yadav

@tobitac i thought it couldn't get worse after this...

There seems to be quite the scandal brewing in our little corner of the left

@sridatta_a @cobbaltt i completely forgot that those dirghas were decomposable in the kannada block! thanks

@sridatta_a @cobbaltt ಓಹೋೕೕೕೕೕೕೕೕೕೕೕೕೕ

gurumukhi is on the right path with the vowel bearers. every brahmic script should do that, but with only one character, not three

@cobbaltt noooooo there should be dedicated plutas that aren't just the number 3. like i should just be able to chain a bunch of dirgha diacritics

@tlonic Welcome to vowel lengthening to fit the metre. Line 3 would make you very mad 😂

@cobbaltt hmm i'll have to reconsider my policy of not thinking about tibetan because it scares me

@tlonic Tibetan is what you're looking for btw

@cobbaltt whenever i see the independent vowel form used to indicate an extra long vowel, i get so mad. it's ugly and tacky!!!

@tlonic Wow can't believe you'd have such a wrong opinion

vowels are an aberration, get rid of this please

and then a language should use the vowel bearer + a virama for glottal stops

@generalusername @hanssium no, zwnj's aren't used in that case, those are used when you *don't* want a conjunct consonant. क्रौ is made up of क + ् + र + ौ. You can use this to see the breakdown: unicode.scarfboy.com/?s=%E0%A4%95%E…

@tlonic @hanssium + for instance, क्रौ is a single syllabic unit in brahmic abugidas but invariably gets broken down to ⟨k‌[zwnj]rau⟩ or something similar in its internal representation, rather than being treated as a basic unit

it pains me to see that complex scripts will always be discriminated in tech against simple, typographically uncomplicated scripts Aksara Jawa will always be "waste of horizontal & vertical space", Mongolian Hudum will always be "too difficult to implement" etc

@generalusername @hanssium is still very small and very compressible compared to most other types of data

@generalusername @hanssium not sure what you mean by latin-like tokens. brahmic scripts use their own codepoints, they are not made up of other characters. the codepoints are higher up so they do use more bytes than Latin characters but that's not really an issue on modern computers; text data in any lang

@learningnothing @cobbaltt i'm not aware of any

@tlonic @cobbaltt are there words in indian languages for the class such as english lacks?

My hottake is that Ḍosā/Dhosā and Dōsɛ/Dōse are different dishes, just like Poroṭṭā and Parāṭhā are now different.

@abenitezburraco @TheStalwart

No effect of writing on syntax complexity ❌ hierarchy depth and sentence length are similar in spoken vs written language ❌the introduction of writing had no discernible effect on clause combining ✅decrease in variance = effect of increased normativity biorxiv.org/content/10.110…

@learningnothing @cobbaltt No, I just meant that as an example of the class.

@tlonic @cobbaltt is that what roti means? will people say roti for idli, puri, pav and rice? "sop" "scoop" "wrap"

@cobbaltt @iwsfutcmd @langofmind Linguistics has gone in all sorts of directions, including cognitive, as it should, but Chomsky has been pretty consistent that his focus is on the formal properties of language. That's why 'intensional' is one of the 'i's of i-language.

@cobbaltt @iwsfutcmd @langofmind "language" here is an abstraction, again, not cognitive, trying to get at the formal language properties of natural language. The "point" is understanding the basic problem of language acquisition. Children do not only learn "particular" languages, they will learn any language

oh come on man that's a low blow (real talk though how many modern mathematicians have read Euclid's Elements?)

@cobbaltt @iwsfutcmd @langofmind any child to learn the grammar of any language with minimal input (c.f. Gold's identification in the limit)

@cobbaltt @iwsfutcmd @langofmind I agree that there's real tension here within the field about what kind of claims UG involves, but UG doesn't come from a cognitive perspective, it's an application of formal language theory to the basic problem of language acquisition: what must language look like in order for

@cobbaltt @iwsfutcmd @langofmind I disagree with this characterization. At least pre-Biolinguistics, Chomsky's program is focused on what Marr (1982) calls the computational level of analysis: what does the system do and why? He sees the "how" question as difficult to answer without knowing the "what" first.

@iwsfutcmd @langofmind ... based on the thesis that language is primarily cognitive and his syntactic theories attempt to model how language is cognitively stored and processed. I've had conversations of pointing out that competing syntactic models - if they explain the data equally well - can ...