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.
If you think Arabic horrible because it has connected letters, so is cursive. But I like both cursive and Arabic. You can easily distinguish separate words as the words' letters are connected, and you don't have to put less space to show that some letters are making a word. It's not optimized for printing and digital fonts I agree. But you can't say it's not useful for daily use. It's so much easier on paper.
Most writing systems are optimized for the way they end up being used in practice by the society of their time. Which has nothing to do with "writing sacred texts" per se, it's more a function of traditional means of writing things down and reading them. E.g. if you do lots of scrolls, top-down writing actually makes a lot of sense.
>As I understand, Arabic alphabet wasn't designed to be practical for daily use and only for writing sacred texts.
You clearly don't. Arabic was the premier language for philosophy, science, and mathematics in the middle ages. Algebra, algorithms, zero, cipher, average, and so on are all etymologically Arabic. One might start to suspect bigotry from a "Panzerschrek."
>Letters have similar form to each other, joints between them makes it harder to take them apart.
Your brain is not powerful enough to pattern-match, but your incapacity is not universal.
>It's objectively so horrible.
Literal billions in the world disagree. I might equally claim that the Arabic abjad is infinitely more beautiful than the pedestrian Latin alphabet, especially when expressed in the ugly and diseased Orc-tongue that is called German. [1] Your lack of taste is not universal.
> Arabic was the premier language for philosophy, science
But this was not for ordinary people (peasants) or even accountants, where practicality matters.
> Your brain is not powerful enough to pattern-match, but your incapacity is not universal
With some training it's possible to read Arabic texts. But this requires more mental load and practice compared to other alphabetical systems.
> Literal billions in the world disagree
Argumentum ad populum. There also billions of people using even worse writing systems - non-alphabetical ones. This doesn't mean that they are as good as alphabetical systems like latin.
The video you linked has nothing to do with practicality. It's about calligraphy, which is an art form. It may look good, but it doesn't matter for daily use when one needs to read and type a lot.
> But this was not for ordinary people (peasants) or even accountants, where practicality matters.
All peasant societies were illiterate, including Latin script adopters. Completely irrelevant to Arabic.
"Or even accountants" - apparently, Arabs didn't trade! It's not like Muhammad himself was, you know, a trader and an accountant...
> With some training it's possible to read Arabic texts. But this requires more mental load and practice compared to other alphabetical systems.
Perhaps your brain is too slow. I and many other bilinguals read Arabic even more quickly and efficiently than Latin-script languages. The words are terser and are read as units as opposed to the inefficient character-by-character Latin system. Speed-reading doesn't exist in Arabic because Arabic is already speedily read.
> It may look good, but it doesn't matter for daily use when one needs to read and type a lot.
You just said it was objectively horrible looking above. But consistency cannot be expected of a jumbled mind.
I wonder by which metric you measure these scripts.
Clearly it can't be on pronounciation or information density.
If "amount of letters" is your pick, then Latin might be "objectively" the best system - you'd just be using a very bad metric.
If you're going to unify all the worlds language into one script, then you'd better pick a good measure for that. If everyone on the world learns it, then it doesn't matter if there are 50 or even 100 different characters.You will have to capture _all_ of the nuances of the languages without blowing them out of proportion in size.
Good luck with that.
How can you describe something as “objectively” horrible-looking? Your opinion on the way that something looks is precisely that: an opinion. I’d add that neither the Arabic nor Latin alphabets were designed for anything. They both evolved organically from other previous alphabets.
I didn't read my comment properly. I have written, why Arabic alphabet is horrible. It's all cursive and letters aren't different enough from each other.
> I’d add that neither the Arabic nor Latin alphabets were designed for anything. They both evolved organically from other previous alphabets.
Many alphabets evolved from a need for daily stuff like accounting, where practicality matters. Arabs didn't have widespread alphabet until they needed writing their sacred texts, so, they have invented an alphabet primary designed for that. Other alphabetical systems were already widespread in the regions currently dominated by the Arabic alphabet and many of them look much better.
> Arabs didn't have widespread alphabet until they needed writing their sacred text
Many tens of thousands of pre-Islamic inscriptions made by Arabs are known. Arabs had been writing for centuries before the Quran, and the current Arabic script is just an iteration of earlier scripts. You might want to educate yourself.
There seems to be an interesting link between the Riemann-Zeta function (therefor prime distribution) and the energy levels in black hole singularities.
Upfront modeling work is always worth it, but that only holds if you actually know your access patterns upfront. Most teams don’t, especially early on.
The article title seems a little too far-fetched but apparently there seems to be an interesting link between the Riemann-Zeta function (therefor prime distribution) and the energy levels in black hole singularities.
Otherwise yes you can authenticate against the other versions with a email/password (really the email doesn't do anything in the open source distribution, just a user identifier but we keep there to be consistent)
We have a similar wrapper for local LLMs on the roadmap.
If you use CLI only - we run claude 4 + gemini on the backend, gemini serving most of the vision tasks (frontend validation) and claude doing core codegen.
We use both Claude 4 and Gemini by default (for different tasks). But the idea is you can self-host this and use other models (and even BYOM - bring your own models).
Exactly. Non-AI projects have always been easy to build without issues. That's why we have so many build systems. We perfected it the first try and then made lots of new versions based on that perfect Makefile.
reply