Font for "Makasar" is missing. #1
Comments
|
Saya adalah orang Makassar, saya akan merasa sangat senang jika bisa membantu. I am a Makassarese, I would feel very happy if I could help. |
|
fwiw, might be worth asking Anshuman Pandey, who has a graphite font at https://github.com/pandey/graphite-fonts in case it may help by providing a starting point for development of a noto font, if he's willing. |
|
You should note that this is not the only script encoded in Unicode which still has no Noto font for it, not even in any other free/open fonts; except possibly some proprietary fonts, or just not-Unicode fonts hacked over ASCII or another script and most often with very partial coverage, such as lack of support for necessary diacritics or ligatures and contextual forms), As of start of 4 February 2022, MOST scripts encoded up to Unicode 14.0 (last version, published on 21 September 2021) are supported with some Noto Fonts, except one complex script from Unicode 13.0 (with a traditional vertical layout, with variable heights for long clusters counting up to 8 base characters, normally not fitting in a single square composition area like Han Sinograms or Hangul clusters). The following supported scripts still have no Noto fonts (and still no other fonts available with an open licence). See also the ISO 15924 Notice of Changes (last updated on 3 December 2021) on the ISO 15924/RA site hosted by Unicode.
Note that there exists some fonts for Khitan Small Script on https://babelstone.co.uk/Fonts/KhitanSmall.html However the following scripts encoded since Unicode 11.0 (at start of this issue in December 2019), are now supported by some Noto fonts (which may still need some additional developments or fixes, so they are in beta, and some are still not hinted or not available in other styles, Serif/Sans/UI, Bold, Italic); only one living script added in Unicode 13.0, Yezidi is well documented and in development (but its usage is shifting, their community is spread across different countries where they are now either endangered minorities or refugees, forced to use other languages and scripts, and with poor support or absence of support for education in their culture; many are then shifting to the Arabic script, as education is mostly provided by islamic schools and mosquees): what was done for enabling Hanifi Rohingya or Dives Akuru should be done for Yezidi for the same reason of preserving endangered cultures (it's more important and more urgent to support them than supporting Elymaic or Cypro-Minoan which are defunct scripts IMHO); Vithkuqi is the only script recently added in Unicode 14.0 which is currently in active development. In February 2022, three additional scripts encoded in Unicode 11.0, 13.0 and 14.0 had Noto fonts released in a first version to support them (with still pending works to fix some issues) :
Also the following scripts are encoded in ISO 15924, but still not in Unicode (so it's for now impossible to define Noto fonts for them, except experimentally with PUA mappings, possibly useful to create Unicode encoding proposals and working documents in PDF form, or test websites). Some of these scripts are planed to be encoded very soon, maybe in Unicode 15.0, much earlier as expected as their addition in ISO 15924 is also very recent, made at the same time as the announcement that these scripts were planed for faster encoding and support (long before other scripts waiting since many years). This is because they have a wide and wellknown modern usage (but were forgotten for too long in the encoding processes, possibly there wer political issues blocking them for ISO 15924). See Proposed New Characters: The Pipeline on the Unicode site:
Working on developping beta versions of open fonts for the previous scripts (but still with PUA mappings, until these scripts are encoded) is possible, if one want to accelerate first release of a suitable Noto font for them, because the proposals are already solid to support a significant part of these scripts and create test documents that will help finalize the Unicode encoding and review its pre-released version before it is finalized for ever. Such documents won't be easily interchangeable (except when using webfonts for HTML documents, or by embedding these fonts in documents, or for providing rendering snapshots that will be useful for these final talks). Other scripts, most of them in the Unicode roadmap for later possible encoding in the SMP and two others in the Unicode roadmap for later possible encoding in the SIP, include :
But also these scripts, roadmapped for possible later encoding in Unicode (but that still don't have any ISO 15924 script code assigned):
Other proposals include ranges of characters proposed for musical notations, mapped as part of the ISO 15924 "Common" script, even if they mostly used in contexts where (traditional) Han ideographs are used. These includes the Unicode encoding proposals for: Chinese Flute Musical Notation, Chinese Lute (or Pipa) Musical Notation, and Jianzi Musical Notation. As well these scripts are tentatively roadmapped for later encoding in Unicode (and they also don't have any ISO 15924 script code assigned), but lack a formal proposal containing enough information to make a solid proposal :
Finally other scripts, which were added to ISO 15924, have been rejected as unsuitable for encoding in Unicode (meaning that they can only be supported by PUA, managed by private agreements such as ConScript), or did not provide enough information for even a pre-allocation in the Unicode roadmap :
Within the scripts above, Egyptian Demotic is probably the most wanted, it has tons of documents. It is one of the three scripts that have the oldest ISO 15924 code assigned in 2004, and still not encoded in Unicode (and still no pending proposal to encode it that can be coherent for being accepted also by ISO). This is strange, given the large number of egyptologists around the world, and tons of documents archived in famous public libraries (notably in France and US for the largest collections, but there are many as well in Israel, UK, and Russia). For now it is supported partly by fonts hacking some other encodings (sometimes ASCII, may be Coptic with a few custom mappings, but Egyptian Demotic also requires lot of specific ligatures, and probably many "variant" composition sequences; but linguists are already using a productive subset without variants or rare ligatures; the problem is that it should be RTL by default, and ASCII or Coptic are LTR; another possibility could be based on a mapping to the existing Unicode encoding of Meroitic Cursive). Using PUA for unencoded scripts will often not render correctly as PUAs are LTR by default and there's unfortunately no RTL subset of PUA - Look at the responses [1] and [2] to "RTL PUA?" by Michael Everson on the Unicode mailing list, 2011-08-19. Two other living script (Shuishu and Naxi Dongba) are also the only two pictographic scripts still living and in use today (in China), but they are highly endangered (despite of their very specific nature: how could they resist for a so long time and did not evolve into some ideographic form like Han, i.e. with logographic/semantic and syllabic features mixed and some simplifications?). These are not "complex" scripts by their composition feature (they are however moderately large), but we lack sources for these pictograms which could be simply digitized and mapped directly. However Unicode does not seem to work on them, as most work in China for Unicode is done for Han by the Ideographics Rapporter Group, visibly not interested in it. There are probably other interested groups in China that could prepare a list of pictograms easily, as long as it remains a living script there, or would help archiving samples in Chinese libraries and elsewhere to preserve them. May be there are people now living outside China that have kept samples (books, or other pieces of art such ceramics, jewelry, tools, clothes, photographs) which would be useful for their urgently needed encoding, and that could be collected on community photo sharing site (with open licences) or as PDF facsimiles, and that could help graphics designers to redraw them (e.g. as SVG shapes) and then collect these glyphs into a beta font useful for making a script encoding proposal in Unicode/ISO/IEC 10646. See: Naxi Geba is also living and endangered, but it is a syllabary (with variable phonetization depending on reader, which may result differences of interpretation and delaying the normative encoding), not pictograms like Naxi Dongba. It is also used to write the Naxi language; it was created to complement or replace Naxi Gongba which was too defective for modern use and sufficient coverage of the language. It is still used by a minority (with no really active support by the local Chinese government and educative institutions in Yunnan, only kept in archives for cultural reason). Ranjana (or Lanydza or Lantsa) is an indic script used in Nepal and Tibet, with a very rich history and tons of documents. It has been endangered in Nepal by the forced adoption of Devanagari in the 1950s-1980s. But it still has fonts already available for it, it is displayed everywhere, and it is actively used and now supported again by the government of the Kathmandu Valley to write the Newa language, so it should be easily encoded (at least the core part even if there are additions later). See: |
|
I have a made an experimental makassar font which can be downloaded here: https://aksaradinusantara.com/fonta/font/Salapa%20Jangang?key=aa166f01886b8b5271fd984ac79a2f5a it needs some refinement but may be useful as a starting point |
|
@adtbayuperdana great to have this. Thank you. Looking at the licence i wasn't 100% sure whether i could make a webfont from this, and serve it with my Makasar character app. Could you confirm? Also, i noticed that the digit glyphs applied to the ASCII range look like arabic digits. Arabic digit code points don't map to glyphs in the font. This may be a bug, given the following passage in the block description of the unicode standard:
Fwiw, here's what i see (ascii code points on the left, arabic on the right): |
|
@r12a the regularization of Salapa is kinda experimental since (to my knowledge) a printed version of the script has never been developed, so i have nothing i can reference to. At times I still tweak a lot of its aspect to look somewhat more authentic. For styles with clearer reference, I have tried to make three other fonts. They are not released anywhere yet, but here's a sample of them: perhaps they are better if you want to make a webfont that are more representative of Makassar? right, the digits should be corrected! although, i saw eastern arabic digits in makassar/buginese manuscript as not very far off from from current iterations of that digit. However, western arabic digits in those manuscripts looks noticeably different than "normal" ones. Whether makassar font should just follow "normal" shapes or have makassar/buginese shapes for the western arabic digits is a thing that i am unsure of. |
|
With Noto Serif Makasar font being available, can this issue be closed? |
|
I'm keeping it open for the moment because for some reason Makasar was missed out when I moved everything to the new build system. Once it's moved across we can get it on Google Fonts and close this issue. |
|
There's now the Nag Mundari script [Nagm/295] without font. It is encoded since Unicode 15.0 (September 2022). See |
|
We’re working on a number of new scripts for 2023, Nag Mundari being one of them. |
I want to bring to your attention that the font for "Makasar",
Unicode plane: U+11EE0 - U+11EFF is missing.
https://en.wikipedia.org/wiki/Makasar_(Unicode_block)
Please upload this font here if possible.
Thank you.
The text was updated successfully, but these errors were encountered: