OMG, I went to HS with the inventor of Julia, we were on the same math team and calculus class. I graduated a year ahead of him. We are the same age but I skipped a grade. Even though I was really good a math, I hated graphs. He on the other hand drew graphs all day long. He was always doing something with mass points and/or complex numbers.
Fast, efficient, simple & easy are the core elements that always win. Julia, having all these, will gain the top position among most popular languages very soon.
@@agentetaeko1422 Never use the Gnu mentality to define languages, most of the times young languages tend to be better. Rust, nix, Go, Kotlin are just some of the examples I can show of how you can do amazing stuff with a small change, they do what C, C++, Java and others do in quite a better way. Also, Julia has been around for quite a while now, most of the big communities in science have migrated to julia, the code is a lot easier to write, read and maintain, it is also quite fast and expressive, something pure python won't achieve, at least not now nor in the near future.
@@IanNMejia Personally, I find python very fast and very expressive. I can not comment on those attributes as they relate to julia, yet, but I do think these are subjective terms. Rust, Go, and others are excellent languages to learn, but I do not think they necessarily do things in a better way than Java and C. What are these things?
@@satyabansahoo1862 Slow in what way? Do you mean in terms of calculations per second as compared to compiled languages? i would certainly not argue against that. I was referring to the above comment " they do what C, C++, Java and others do in quite a better way". Like i said, I found this to be subjective and not the best way to compare languages. I think i have worked with some pretty large datasets, and i did not choose python as a tool to work with all of them, but i also did not make that decision based solely on compuational effiency. Sometimes the ability to quickly train others/write/edit/release/update code is as important or more important than which language gives a result the fastest.
From this presentation julia sounds like unicorn dreamland. Computational modeling is inherently hard. Having worked extensively on different versions of finite element methods I am extremely skeptical on whether the math can simply be separated from the implementation as shown on the slides. In the performance world each problem has its own optimal algorithm, and it is exactly the very non-trivial machine-level tweaks like manual hash management, fine-tuning of relaxation paramteters, choosing the forwards integration scheme, etc ... that make the task hard. It is of course cute to be able to write equations as equations in the code, but that has never really been a problem. One simply writes the equations in some function, extensively tests that function, and never looks at its contents again.
Writing equations as equations in the code is not Julia's core idea (you can do it if you want but it is not the point), readeable code like Python's and good performance define what Julia is.
> _"readeable code"_ @@edgarlop aaah yes, I was mesmerized by what was written above, then ur comment snapped me back to reality, * yes, readability. * Also, no, ppl don't work like what's written above ... * since the reason Python is popular is because a huge proportion of ppl using it in academia, don't write all functions themselves, they use premade functions, and assemble equations from them * and those premade functions are getting increasingly complex * julia simplifies the non-mathy part, to counter some of that increasing complexity am i right??
Correct me if I am wrong; Julia is NOT fast on the first run. It becomes fast on the second run. And that is the big deal. It means any changes to Julia code has to be JITted first, the loops unrolled and the results cashed. It is like the Julia JIT compiler calculates the results of loops and arithmetics in the first run (slow) and in the second run, it outputs the cashed results.
You are somewhat correct. There is some JIT overhead for the first execution of a piece of code. Typically, the overhead is small and does not matter for long running simulations. As far as I am aware, the JIT compiler does not typically cache results. That would not be effective in most cases because the results of a function are not typically known in advance. The JIT compiler generates machine code. It would be accurate to say that Julia (especially 1.9 +) caches much of the JIT compiled machine code for packages so that there is less JIT overhead. This precompilation process may require a few seconds or a minute in extreme cases during the initial installation. The benefit is that it greatly reduces JIT overhead for installed packages.
Unfortunately he told us nothing at all about Julia. Why should I use it? What makes it different from Go or Rust or Java? Is it general purpose or domain specific like R or MatLab?
Not really that kind of video it seems (which is a bit strange, since presumably the people watching a ted talk about a language would want to actually learn about the language). There is lots of information elsewhere though.
I mean you can't exactly read the code on the screen because it's too small, so I think what he was referencing is that a lot of code for complicated math and physics applications looks daunting. But honestly, all code is like that until you understand it: looking at someone else's code for the first time is challenging because you have to decipher what is going on and interpret it, but if Julia turns out to be as effective as people say it is, then the readability will be much better than some abstract, long, messy code in another numerical simulation language, since you can write in equations and it is close to being as clean of a language as Python.
I dunno. This simple bit seems arcanish to me. s1 = "The quick brown fox jumps over the lazy dog α,β,γ" # again with a regular expression r = replace(s1, r"b[\w]*n" => "red") show(r); println() #> "The quick red fox jumps over the lazy dog α,β,γ"
I tried to learn Julia since I started to gain interest in Data Science and Machine Learning. But the syntax and the semantics of the language look very cryptic and strange (1-based indexing for arrays for example). So I quickly gave up. For someone like me, who has programmed in a lot of C-like languages, Python's syntax is a lot more tolerable than Julia's.
@@தமிழோன் From my experience, Python is for programmers, R is for statisticians, and Julia for scientists. That's drawing the lines a bit too harsh, but you get the direction. If you do Machine Learning, Python is great and has an excellent ecosystem of libraries. The libraries you use have their critical components optimized, they are written in C under the hood. However, Julia can run many scientific calculations natively with great performance which allows for flexibility outside of ML as well -- for example n-body gravity calculations for asteroid trajectories, or nucleotide searches in genetics. It does have serious drawbacks as well though: It's processing speed tends to come with very high memory requirements for example. "It's aimed at non-programmers" is very true, many statisticians, geneticists and astronomers don't see programming as their primary job, but rather as a tool to aid with their job.
If you are working with scientific computing, Julia has advantage because it is designed for it. As an example you can use DifferentialEquations.jl to work stuff related to differential equations.
@@maggieselbstschopfer1956 Indians who are good at math & logic *= may have larger brains with high density neural circuity 一> Need a bigger skull to hold these brains. 😉 I confess I am a big head asian (not indian) myself 😂 while not working for MIT (I wish someday, before I die, I had the "Freedom" to do so).
Because they have a CUDA.jl package and an AMDGPU.jl package. I wouldn't think one would use an IGpu for parallel computing, so I think CUDA and AMDGPU should have one covered. Also, OpenCL is still pretty well supported in Julia.
There's only 2 stages left to achieve. First the ability to use more varied pseudo code Then with some other software engine the ability to turn words and decisions into optimal code.
That's very hard to do, if not impossible, because natural language is ambiguous. You have to train yourself to be very specific in what you want when you're programming.
Well, I don't know. I am invested into Python, LabVIEW and C. Sometimes I have to use some Java or C#, for some platform specific stuff. Why should I add Julia to my portfolio? The languages like C and Python are not really hard to understand surely not much harder than Julia. The issue with C is that the problem of dependency management has not been resolved, but Python does that. Also Python has some projects that allows for really fast computing when there is need for that. I like the idea, but I feel like we're at a point where this language can only appeal to academic paper generator groups, even if it would be the best "compromise" between ease of use and performance - which doesn't have to be the case (look at Rust?).
1. The dependency problem -- Pkg, Julia's package manager, uses virtual environments for literally everything which can be activated with all the appropriate dependencies. 2. Not only is Julia as fast as C, it is as simple as Python. 3. Dispatch -- Julia is aimed at scientific computing, being able to use parametric polymorphism on just about anything means that any type can be applied to any application. Further more, explicitly imported methods and types can be extended -- and this is what he means when he says that code can be composable and work together without the developers creating it that way. 4. Paradigm -- Julia " uses multiple dispatch as a paradigm" although the language is by nature functional, you can create methods with dispatch from within a function and return a new type with those methods as children of a new type. A.K.A. You can do object-oriented programming when you want, functional programming when you want, and very statistical syntactical programming when you want. Heck you could even make pointers in the language and go full iterative if you need to. 5. Of course Julia is fast. It is also written in itself. Not only is it fast, but it is very fast to develop, as well.
Furthermore, you can get Lisp-style functional with Julia using macros and meta, so it is quite powerful because you could combine meta parsing with decorators and subtypes. I would say if you enjoy C and Python, that is fine, what I like about Julia is that I can be creative in how I approach any given programming problem, and the language isn't ever in the way of that creativity. However I want to store, modify, or move data is completely up to me. I am not forced to use structs, classes, or decorators.
@@emmaccode Julia can't be faster than C, I guess that you don't do low-level programming. What could be true is that it is easier to write faster code in Julia than in C.
@@leoalmeida2583 Julia CAN be faster than C, you should look at the Julia benchmarks for the data that supports that. Saying that Julia can't be as fast as C as you say with data would be one thing, but the data seems to back me up on this -- Julia is usually a bit slower than Rust, but it can, depending on the application, run faster than C, or in some cases slower than LUA with JIT, it just depends on the scenario. And define low-level programming... As in iterative? I write NASM Assembly, man, I'm pretty sure I can write C. What I'm saying is factual and yet still I am hearing you tell me it is not, I can't say much other than do your own research on the language benchmarks and come to a conclusion. You also underestimate LLVM, it is quite a robust compiler library.
@@emmaccode no matter what code you write on Julia, I can achieve the same performance with C (C++ and Rust), probably with more work required. I never wrote any Julia code but I can imagine that you can't write for example a memory pool with Julia because of the limitations of being too high level. Yes, Julia can be faster than C and in many cases will be faster. As someone once said "C doesn't give you performance, it gives you control over performance."
Fairly weak presentation, especially given the audience. Claiming you can have your cake and eat it too, is a bit too bold without any further explanation. From the looks of it, Julia seems to be a highly constrained and optimized for the computational modeling domain. It's highly likely its strengths, namely high-level programming without compromising performance, don't carryover to other domains.
Just as I thought I had found the perfect programming language, I saw the keyword "end". (Terribly cluttered in the eyes of someone looking for something cleaner than python.)
"cleaner than python" is a high goal here. I also wondered why they didn't use pythons approach of code block definition. But hey, you could write your own macro (meta-program) in julia that lets you omit the "ends"
Nothing useful here. I expect something more revolutionary like human understand-able graphical programming nodes, built with easy to make modules, interconnected with simple math formula "links" / lambda funcs , sort of..... I may be asking for too much..... 😅😅
Nothing really special here. Pls point me out if I am wrong. I expect something more revolutionary from MIT, like human understand-able graphical programming nodes, built with easy to make modules, interconnected with simple math formula "links" / lambda funcs , sort of..... I may be asking for too much..... 😅😅
Julia: clean like python, fast like C. Not for use in software dev, but rather for use in science, data, and math.
OMG, I went to HS with the inventor of Julia, we were on the same math team and calculus class. I graduated a year ahead of him. We are the same age but I skipped a grade. Even though I was really good a math, I hated graphs. He on the other hand drew graphs all day long. He was always doing something with mass points and/or complex numbers.
what are mass points??
@@yash1152 They are lunar coordinates equated by orbital inclination denoted by the equatorial plane
@@francescozhu917 wow, do such things really come in HS (high school)?? (which i am thinking to be equal to 12th standard)
Fast, efficient, simple & easy are the core elements that always win. Julia, having all these, will gain the top position among most popular languages very soon.
Julia is not just the programming language for math, it is the math.
Julia programming language looks easy like python and its faster like c programming .. i am very much interested to learn
Its kinda new, so its better to learn something more old and come to julia if you feel like you need this lang
@@agentetaeko1422 Never use the Gnu mentality to define languages, most of the times young languages tend to be better. Rust, nix, Go, Kotlin are just some of the examples I can show of how you can do amazing stuff with a small change, they do what C, C++, Java and others do in quite a better way. Also, Julia has been around for quite a while now, most of the big communities in science have migrated to julia, the code is a lot easier to write, read and maintain, it is also quite fast and expressive, something pure python won't achieve, at least not now nor in the near future.
@@IanNMejia Personally, I find python very fast and very expressive. I can not comment on those attributes as they relate to julia, yet, but I do think these are subjective terms. Rust, Go, and others are excellent languages to learn, but I do not think they necessarily do things in a better way than Java and C. What are these things?
@@ChrisWoodBandit then u might not have worked big data or heavy programs... if so, u will see how slow it is.
@@satyabansahoo1862 Slow in what way? Do you mean in terms of calculations per second as compared to compiled languages? i would certainly not argue against that. I was referring to the above comment " they do what C, C++, Java and others do in quite a better way". Like i said, I found this to be subjective and not the best way to compare languages. I think i have worked with some pretty large datasets, and i did not choose python as a tool to work with all of them, but i also did not make that decision based solely on compuational effiency. Sometimes the ability to quickly train others/write/edit/release/update code is as important or more important than which language gives a result the fastest.
Ease, Speed.. and there is a third factor as well, i.e. Size. Julia has scope of improvement in generating a smaller size of programs.
This is a very important TEDx Talk. Why so few in the audience?
The cool thing is his first program was Hello world.
Julia is an epitome of human ingenuity. we humans always seeks for perfections and this is the result of it.
My name is Julia so this is super fun
Writing code with JuliaLang is easy. Optimization and morphism design is hard.
umh, what is morphism design?
From this presentation julia sounds like unicorn dreamland. Computational modeling is inherently hard. Having worked extensively on different versions of finite element methods I am extremely skeptical on whether the math can simply be separated from the implementation as shown on the slides. In the performance world each problem has its own optimal algorithm, and it is exactly the very non-trivial machine-level tweaks like manual hash management, fine-tuning of relaxation paramteters, choosing the forwards integration scheme, etc ... that make the task hard. It is of course cute to be able to write equations as equations in the code, but that has never really been a problem. One simply writes the equations in some function, extensively tests that function, and never looks at its contents again.
Writing equations as equations in the code is not Julia's core idea (you can do it if you want but it is not the point), readeable code like Python's and good performance define what Julia is.
> _"readeable code"_
@@edgarlop aaah yes, I was mesmerized by what was written above, then ur comment snapped me back to reality,
* yes, readability.
* Also, no, ppl don't work like what's written above ...
* since the reason Python is popular is because a huge proportion of ppl using it in academia, don't write all functions themselves, they use premade functions, and assemble equations from them
* and those premade functions are getting increasingly complex
* julia simplifies the non-mathy part, to counter some of that increasing complexity
am i right??
Correct me if I am wrong; Julia is NOT fast on the first run. It becomes fast on the second run. And that is the big deal. It means any changes to Julia code has to be JITted first, the loops unrolled and the results cashed. It is like the Julia JIT compiler calculates the results of loops and arithmetics in the first run (slow) and in the second run, it outputs the cashed results.
You are somewhat correct. There is some JIT overhead for the first execution of a piece of code. Typically, the overhead is small and does not matter for long running simulations. As far as I am aware, the JIT compiler does not typically cache results. That would not be effective in most cases because the results of a function are not typically known in advance. The JIT compiler generates machine code. It would be accurate to say that Julia (especially 1.9 +) caches much of the JIT compiled machine code for packages so that there is less JIT overhead. This precompilation process may require a few seconds or a minute in extreme cases during the initial installation. The benefit is that it greatly reduces JIT overhead for installed packages.
WoW....!!...Julia Got Super Power...!
Unfortunately he told us nothing at all about Julia. Why should I use it? What makes it different from Go or Rust or Java? Is it general purpose or domain specific like R or MatLab?
Not really that kind of video it seems (which is a bit strange, since presumably the people watching a ted talk about a language would want to actually learn about the language). There is lots of information elsewhere though.
I felt the same, what was the topic supposed to be?
can it be used for red team tool development?
Let's always do good 🙏
Why is he reading? He is usually a very good speaker.
Maybe lack of time? Very urgent to read someting but no enough details for listeners to understand.
How come this man shows messy code and says "oh look how messy this is" but the code is also written in julia?
I mean you can't exactly read the code on the screen because it's too small, so I think what he was referencing is that a lot of code for complicated math and physics applications looks daunting. But honestly, all code is like that until you understand it: looking at someone else's code for the first time is challenging because you have to decipher what is going on and interpret it, but if Julia turns out to be as effective as people say it is, then the readability will be much better than some abstract, long, messy code in another numerical simulation language, since you can write in equations and it is close to being as clean of a language as Python.
I dunno. This simple bit seems arcanish to me.
s1 = "The quick brown fox jumps over the lazy dog α,β,γ"
# again with a regular expression
r = replace(s1, r"b[\w]*n" => "red")
show(r); println()
#> "The quick red fox jumps over the lazy dog α,β,γ"
the source code julia is really nice to read.
Wow, what a setting to host a Ted talk!!
It sounds like a declaration of independence or something like that.
Have you tried it yet?
manifesto. ideology.
Isn't "codes" something different? Since when is it used as plural for code?
This is much later but it's very common in high performance computing and numerical computing
cameraman is amateur
I agree that they probably don't have much experience in terms of recording talks like of this sort. Makes the talker look weird :D
Michael Bay
He seems to work really hard, btw :)
He watched the Matrix too many times
@@aparup53 ... *spittake* :D
I tried to learn Julia since I started to gain interest in Data Science and Machine Learning. But the syntax and the semantics of the language look very cryptic and strange (1-based indexing for arrays for example). So I quickly gave up. For someone like me, who has programmed in a lot of C-like languages, Python's syntax is a lot more tolerable than Julia's.
R is 1-based indexing as well and is here at the 20 most used languages. 0 indexing is really weird for nom programmers.
@@igordemetriusalencar5861 I see. So R, and Julia are mostly aimed at non-programmers then.
@@தமிழோன் From my experience, Python is for programmers, R is for statisticians, and Julia for scientists. That's drawing the lines a bit too harsh, but you get the direction. If you do Machine Learning, Python is great and has an excellent ecosystem of libraries. The libraries you use have their critical components optimized, they are written in C under the hood. However, Julia can run many scientific calculations natively with great performance which allows for flexibility outside of ML as well -- for example n-body gravity calculations for asteroid trajectories, or nucleotide searches in genetics. It does have serious drawbacks as well though: It's processing speed tends to come with very high memory requirements for example. "It's aimed at non-programmers" is very true, many statisticians, geneticists and astronomers don't see programming as their primary job, but rather as a tool to aid with their job.
@@odw32 What about LISP??
@@AcidiFy574 No pls not lisp.
Julia: walks like Python, runs like C!
Please someone teach me Julia.
There is a few youtube channels are available to learn julia language ..
Jayakumar Nagaraj Thanks a lot handsome. You teach me I will teach you too
Why not my friend.. I am happy to help others..
Im glad I don't make JL tutorials...
Now that...
That would be bad.
@@emmaccode rofl
Can anyone tell me the advantages of Julia compared to Rust?
Use Rust when you're building software. Use Julia when you're analyzing datasets.
If you are working with scientific computing, Julia has advantage because it is designed for it. As an example you can use DifferentialEquations.jl to work stuff related to differential equations.
@@loreleipenn That's the best answer that I received! Thank you!
@@dmitriidemenev5258 My pleasure that I was able to share that bit of knowledge. 😄
You are welcome. 🙂
8:06 Code is math
ohsommme
Yo Julia!
Viral Shah An Indian 🇮🇳 PhD Student at Santa Barbara is the CO-FOUNDRR & CO-CREATOR of JULIA.
Yes. He is. Should be proud of him!
Yes he is, the co-founder isn't one guy, there are like 4 of them
Why are Indians so bighead?
@@maggieselbstschopfer1956 what's a big head? I don't understand the Slang (Short Language) you're using.
@@maggieselbstschopfer1956 Indians who are good at math & logic *= may have larger brains with high density neural circuity 一> Need a bigger skull to hold these brains. 😉
I confess I am a big head asian (not indian) myself 😂 while not working for MIT (I wish someday, before I die, I had the "Freedom" to do so).
Why did Julia drop OpenCL support?
Because they have a CUDA.jl package and an AMDGPU.jl package. I wouldn't think one would use an IGpu for parallel computing, so I think CUDA and AMDGPU should have one covered.
Also, OpenCL is still pretty well supported in Julia.
There's only 2 stages left to achieve. First the ability to use more varied pseudo code Then with some other software engine the ability to turn words and decisions into optimal code.
That's very hard to do, if not impossible, because natural language is ambiguous. You have to train yourself to be very specific in what you want when you're programming.
@@wyqtor or to gain a more specific grasp of what the user wants the system could also question the user.
domain specific languages are a thing, and with julia's metaprogramming support (in theory, i haven't actually used it) you could make any sort of dsl
sounds like a trainwreck! articulation and precision is much more valuable in computation than having a language try to guess what you want to do.
Now i know why i find it is so hard to code.... Cos most lang when written is not intuitive by nature as humans learn to communicate with one another
Well, I don't know. I am invested into Python, LabVIEW and C. Sometimes I have to use some Java or C#, for some platform specific stuff. Why should I add Julia to my portfolio?
The languages like C and Python are not really hard to understand surely not much harder than Julia. The issue with C is that the problem of dependency management has not been resolved, but Python does that. Also Python has some projects that allows for really fast computing when there is need for that.
I like the idea, but I feel like we're at a point where this language can only appeal to academic paper generator groups, even if it would be the best "compromise" between ease of use and performance - which doesn't have to be the case (look at Rust?).
1. The dependency problem -- Pkg, Julia's package manager, uses virtual environments for literally everything which can be activated with all the appropriate dependencies.
2. Not only is Julia as fast as C, it is as simple as Python.
3. Dispatch -- Julia is aimed at scientific computing, being able to use parametric polymorphism on just about anything means that any type can be applied to any application. Further more, explicitly imported methods and types can be extended -- and this is what he means when he says that code can be composable and work together without the developers creating it that way.
4. Paradigm -- Julia " uses multiple dispatch as a paradigm" although the language is by nature functional, you can create methods with dispatch from within a function and return a new type with those methods as children of a new type. A.K.A. You can do object-oriented programming when you want, functional programming when you want, and very statistical syntactical programming when you want. Heck you could even make pointers in the language and go full iterative if you need to.
5. Of course Julia is fast. It is also written in itself. Not only is it fast, but it is very fast to develop, as well.
Furthermore, you can get Lisp-style functional with Julia using macros and meta, so it is quite powerful because you could combine meta parsing with decorators and subtypes.
I would say if you enjoy C and Python, that is fine, what I like about Julia is that I can be creative in how I approach any given programming problem, and the language isn't ever in the way of that creativity. However I want to store, modify, or move data is completely up to me. I am not forced to use structs, classes, or decorators.
@@emmaccode Julia can't be faster than C, I guess that you don't do low-level programming. What could be true is that it is easier to write faster code in Julia than in C.
@@leoalmeida2583 Julia CAN be faster than C, you should look at the Julia benchmarks for the data that supports that.
Saying that Julia can't be as fast as C as you say with data would be one thing, but the data seems to back me up on this -- Julia is usually a bit slower than Rust, but it can, depending on the application, run faster than C, or in some cases slower than LUA with JIT, it just depends on the scenario.
And define low-level programming... As in iterative? I write NASM Assembly, man, I'm pretty sure I can write C. What I'm saying is factual and yet still I am hearing you tell me it is not, I can't say much other than do your own research on the language benchmarks and come to a conclusion.
You also underestimate LLVM, it is quite a robust compiler library.
@@emmaccode no matter what code you write on Julia, I can achieve the same performance with C (C++ and Rust), probably with more work required. I never wrote any Julia code but I can imagine that you can't write for example a memory pool with Julia because of the limitations of being too high level. Yes, Julia can be faster than C and in many cases will be faster. As someone once said "C doesn't give you performance, it gives you control over performance."
* pandemic models too
julia is the language who introduced the term "time to first plot".
so thanks but no thanks
What is mathematica for
I think this is open source and mathematica not
Mathematica is mainly for symbolic computing
Mathematica is even slower than Python in most cases
@@blugio oh okay
@@MarcelloZucchi91 okay
C++ common in the business world? Not. But Julia seems like a very good idea.
In defence. Finance. Yes. C/c++ rules the world. If you don't know that... You're not a programmer...
@@jonathanmoore5619 Who said anything about C?
@@jonathanmoore5619 Yes , large parts of tensorflow is also in C++
@@jonathanmoore5619 And you know programming? Is that why you've stitched a C to C++ like C/C++ when they are widely different languages THESE DAYS.
@@balen7555 don't be a silly billy. You're too young to know anything about anythiink! Bedtime now... Bed ways is best ways for you, little droogy.
If you wonder about Julia as a programmer, this video is not what you are looking for. Leave now. Don't waste your time.
So Julia's syntax is simply math symbols?
No, but it supports them!
And emojis
For example:
♥️(😝) =2*😝
Is valid Julia code
Fairly weak presentation, especially given the audience. Claiming you can have your cake and eat it too, is a bit too bold without any further explanation. From the looks of it, Julia seems to be a highly constrained and optimized for the computational modeling domain. It's highly likely its strengths, namely high-level programming without compromising performance, don't carryover to other domains.
❤
soy venezoalno , un sistema aprendizaje puede tener alma en el chip segun su intruciones y parametros de comportamiento logiica
Just as I thought I had found the perfect programming language, I saw the keyword "end". (Terribly cluttered in the eyes of someone looking for something cleaner than python.)
"cleaner than python" is a high goal here.
I also wondered why they didn't use pythons approach of code block definition. But hey, you could write your own macro (meta-program) in julia that lets you omit the "ends"
I personally hate Python for not having curly braced or ended blocks
I sold my car [as a Julia co-founder]
2:53 oh scratch, i dont like it at all
what a flop
Wait until he hears about how aerosols are actually cooling the Earth LOL
As a programmer I tell you, do not waste your time with this. Has nothing wort it in terms of information you need
Definitely not the video to come to for information.
Nothing useful here. I expect something more revolutionary like human understand-able graphical programming nodes, built with easy to make modules, interconnected with simple math formula "links" / lambda funcs , sort of..... I may be asking for too much..... 😅😅
Something like LabVIEW? It is not opensource however, used on real hardware though and by CERN.
I love how newbies are comparing Julia with C 😂
?
He is kind of cute, but I can not get rid of the feeling that he is also a little poser
Nothing really special here. Pls point me out if I am wrong. I expect something more revolutionary from MIT, like human understand-able graphical programming nodes, built with easy to make modules, interconnected with simple math formula "links" / lambda funcs , sort of..... I may be asking for too much..... 😅😅