needs to be done by a certain day or anything.

@BWM I changed the tick on U+4E28 and U+2F01 to go to the right.  I think this is what you're talking about, yes?

Comment by Greenstar967 2nd september 2021

:)

also btw i'd prefer "merrybot" over just "merry"

Comment by merrybot_v2 2nd september 2021

Yep, GS987…

Comment by Bryndan W Meyerholt (BWM) 2nd september 2021

@Merrybot I changed my comment to reflect that

@BWM alrighty perfect then

Comment by Greenstar967 2nd september 2021

Added Seussian.  The ligature character is 3-wide.

CCC: 8330

I've also added the PUA assignments to the font description - hopefully that'll make collaboration and proposals easier!

Comment by Greenstar967 2nd september 2021

ah yes that helps. thanks :)

Comment by merrybot_v2 2nd september 2021

Np! :)

Also added Aurebesh

CCC: 8379

Comment by Greenstar967 2nd september 2021

aaaaa it just keeps getting bigger

Comment by merrybot_v2 2nd september 2021

don't forget to remove the [tba] for aurebesh then

Comment by merrybot_v2 2nd september 2021

First part of the canIPA encoding, concerning the vowels.

Comment by erictom333 2nd september 2021

Looks good so far - thank you again for this!

I've also removed the TBA from Aurebesh, thanks :)

Comment by Greenstar967 2nd september 2021

looks like i have some fonting to do! :)

Comment by merrybot_v2 3rd september 2021

Added: Glagolitic, a Georgian character, and a few more kana

Changed: a Syriac glyph

CCC: 8473

Comment by Greenstar967 3rd september 2021

Changed and fixed some Latin letters with stroke

Comment by Greenstar967 3rd september 2021

Updated description again to reflect the creation of the Plane F font - though I anticipate updates related to this series as a whole will be kept here, in the Plane 0 font.

Comment by Greenstar967 3rd september 2021

Fixed more Latin

Comment by Greenstar967 3rd september 2021

@Eric do you have the mu without right leg anywhere? (in VE2)

Comment by Greenstar967 3rd september 2021

Sorry, my error. I'll put it in F2045 and move everything after it up one spot.

Comment by erictom333 3rd september 2021

Also, you can add links to the description.

Comment by erictom333 3rd september 2021

I didn't know you could do that, thanks!

And I'll edit for the character as well

Comment by Greenstar967 3rd september 2021

Added: stylistic numerals (F050~F059) and some CJK punctuation.  These sytlistic numerals are unified with the canIPA numerals, though they are used in many other contexts.

Changed: a few modifier letters, and Latin small M where it's 7 pixels wide, to make them look nicer

Fixed: iota with stroke and modifier iota

CCC: 8487

Comment by Greenstar967 3rd september 2021

Added and changed some N'Ko

CCC: 8517

Comment by Greenstar967 3rd september 2021

Debating if I should unify these two canIPA characters (red) with two Nordic dialectological characters (blue).

Comment by erictom333 5th september 2021

those blue characters are actually nonexistent in unicode and the first three columns of the yellow range are taken up by Glagolitic Supplement. that said i do see other unifiable characters here (such as 1E0AA with vowel I-B3, unless that's what you did originally)

Comment by merrybot_v2 5th september 2021

It's a proposal, not in Unicode at the moment, but I believe Eric has it in his PUA, or at least some of the characters.  I think maybe we could unify them - with F17E and F1C0?  Maybe I could add the whole range from F14D~F1D2 if that's the case.

Comment by Greenstar967 5th september 2021

See here for proposal

Comment by Greenstar967 5th september 2021

Added basic Georgian (the U+10A0 block)

CCC: 8596

Comment by Greenstar967 5th september 2021

Added: isolated 2-part Arabic ligatures (and akbar) in APF-A

Changed/Fixed: some more unresolved Latin things

CCC: 8693

Comment by Greenstar967 6th september 2021

Added more CJK stuff

Changed and fixed some Latin xC glyphs - changed w with hook to fit with the slight modification of other thin but compact letters, and fixed v with curl (it didn't have one before)

CCC: 8729

edit: and in doing so I found 2C7D is a modifier letter capital V - Eric I'm not sure if you want F0FC unified with that...but I thought it might be useful information anyway.

Comment by Greenstar967 6th september 2021

Yes, I have letters from that proposal in my PUA. I'm also wondering where I even got U+F120 (d with left facing hook); I can't find it anywhere.

Comment by erictom333 6th september 2021

edit: It is in the section for obsolete IPA letters (from F119 to F128).

Comment by erictom333 6th september 2021

Nasals, semi-nasals, stops, semi-stops, and stop-strictives (in Canepari's words). F2165 is F2167 without the long leg.

Comment by erictom333 7th september 2021

Thanks!

Comment by Greenstar967 8th september 2021

I like how @GS987 and @erictom333 writes letters, and I love how this font is going. I wonder what'll be next? Yi? Khmer? Rest of Devangari? Who knows?

Comment by Bryndan W Meyerholt (BWM) 8th september 2021

Ah well I didn't see your comment until now but turns out the answer was Thai (plus more APF-A).

Devanagari is a high priority for me but I don't know how the combining two-part vowels will work (or those for the majority of the Indic/Brahmic scripts for that matter).  I'm not sure if it's within FS's capabilities to even have them formatted correctly, as is the case with Arabic positional forms for example, but I'm trying to see what I can do about that.  Right now I'm just doing what I can with regards to that.

I'm also trying to get both kana systems in somewhere along the line, hopefully soon.

Yi and Khmer aren't of particularly high priority at the moment, but if there's enough demand for either/both of them, I could start focusing more heavily on those.

CCC: 8840

Comment by Greenstar967 8th september 2021

For the left-facing and surrounding diacritics, you could do as I did in Pixel Sans. Instead of left-facing diacritics, I encode a superscript version in the top-left of the character cell, and put the proper diacritic somewhere else. I have dedicated F400-F49F specifically to Devanagari presentation forms.

On the topic of presentation forms, can we talk about Latin presentation forms? My range in F800-F88F covers different scope and has different organisation to yours in F020-F07F. Mine covers Latin and Cyrillic, but not ASCII numbers and punctuation, and covers all possibilities for each letter (including the "default"), but can have some unifications (e.g. IPA script g).

Comment by erictom333 8th september 2021

My F020~F0BF range is more like glyph variations rather than presentation forms.  I envisioned it as having F060~F0BF be a 1:1 mapping of ASCII 0020~007F (and F020~F05F be sort of miscellaneous), but I don't particularly like this solution too much, as some characters may have more than one variation and some may have none.

Are you suggesting we talk about Latin presentation forms because you want to unify/make compatible our blocks?  And then also while we're on the topic of compatibility, when should I encode the rest of your proposal for F0C0~F2FF that you said to ignore for now, if ever?  I don't mind whatever you decide, but I've already encoded some that I saw as unifiable with some of the canIPA characters, so it might make sense to make the rest compatible.

Also, thanks for the Devanagari suggestion

Comment by Greenstar967 8th september 2021

Yes, I was thinking unification of our glyph variants, though I'm not sure about the specifics. Some forms are not necessary, as they can be unified elsewhere (script g with 0261, script r with AB4B, script x with F23D, script y with F1C8). And go ahead with the rest of F0C0-F2FF.

Also, not sure if you've caught this but A7FA is unified with the Soviet minority and canIPA Latin sha.

Comment by erictom333 9th september 2021

Second part of the canIPA: constrictives, trills, taps, flaps, laterals, and additional symbols. I might have forgotten a diacritic or spacing mark.

Comment by erictom333 16th september 2021

Thanks for the rest of the canIPA; it will be updated in time.

Meanwhile:

Added: All U14 Latin in the BMP (I think), plus more U14 Arabic, combining marks, and the som currency symbol.  Also added a variation of long s at U+F0B3.

Fixed: U+033C (for some reason it was upside-down)

CCC: 8871

Comment by Greenstar967 18th september 2021

wait is unicode 14 in fontstruct now? or could you only do that because it supports the entirety of the existing unicode 13 blocks

Comment by merrybot_v2 18th september 2021

Not to my knowledge, I just added to the previously existing blocks.

Comment by Greenstar967 18th september 2021

Apologies for the slower progress recently; I've had exams and will likely be working on this series in waves.  That is, a few weeks of a lot of progress followed by a few weeks of not much progress, and repeat.

I've been working mostly on the Plane F font and U14 additions.

A few questions/comments regarding the canIPA @Eric:

F2085 has a middle tilde and F20D0 has a reversed middle tilde I think.

Should we consider the second "column" of glyphs in the proposal?  For example, in 10.3.2, at glyphs 50 and 51 there are c and dotless j with stroke with a reversed middle tilde.  These don't appear as "main" glyphs, but I don't think they're variants exactly, more like semantic phonetic differences.

I have so far been excluding glyphs with cedillas and leaving holes where they should be.  The reason for this is that Unicode, since I believe 3.0, has made the commitment to not add any future letters with diacritics, except strokes, middle tildes, and those sorts of things.

Are F2093 and F208C the same glyph?

Should we add the glyphs at 10.3.2-41 and -43?

Should we disunify some of the glyphs in section 10.3.4?  The hooks look slightly different in my estimation.

Also, not regarding the canIPA, how do you propose we do glyph variation unification?  I agree it should probably have some sense of uniformity but at the same time I think we have different forms of different letters.

Comment by Greenstar967 22nd september 2021

Sorry for the late reply.

- Yes they do, thanks for pointing that out.

- Yes. I've chosen to disunify them because they constitute distinct characters, and because I've added all the other hyperphonetic (double square brackets) characters.

- I've included them per merrybot's request, as fonts like theirs and some of mine are not monospaced, and anyway some of the ligature characters have irregularly placed cedillas (eg. tş vs ʦ̧).

- Apparently they are. However, this may have been an error on Canepari's part; F2093 should have had its left leg longer than its right, as it is phonologically closely related to and based off ɲ, as opposed to F208C being based off ɳ.

- That's an ommision on my part, so yes.

- No.

- I've started listing all possible variants of letters, so we can then remove all that are inappropriate or could be unified.

My principles are: list all possible variants, including the one already in the font (unless there is no variation whatsoever); if a variant form is a distinct letter in some orthography, unify it with that (e.g. IPA script g, Teuthonista script r), unless it is also ambiguous (e.g. Latin script a - looks like script a in print fonts, and Greek alpha in script fonts).

For example, there are 3 variants of capital A: triangle-body A (e.g. the font used here, Pixel Sans, this font's alt); triangle-top A (NPS Octagon, merrybfont); and roundtop A (this font, 5x7 Type). As the roundtop A can be unified with a character used in the EPA disctict from regular A (F269), it will not be used, so the other two As will be placed as glyph variants.

Comment by erictom333 2nd october 2021

I may put you in charge of the glyph variants then, since it seems like you have a plan, and I don't have much of a preference on how we do this.  I'm willing to unify, though, and I can provide any necessary input as it relates to my font.

Comment by Greenstar967 5th october 2021

Here's what I've got. Circled characters can be unified with existing characters.

Comment by erictom333 11th october 2021
Comment by erictom333 11th october 2021
Comment by erictom333 11th october 2021

And here's my proposed layout of basic letters, numbers, and punctuation in F020-F0BF.

Comment by erictom333 13th october 2021

That works for me!

Comment by Greenstar967 23rd october 2021

Now that FontStruct supports Unicode 14, could the Latin Extended-F and G letters go into their respective blocks?

Comment by Bryndan W Meyerholt (BWM) 25th october 2021

They will likely be duplicated over to a plane 1 font shortly, but I won't delete them from here until the download glitch that affects SMP characters is fixed.

Comment by Greenstar967 25th october 2021

@Greenstar967 that glitch should be fixed now. You may need to make a small change to your design and save it, to clear the cache, before you download.

Comment by Rob Meek (meek) 25th october 2021

Ah perfect!  Thank you very much!

Comment by Greenstar967 25th october 2021

This all works (thanks again Mr. Meek!) so Latin xF and xG should exist in a separate plane 1 font by the end of the week hopefully.

Comment by Greenstar967 25th october 2021

Surprise! - it's the end of the week apparently because I got stuff done early (for once).  The Plane 1 font is now up with U14 Latin xF and xG added, and font descriptions will be changed to reflect this addition shortly.

Comment by Greenstar967 27th october 2021

I'm happy to announce that after almost 3 months of collaboration, the Latin Orthographic Extensions block (F000~F2FF) has been completed with stuff like:

- glyph variants

- phonetic extensions

- Soviet minority letters

- other orthographic extensions

I also made some minor tweaks to a few Unicode Latin letters (nothing too major), either for the sake of consistency with the PUA extensions, or because I actually made an error in the glyph itself.  The more glyphs you have, the more oversights there are; who knew?

I also realize I haven't encoded a new script in a while.  I've been putting off the two kana sets (mostly because they're surprisingly difficult and time consuming), but that has to be done at some point so I may do that along with UCAS or something.  And also maybe finish up the canIPA finally, that too.

CCC: 9155

Comment by Greenstar967 7th november 2021

yeah i should also finish up the canIPA in merrybfont

i am currently about halfway through the constrictives i think

Comment by merrybot_v2 8th november 2021

More additions!

- Arabic xB

- the rest of APF-A (except the second half of the three-part ligatures, and some the U14 word ligatures that take about 20 minutes each)

- some UCAS

- like five kana

- U14 Latin, Glagolitic, Arabic, punctuation, etc. that were added to already-existing blocks

CCC: 9641

Comment by Greenstar967 15th november 2021

Also changed/fixed (han)dakuten to be more legible (why are kana way harder than I thought they'd be)

Comment by Greenstar967 15th november 2021

When CJK is done, I wonder if you'll also create variants (Traditional Chinese, Japanese, Korean, etc.) cause they can look different when written in Japanese vs Chinese texts…

Comment by Bryndan W Meyerholt (BWM) 15th november 2021

I'll do the compatibility forms in F900~FAFF and 2F800~2FA1F but I wouldn't really know which other variants to include beyond that.

Comment by Greenstar967 16th november 2021

Katakana fu is the 10,000th character!

Comment by Greenstar967 17th november 2021

Added:

- more kana

- more UCAS

- Georgian supplement

Edited:

- some letterlike symbols

Fixed:

- subscript j in Latin xC

CCC: 10168

Comment by Greenstar967 18th november 2021

Added:

- most kana

- a few more CJK characters

- most of Phags-Pa

- UCAS (finished, as well as UCAS extended)

- APF-A (everything except some U14 glyphs)

Changed:

- smoothed some glyphs

CCC: 10454

Comment by Greenstar967 26th november 2021

Maybe all of Plane E? It's just two blocks allocated at the moment: Tags, and Variation Selectors Supplement

Comment by Bryndan W Meyerholt (BWM) 29th november 2021

@bwm and neither of those are intended to actually display their characters

Comment by merrybot_v2 29th november 2021

I probably won't include them, or I may just have a blank font or something similar...not entirely sure what use it would provide though.  The Plane E section in the description is mostly for completion's sake

Comment by Greenstar967 29th november 2021

I have Ideas for visual characters within in the SSP:

For the Tags block, you could include a letter inside a tag-like object to indicate what it is (The space tag could have "SP", the begin tag could have "BEGIN", and the end tag could have "END")…

And for the Veriation Selectors Supplement, it could contain "VSxyz" in a dotted box, where "xyz" is a number up to 256 (U+E013A could display "VS75" in a dotted box)…

Comment by Bryndan W Meyerholt (BWM) 30th november 2021

That may work

Comment by Greenstar967 30th november 2021

Most kana have been added - these would be everything (pretty much) you'd use in Modern Japanese, plus the yi/ye/wu stuff in the PUA from F7C0~F7FF recommended by Eric (which are also in the Plane 1 font.  More of this to come soon - next on the kana list is wi/we, the v-series, halfwidth, and circled kana.  I also need to prepare myself for the 1B000 block, also known as, "Help God please I don't want to encode kana anymore because it's way more difficult than I thought and I probably still got stuff wrong anyways."

Enjoy :')

CCC: 10536

Comment by Greenstar967 5th december 2021

Added Mandaic, two more APF-A ligatures, Devanagari, and a bunch more Vedic stuff

Also shifted around the modifier letters and superscripts/subscripts a bit to look (hopefully) more clean and symmetrical

CCC: 10796

Comment by Greenstar967 9th december 2021

I wonder where Devangari/Bengali/Gujrati etc half forms and ligatures will be encoded, if you choose to include them. Consonant clusters occur a lot in Devangari/Bengali (Including copies of left-facing accents that take up a full letter's width, halves of two-part vowel marks not in Unicode, etc)

I'm not sure how many codepoints that will take up, but I think it'll grow well beyond 512 when all those languages are included (Those that use complex text rendering)…

Comment by Bryndan W Meyerholt (BWM) 9th december 2021

Conjuncts and other presentation forms will probably be in a separate PUA block akin to Eric's Devanagari Presentation Forms

Comment by Greenstar967 9th december 2021

I wonder if changing ௷ and ௸ to be 2 letters wide to make them more readable/legible, and changing UCAS ᙭ to be the same size as the others for consistency would be a good idea…

Comment by Bryndan W Meyerholt (BWM) 10th december 2021

I fixed UCAS and Tamil, and added some of my own characters to the Latin Orthographic Extensions block (and filled it completely).  I also added a PUA punctuation symbols block at F300~F33F, and reserved FE000~FEFFF for my own conscripts (there are 9 major ones, and I have to figure out how to allocate them).  I will almost certainly add another Latin block somewhere in the future.

CCC: 10845

Comment by Greenstar967 12th december 2021

@Eric I was reading a comment you made here from months ago

Can I make a reservation for part of the supplementary PUA? I want to reserve all of EE000-EEFFF for my myriad of personal scripts, as per Pixel Sans's PUA (E000-EFFF). These include: Minecraftian (E420-E6FF, a Korean-like syllabary with 676 syllables), Hylian (E700-E77F), Old Gerudo (E780-E7DF), Latin extensions for phonetic use (EA00-EBFF), Mitrata Lipi (EE00-EEFF), various symbols, and a litany of smaller scripts from years ago.

I'm not sure what you want me to do with which range exactly.  EE000~EEFFF isn't in the PUA; I'm not sure if you mean E000~EFFF or FE000~FEFFF, or maybe you do actually mean EE000~EEFFF.  I'm also not sure what you mean by "reservation".  Do you want me to encode them here as part of the compatibility arrangement or did you mean you don't want me to encode anything there?  If the former, can you provide code charts or some other information, and if the latter, I could probably encode some of my own conscripts there if that's ok.  Apologies for the misunderstanding.

Comment by Greenstar967 14th december 2021

Yes, I meant FE000-FEFFF. And by "reservation", I meant that you can encode my characters there if you want to. If you want to encode some conscipts of your own, feel free to do so in another part of the PUA. I have code charts available.

Comment by erictom333 14th december 2021

Ok perfect, can you send those?

Comment by Greenstar967 15th december 2021

Notes:
- Dotted circles indicate diacritics, unless said otherwise.
- For A56, the places of articulation (FE0D0-FE0DB, FE0E9-FE0ED) denote where and in what orientation the methods of articulation and other modifiers (FE0DC-FE0E8) attach. This can be implemented without complex rendering by orientating the places of articulation such that the methods of articulation and modifiers would be placed on the right. See Pixel Sans for an example.
- Tone letters should occupy the height range of the letter 'x'.
- Yoshiese and Minecraftian both have East Asian width.
- Minecraftian syllables (FE45C-FE6FF) are composed of zero or one initial consonant (null or FE440-FE44B), one vowel (FE44C-FE44F), and zero or one final consonant (null or FE450-FE45B). Syllables with a vertical bar vowel have the initial consonant to the left and the final consonant to the right; syllables with a horizontal bar vowel have the initial consonant above and the final consonant below. (Do not use Pixel Sans as an example, you have enough space.) I'm telling you this as I'm not drawing all 676 Minecraftian syllables.
- Some Legend of Zelda symbols (FE804-FE83A) and SSB and FSB series symbols (FE880-FE8FF) are too complicated to draw in such a small space. I will provide images should you do these.
- FEA00-FEBFF are not allocated, they have been moved to F2400-F27FF after the canIPA block. The space is still reserved. I will provide code charts later.
- Script Gerudo (FED80-FEDAF) joins like Arabic; contextual forms are in FEDBO-FEDFF.

Comment by erictom333 16th december 2021

2/7

Comment by erictom333 16th december 2021

3/7

Comment by erictom333 16th december 2021

4/7

Comment by erictom333 16th december 2021

5/7

Comment by erictom333 16th december 2021

6/7

Comment by erictom333 16th december 2021

7/7, sorry I had to split it up into seven because it would have been too big

Comment by erictom333 16th december 2021

That's alright, thanks for the charts!  I'll update you if I need anything.

Comment by Greenstar967 16th december 2021

Hey can you type out the names and ranges as well?  I just want to make sure I'm getting it right.

Comment by Greenstar967 17th december 2021

Websites/sources from where you got these would also be helpful if you have them

Comment by Greenstar967 17th december 2021

FE020-FE02F Raichi-2
FE030-FE06F Orange Sheet Ideographs
FE070-FE08F A20
FE090-FE0CF Aereian
FE0D0-FE10F AS6
FE110-FE15F Pargy
FE160-FE17F Early Gerudo
FE180-FE19F Tone Letters
FE1A0-FE1BF Additional Combining Diacritics
FE200-FE3FF Yoshiese Ideographs
FE400-FE41F Minecraftian Supplement
FE420-FE6FF Minecraftian
FE700-FE77F Hylian
FE780-FE79F Gerudo
FE7A0-FE7DF Gerudo Presentation Forms
FE7E0-FE7FF L8
FE800-FE87F Legend of Zelda Symbols
FE880-FE8BF Super Smash Bros. Series Symbols
FE8C0-FE8FF Fantendo Smash Bros. Series Symbols
FE900-FE97F Additional Mathematical Operators
FE980-FE9BF Additional Symbols
FE9C0-FE9FF Additional Number Forms
FEC00-FEC0F Gothic Ext
FEC10-FEC3F Arabic Ext-C
FECC0-FECFF Cyrillic Ext-E
FED00-FED5F Tatar
FED80-FEDAF Script Gerudo
FEDB0-FEDFF Script Gerudo Presentation Forms
FEE00-FEEFF Mitrata Lipi
FEF00-FEF3F Featural Minecraftian
FEF40-FEF8F African Transitional

Sorry, I don't have any websites that they are on. All of them have been created on paper.

Comment by erictom333 17th december 2021

Oh thanks!  Are they all yours then?  I'd imagine the video game stuff isn't but if you created those too that's cool.  Also, is the FE0D0 block "AS6" or "A56" because you've written it two different ways

Comment by Greenstar967 17th december 2021

The Zelda and Super/Fantendo Smash Bros symbols aren't mine, but all of the rest are mine. The block starting at FE0D0 is A56 (Alphabet 56); that was a typo.

Comment by erictom333 18th december 2021

Oh cool, thanks!

Comment by Greenstar967 18th december 2021

Added Tifinagh, Syriac, and Modern Cherokee

CCC: 11065

Comment by Greenstar967 19th december 2021

Added Kinya (the alphabet, not the syllabary just yet)

I also redid a lot of the more octagonal Latin letters since they weren't my favorite and in my view were inconsistent with the design of most of the other scripts.  I took the more triangular versions in my glyph variants PUA block and swapped them out here.  More work still to be done on this.

CCC: 11152

Comment by Greenstar967 19th december 2021

The Great Swap is mostly completed at this point, with only a few more minor edits to be made (probably missed some stuff as well).

I also did a lot of glyph smoothing, making things look nicer, etc. outside of Latin - especially where Greek and Cyrillic were homoglyphic with Latin (e.g. capital Latin x, Greek chi, Cyrillic ha).

I also also fixed three UCAS glyphs according to this Unicode errata page.

Comment by Greenstar967 20th december 2021

Added some more stuff to the PUA

CCC: 11244

Comment by Greenstar967 22nd december 2021

Added additional extensions for Tengwar, Aurebesh, Cistercian Numerals, and Klingon, according to this font's PUA assignments.  May as well add the whole (U)CSUR as well, although if I do, the Plane F assignments will be moved to Plane 10.  I'm looking at MUFI to see what I can do with that as well.

CCC: 11288

Comment by Greenstar967 22nd december 2021

(Formally) added the Variation Selectors block by marking them as VS###.  If variation selectors are ever implemented and work properly within FS (or maybe they do and I just haven't found how), this may change.

CCC: 11304

Comment by Greenstar967 1st january 2022

Variation selectors are used to specify glyph variant for the preceeding letter. Only VS 1-3, and 15-16 are used, 15 makes it appear as text form, 16 makes it appear as an emoji.

Also, VS 17-48 are used in ideographic variation sequences in the  idergraphic Variation Database (IVD)…

Since FontStruct does not support OpenType (yet), they appear as-is with the preceeding codepoint the exact same for the time being

Comment by Bryndan W Meyerholt (BWM) 1st january 2022

When I said OpenType, FontStruct supports the format, but not yet the features…

Comment by Bryndan W Meyerholt (BWM) 1st january 2022

A few errata in the PUA Latin extensions (that may have been caused by bad handwriting and/or erasing on my end):

- F120 is a d with the hook coming out of the bowl to the left.

- F121 is a g with the tail bent slightly and a stroke.

- F125 is a reversed normal lezh, not a reversed F123.

- F126 has the hook coming out of the left of the t, pointing down.

- F134 is a gamma with stroke, not a ram's horn with stroke. Hence, the superfluous gamma with stroke in plane 10 should be removed.

- F136 should have the bar (actually not a bar) slightly lower.

- F139 is a p with the curl coming out of the left.

- F13D is a spacing modifier left-pointing arrow.

- F13E is a ɉ with the hook coming out of the dot.

- F140 is a reversed o with ogonek.

- F143 is a small capital AO ligature.

- F1E6 is a bilabial click with two dots.

Above are the correct fforms of the letters.

Comment by erictom333 1st january 2022

Also F1E9 is an inverted Ƃ.

Comment by erictom333 1st january 2022

The variation selectors (including those in the SSP) are double-width since that's the most convenient way I've found to do them

Comment by Greenstar967 1st january 2022

Also, I understand how variation selectors are meant for use in Unicode, but I guess my point was that they don't work as FS intends, or maybe they do and I was supposed to implement them myself.  In any case, if FS ever does support them correctly, I'll remove the glyphs and have them work as they're supposed to.  The [VS###] are just placeholders for now.

And I'll fix those BMP PUA errata, thanks!

Comment by Greenstar967 1st january 2022

Hopefully fixed all the errata, thanks for the information

Comment by Greenstar967 3rd january 2022

I've changed the names of blocks (in all planes) of extensions to scripts that already exist in Unicode.  So, for example, Cyrillic Extended-D has been changed to Extended Cyrillic-A.  This is to prevent conflict with future Unicode versions (U15 will most likely be adding Cyrillic Extended-D), though this isn't the most elegant solution, and may change.

Comment by Greenstar967 12th january 2022

You could also try Script Extended-1, -2, -3, and so on…

Comment by Bryndan W Meyerholt (BWM) 12th january 2022

I though of that too.  It might be a better idea, and what I decide to name them doesn't have much consequence anyway unless this becomes a standarized PUA set like (U)CSUR or MUFI...and it'll need a lot more polishing and finalizing before that can happen lol

Comment by Greenstar967 12th january 2022

Alrighty, first update in almost a month.  But for good reason, I hope....

I've added almost all of the Plane 0 CSUR glyphs, which can be found here.  For those who don't know, the CSUR (ConScript Unicode Registry) is/was a somewhat widely-used private use allocation for constructed scripts.

I say "almost" because it turns out some of the proposals were made in the late 90s (25+ years ago!), and as such it was incredibly difficult to find a lot of the glyphs described in the glyph name lists.  So difficult that I regularly stayed up until 4 in the morning or so, scouring various libraries and the deepest corners of the internet for any bit of information, usually with little success.  Due to the fact that I'm currently attending college, and that FS is, ultimately, a hobby, I am left with some "holes" in the CSUR - where some glyphs should appear, but don't.

These holes are:

Syai: E1D4~E1DD, E1DF, E1E2~E1E9, E1EB~E1EC, E1EE, E1FB~E1FC.  Additionally, though included, E1EA, E1ED, E1EF may not be placed correctly.

Amman-Iar: E2B5~E2B9.

Streich: E2C0~E2CE.  Full block.

Serivelna: E4C0~E4E8.  Full block.

Ssûraki: E5A0~E5BC.  Full block.

Ferengi: E600~E610.  Thank you @merrybot_v2 for the rest of the Ferengi block, which I couldn't find myself!

Aiha: F8C5~F8C7.  Full block.

If anyone has any information about any of these glyphs (or if there's something I encoded incorrectly) I would be extremely grateful for your help!

I will also keep looking for these elusive glyphs when I have the time.  Maybe there's somewhere I haven't looked, or something obvious I'm missing.

Additionally, I've added some more font specs (really just font name symbols) from F800~F81F, fixed the Cyrillic Multiocular O (U+A66E), altered Latin R and W, and added more symbols for the Klingon and Aurebesh blocks.  I've also added some extra glyphs for other CSUR blocks that I found at the UCSUR's website here.

More to come!

CCC: 12772

Comment by Greenstar967 5th february 2022

*also added Braille Extensions from the UCSUR, since I forgot to include that

Comment by Greenstar967 5th february 2022

So with the addition of these previous seven comments, I feel the time is right to publicly address the blatant font clones of Erictom's, my, and others' works (especially Pixel Sans and GS Unicode 2.0).  I actually have addressed this in the comments sections of some of these clone fonts, but I thought it might be worth it to have a record here anyways, just in case those fonts are eventually deleted, and since mine won't be (at least by my own hand).

Perhaps surprisingly, I don't actually mind that others are copying my works, at least for now.  With FontStruct's current system, the overwhelmingly disproportionate ratio of downloads/comments here compared to the clones, the dates each font was added, etc., I don't think there is any confusion issue here, so I won't go much further in this respect.  And besides, with all the proposals and feedback, this font is more of a collaborative effort at this point, at least in my eyes.

However, I do want to specifically call out clones that allow for commercial gain, in somewhat harsh terms, though not so harsh to allow for harrassment (which I obviously don't condone).  Circle Sans, whose license, if I understand it correctly, is a good example of this.  It also might lend attribution to the creator of clone font, rather than Erictom333, the creator of Pixel Sans (the original).  I'm not in a position to take any actual action to fonts that aren't mine, but I thought it useful to point this out anyways.

To be clear, I'm not mad or angry or anything like that.  Slightly concerned, if anything, but still concerned enough to post something publicly.  I assume it's an honest mistake, or experimentation with FontStruct's capabilities, or something else that was intended to be harmless, so that's about all I'll say for now.

Comment by Greenstar967 14th february 2022

If you haven't done so already, I wonder if these 32 could be done (within the PUA areas)

They are highlighted in the sample. I saw them in a lot of the fonts with polytonic greek…

Comment by Bryndan W Meyerholt (BWM) 14th march 2022

I can do that

Comment by Greenstar967 15th march 2022

What about Kaotican/other MotS (Music of the Spheres) scripts? https://www.youtube.com/watch?v=PwEM1aNz0yQ [Use this video for the names of the Kaotican letters, video created by me] Maybe you could add those in? [Scripts by Coldplay]

Comment by Fluxxor392 Tue, 22nd march

Also throw in the triangle symbol as "KAOTICAN LETTER GLOTTAL STOP"

Comment by Fluxxor392 Tue, 22nd march

Will do

Comment by Greenstar967 Wed, 23rd march

Well it's been almost over three months since my last update, but I've definitely still been working on this font!  It's just been mixed in with a bunch of college stuff and other general business.

Not much of Plane 0 has changed - more characters have been added and shifted around in the range of EF80~F47F, but I haven't really added any additional blocks.  That will hopefullt change soon, although I'm focusing mostly on private use requests for now.

Plane 1 has had some minor additions which are listed on that font's thread.

Plane F has also had a lot more additions, including large chunks of the F2xxx subplane (although still more has yet to be added).  The rest of the FExxx subplane should be next.

CCC: 13002

Comment by Greenstar967 Fri, 13th may

U15 is out as of yesterday!

Though the updates have been less frequent recently I wanted to assure you all that I am still alive, well, and working on this font even though it was last updated a month ago.  I've been working largely on my own Latin/phonetic extensions, as well as more additions to the canIPA.

The canIPA bit has been especially tedious because I had to read through 3193 pages just to make sure I didn't miss any characters.  And some of those that I missed just have a serif distinction or smaller from the ones I didn't.

But I will be back up and running smoothly soon.  College has been a thing so that takes precedence but I'm still spending a good amount of time doing font stuff in the meantime.

Comment by Greenstar967 Thu, 15th september

Also of Interest

GlyphsApp

Get the world’s leading font editor for OSX.

More from the Gallery

GS Unicodeby Greenstar967
GS Unicode 2.0 (Plane 0)by Greenstar967
UnStructed: Syriac Supplementby Greenstar967
GS Unicode 2.0 (Plane E)by Greenstar967
Horizontally Phasedby funk_king
xe A Stroke of Geneus1by Logan Thomason (xenophilius)
fs Insatiableby thalamic
Luthien Pixelby Frodo7

From the Blog

News

Gridfolk: Interview with Zephram

News

Heavy Competition Results

News

Heavy Competition

News

Gridfolk: Interview with Jiri Novak