Rendered at 11:15:33 GMT+0000 (Coordinated Universal Time) with Cloudflare Workers.
abdullahkhalids 8 hours ago [-]
This problem is not limited to Arabic. Variants of the arabic alphabet are used by Persian (including Iranian and Dari dialects), Mazanderani, Qashqai, Luri, Gilaki, Kurdish (excluding Kurds in Turkey), Talysh, Azerbaijani (in Iran), Pamir languages, Pashto, Urdu, Balochi, Sindhi (in Pakistan), Punjabi (in Pakistan), Uzbek (in Afghanistan), Turkmen (in Afghanistan), Saraiki, Hindko, Brahui, languages spoken in Kashmir.
Whole languages are dying out because people are unable to express them properly on computers. Even popular software that dominate these speakers does not care to improve their experience. For example, Urdu has traditionally been written in the Nastaliq form [1], but is usually is rendered everywhere in the Naskh form [2]. There is no way to change this, for example, in Android without basically rooting it and changing the system fonts.
> There is no way to change this, for example, in Android without basically rooting it and changing the system fonts.
I am really surprised Android won't let the user select their own system font. This is a huge accessibility problem, especially for dyslexics.
Conscat 5 hours ago [-]
I feel like I've never gotten a compelling explanation for why Nastaliq is hard/unavailable. I'm not an expert on abjads, but it doesn't look harder to render then Naskh (and it self-evidently is possible since the fonts exist). Does anyone here know why they make it difficult? Urdu is much less obscure than, say, Sharada or other languages with Unicode support. I think Punjabi is also often written in Nastaliq when it's not in Gurmukhi or Roman.
bradrn 9 minutes ago [-]
In Naskh, each letter has only four forms (for the most part — there are a few ligatures etc. but I think ‘only four forms’ remains basically true). The choice between forms is determined almost entirely by position within a word (initial/medial/final/isolated). All the letters are aligned along the baseline and connect to each other in basically the same way.
By contrast Nastaliq is a much more complicated style. Many letters and letter combinations take on several different forms depending on which other letters surround them. Letter joins are usually diagonal, so letters earlier in a word need to be shifted above the baseline by a variable amount. Having to shift letters vertically as well as horizontally greatly complicates other aspects of the style too.
(I recall seeing a nice table some time ago showing all the various different possibilities for letter joins in Nastaliq. Unfortunately I can’t seem to find it again. Still, you might get some idea by consulting the documentation of one of the existing Nastaliq fonts, e.g. Awami Nastaliq: https://software.sil.org/awami/what-is-special/)
mchaver 2 hours ago [-]
My guess would be line height is a challenge and Naskh already exists. Then probably because these scripts are not used often in the places that are centers of software/OS development.
mohamedkoubaa 8 hours ago [-]
I don't know why people look down their noses at Arabizi
mchaver 2 hours ago [-]
Probably because it's a work around and not what most people want to do. Imagine someone telling you you have to type English in Cyrillic. I know if I could no longer type out Chinese characters and had to use pinyin it would feel very odd and like something was taken away.
vessenes 6 hours ago [-]
I don’t know either, but I am aware that in glyph based languages (and this article makes the case that Arabic has some glyph-like features), there is considerable social discussion about the equivalents, like pinyin. Detractors worry that sound-based (where sounds are based on the latin / western orthography) approaches to writing change something fundamental in people’s brains as distinct from more native versions.
In Chinese for instance, you can use a keyboard that combines radicals - parts of a character, or you can use a keyboard that combines phonemes. Those seem likely to change literally how you think in your language. There may be related concerns for Arabic.
That said, one of the complaints in the blog is that two different codepoints render to the same exact letter / phrase / word — this is not a problem unique to Arabic in Unicode, and there are known approaches: I’d expect (I’m not a Unicode expert by any means) that more work on the tech stack for rectification (I’m sure there’s a technical Unicode word for this process of matching codepoints for e.g. search and uniqueness of rendering) would likely be useful for Arabic, and relatively seamlessly flow in many places.
e28eta 4 hours ago [-]
> I’m sure there’s a technical Unicode word for this process of matching codepoints for e.g. search and uniqueness of rendering
That’d be Unicode Normalization. I don’t have an opinion on the best source for more details, so here’s a link from unicode.org https://www.unicode.org/reports/tr15/
I don’t know enough to know whether or not there are still Arabic-specific issues, either in the spec or the implementations.
The example in the article of copy/paste/search is interesting. I think it’s equally likely to be a RtL issue as a normalization bug, but I haven’t done anything significant with either topic.
abdullahkhalids 8 hours ago [-]
Because people don't want to abandon hundreds or thousands of years of culture for a completely solvable problem.
pseingatl 8 hours ago [-]
For a while, Arabizi was wildly popular and universally used on feature phones. When mobiles became smarter, it was used less. Japanese has romaji and Mandarin has pinyin. Arabic's Arabizi would increase literacy rates and solve all these digital problems.
avadodin 5 hours ago [-]
Romanization is a separate issue to using fixed glyphs.
There was a theory in the XIX / early XX century that full literacy was impossible without the Latin script but such claims are ridiculous especially for Arabic which is an alphabetic script already. China has higher literacy rates than Vietnam, for example.
I don't think the many composition rules of Unicode are really necessary, though. Maybe as an extension for academic work or artistic compositions but not for computing.
If all we had were movable types, all of these language users would find a way to write their language that wouldn't require a Turing-complete computer on each glyph. Now the Unicode gods pander to some of these computer-hostile scripts making the users of different scripts feel slighted.
cyphar 7 hours ago [-]
The vast majority of Japanese and Mandarin speakers are also not in favour of replacing their current writing systems (which give them a link to thousands of years of their own history) in favour of simplified systems. I suspect it is the same for Arabic speakers.
numpad0 5 hours ago [-]
Romaji/pinyin are widely used for typing the actual written scripts. They're not seen as alternate written scripts outside of edge case scenarios(like chats in FPS)
throwaway27448 3 hours ago [-]
I generally agree with what you're saying, but there is rather famously a simplified form of chinese that was designed specifically to increase literacy rates.
smitty1e 2 hours ago [-]
This seems an esoteric problem for the outsider.
But consider how cursive is dying out in (at least American) English, and how many centuries of writing will become unintelligible to the casual reader as a result.
All of these important cultural artifacts require maintenance.
cenamus 1 hours ago [-]
This has pretty much already happened for the older style of German cursive, called Kurrent. Partly also because the Nazis got rid of it.
Tons of old documents written in it, basically impossible to decipher for anyone that only learned to write "modern" cursive or even print.
a_t48 59 minutes ago [-]
Oh hey, second chance queue, nice. I'd originally searched this up because I was curious about how Arabic worked on early computers.
pseingatl 8 hours ago [-]
For a while, Arabizi was wildly popular and universally used on feature phones. When mobiles became smarter, it was used less. Japanese has romaji and Mandarin has pinyin. Arabic's Arabizi would increase literacy rates and solve all these digital problems.
yorwba 2 hours ago [-]
Many Arabic-speaking countries already have very high literacy rates. Meanwhile Somali is a related language officially written in the Latin alphabet, but Somalia has a literacy rate around 50%: https://ourworldindata.org/grapher/cross-country-literacy-ra...
So adopting Arabizi without increasing access to education can be expected to do roughly nothing for literacy, whereas with a good education system, people can learn to read and write in Arabic script just fine.
harshreality 7 hours ago [-]
(2017)
How much of this is still a problem with modern software/font stacks and harfbuzz?
Karliss 6 hours ago [-]
The fact that the article was able to show correct version in regular text is pretty good indicator that if done correctly those are more or less solved problems. I don't disagree that there are probably plenty of times when those mistakes are repeated and solutions not used widely enough (more often for Arabic scripts than other languages), but even for 2017 it feels more like anecdotal examples of what can go wrong ignoring existing technical details. But those mistakes largely come down to having someone who cares and understands the language and technology not for the lack of solutions. There are probably plenty of interesting edge cases that might not be handled perfectly even though solutions for basic cases exist, but article doesn't come even close to discussing those technical details especially if it's only conclusion is "computers introduced more problems, notably because of Unicode".
> The inflexibility persisted and has arguably only become more aggravated in the 20th century
What about 21th century? Digital printing can overlap characters just fine. And modern fonts support context sensitive ligatures and glyph substitutions.
Second/third example those seemed to be caused by more by someone who doesn't understand the language copy pasting stuff.
PDF -> that's just PDF being bad. Text and text search in PDFs tends to mes up even or English.
> with unicode number U+0623, but one can also type أ, which is an alif and a high hamza, represented by unicode numbers U+0627 and U+0654.
That's what Unicode normalization and locale settings are for. Same thing applies to large fraction of latin based scripts other than English, anything which has letters with diacritic marks.
> for كثيره and كثيرة will in most cases yield different results
Similar thing in almost any non English language for example cafe and café or ABC and ⒶⒷⒸ. Although at least some systems handle it reasonably. Not sure how much it is heuristics based on large data (hard to scale across software), and how much it's good application of Unicode character decomposition/normal form tables. Which Arabic letters lack appropriate Unicode decomposition (and other) tables and what are the best practices of unicode normalization/decomposition/locale handling for search (applicable for all languages) are more interesting and actionable topics.
> Not even the simple idea of CJK has been implemented.
Many users of CJK language would argue that CJK unification was a mistake. If different languages prefer different forms of the glyph, they should better be separate characters. Having separate Chinese and Japanese fonts because CJK unified too much just introduces additional points of failure.
numpad0 6 hours ago [-]
It's... interesting how the author sees Han Unification as a feature, when it's just a longstanding and politically charged bug. CJK languages are mutually unintelligible, so displaying CJK texts in wrong fonts won't do anything meaningful; it won't make texts in one language readable to speakers of other languages.
yorwba 3 hours ago [-]
Indeed displaying CJK texts in wrong fonts won't do anything to change the meaning and people who can read it in one font can read it in any font. They might complain that it looks ugly because that one stroke should be slightly longer and have a different angle, but those are ultimately aesthetic preferences that don't affect readability.
Even before Unicode, it was established practice that documents mixing Chinese and Japanese would use the same encoding for both and roughly nobody would bother to pick an ugly font for the foreign-language text to make it look appropriately different.
Unicode rightly decided that the fine details of appearance are left to fonts. Otherwise you'd also need e.g. a bunch of extra codepoints so that early-20th-century handwritten letters in German can have their look accurately preserved: https://en.wikipedia.org/wiki/S%C3%BCtterlin
numpad0 1 hours ago [-]
You can't mix encodings in a single file. A file has one encoding only. It was not possible before Unicode to mix two languages in a single file, whether the languages involved were Chinese or Japanese or French(English was an exception).
Now, if a file was encoded in Unicode, and/or if it was in such document format that support inline font specification, such as HTML, then you could mix two languages without having to stick to one language by e.g. wrapping <font face=Helvetica>paragraphs and words</font> <font face=Futura>with tags</font>.
My point is, it seems that the author is not aware that each of CJK languages are only understood within each countries, in both writings and speeches, and that's somewhat peculiar.
Panzerschrek 5 hours ago [-]
Why do people still use this horrible-looking and hard to process alphabet? Why not switching to latin (as some countries did) or at least to reform it so that it's easier to type and to read?
cenamus 1 hours ago [-]
Because pretty much no language besides Latin actually maps nicely to the Latin script. Pretty much all languages use digraphs, diacritic symbols or completely new letters in the Latin script (even English, œ and æ, for some time at least).
ah27182 3 hours ago [-]
What’s the point of insulting a script as “horrible looking”. What a silly comment, please grow up.
Panzerschrek 4 minutes ago [-]
It's objectively so horrible. Letters have similar form to each other, joints between them makes it harder to take them apart. As I understand, Arabic alphabet wasn't designed to be practical for daily use and only for writing sacred texts.
Whole languages are dying out because people are unable to express them properly on computers. Even popular software that dominate these speakers does not care to improve their experience. For example, Urdu has traditionally been written in the Nastaliq form [1], but is usually is rendered everywhere in the Naskh form [2]. There is no way to change this, for example, in Android without basically rooting it and changing the system fonts.
[1] https://en.wikipedia.org/wiki/Nastaliq
[2] https://en.wikipedia.org/wiki/Naskh_(script)
I am really surprised Android won't let the user select their own system font. This is a huge accessibility problem, especially for dyslexics.
By contrast Nastaliq is a much more complicated style. Many letters and letter combinations take on several different forms depending on which other letters surround them. Letter joins are usually diagonal, so letters earlier in a word need to be shifted above the baseline by a variable amount. Having to shift letters vertically as well as horizontally greatly complicates other aspects of the style too.
(I recall seeing a nice table some time ago showing all the various different possibilities for letter joins in Nastaliq. Unfortunately I can’t seem to find it again. Still, you might get some idea by consulting the documentation of one of the existing Nastaliq fonts, e.g. Awami Nastaliq: https://software.sil.org/awami/what-is-special/)
In Chinese for instance, you can use a keyboard that combines radicals - parts of a character, or you can use a keyboard that combines phonemes. Those seem likely to change literally how you think in your language. There may be related concerns for Arabic.
That said, one of the complaints in the blog is that two different codepoints render to the same exact letter / phrase / word — this is not a problem unique to Arabic in Unicode, and there are known approaches: I’d expect (I’m not a Unicode expert by any means) that more work on the tech stack for rectification (I’m sure there’s a technical Unicode word for this process of matching codepoints for e.g. search and uniqueness of rendering) would likely be useful for Arabic, and relatively seamlessly flow in many places.
That’d be Unicode Normalization. I don’t have an opinion on the best source for more details, so here’s a link from unicode.org https://www.unicode.org/reports/tr15/
I don’t know enough to know whether or not there are still Arabic-specific issues, either in the spec or the implementations.
The example in the article of copy/paste/search is interesting. I think it’s equally likely to be a RtL issue as a normalization bug, but I haven’t done anything significant with either topic.
There was a theory in the XIX / early XX century that full literacy was impossible without the Latin script but such claims are ridiculous especially for Arabic which is an alphabetic script already. China has higher literacy rates than Vietnam, for example.
I don't think the many composition rules of Unicode are really necessary, though. Maybe as an extension for academic work or artistic compositions but not for computing.
If all we had were movable types, all of these language users would find a way to write their language that wouldn't require a Turing-complete computer on each glyph. Now the Unicode gods pander to some of these computer-hostile scripts making the users of different scripts feel slighted.
But consider how cursive is dying out in (at least American) English, and how many centuries of writing will become unintelligible to the casual reader as a result.
All of these important cultural artifacts require maintenance.
https://en.wikipedia.org/wiki/Kurrent
Tons of old documents written in it, basically impossible to decipher for anyone that only learned to write "modern" cursive or even print.
So adopting Arabizi without increasing access to education can be expected to do roughly nothing for literacy, whereas with a good education system, people can learn to read and write in Arabic script just fine.
How much of this is still a problem with modern software/font stacks and harfbuzz?
> The inflexibility persisted and has arguably only become more aggravated in the 20th century
What about 21th century? Digital printing can overlap characters just fine. And modern fonts support context sensitive ligatures and glyph substitutions.
Second/third example those seemed to be caused by more by someone who doesn't understand the language copy pasting stuff.
PDF -> that's just PDF being bad. Text and text search in PDFs tends to mes up even or English.
> with unicode number U+0623, but one can also type أ, which is an alif and a high hamza, represented by unicode numbers U+0627 and U+0654.
That's what Unicode normalization and locale settings are for. Same thing applies to large fraction of latin based scripts other than English, anything which has letters with diacritic marks.
> for كثيره and كثيرة will in most cases yield different results
Similar thing in almost any non English language for example cafe and café or ABC and ⒶⒷⒸ. Although at least some systems handle it reasonably. Not sure how much it is heuristics based on large data (hard to scale across software), and how much it's good application of Unicode character decomposition/normal form tables. Which Arabic letters lack appropriate Unicode decomposition (and other) tables and what are the best practices of unicode normalization/decomposition/locale handling for search (applicable for all languages) are more interesting and actionable topics.
> Not even the simple idea of CJK has been implemented.
Many users of CJK language would argue that CJK unification was a mistake. If different languages prefer different forms of the glyph, they should better be separate characters. Having separate Chinese and Japanese fonts because CJK unified too much just introduces additional points of failure.
Even before Unicode, it was established practice that documents mixing Chinese and Japanese would use the same encoding for both and roughly nobody would bother to pick an ugly font for the foreign-language text to make it look appropriately different.
Unicode rightly decided that the fine details of appearance are left to fonts. Otherwise you'd also need e.g. a bunch of extra codepoints so that early-20th-century handwritten letters in German can have their look accurately preserved: https://en.wikipedia.org/wiki/S%C3%BCtterlin
Now, if a file was encoded in Unicode, and/or if it was in such document format that support inline font specification, such as HTML, then you could mix two languages without having to stick to one language by e.g. wrapping <font face=Helvetica>paragraphs and words</font> <font face=Futura>with tags</font>.
My point is, it seems that the author is not aware that each of CJK languages are only understood within each countries, in both writings and speeches, and that's somewhat peculiar.