Pluto 2:23:49

Am I right in understanding of Julia convention, that the birthday function should have actually been named birthday! because it acts to change its argument?

Thanks

Thank you!!!

Funny when you defined g as the plus function: g = + and then used g(3, 4), I stopped the video to try 3 g 4 - like 3 + 4. If I let played a few seconds longer, I would have seen, nope. I wasn't thinking like someone in audience.

Great tutorial! It would be great to know which minutes of the video correspond to which topic.

Hi can you show how you mesh your model in 3D ?

@sanbalestrini hi

Amazing! It's very simple to install!

Great presentation! Thanks for your hard work, and I have read almost all of your papers on DEX routing and pathfinding.

2:15 That's what she said.

07:00:33 horus

Swag

This whole concept can clearly be shown within a few lines of super simple to understand code. Showing the reason this works so well. I know this is what you think you have here but, to a new coder this is far to complicated. When showing/explaining this style of programming, I also like to be going over top down programming. They fit together well.

not enough people using this.

shit is built in fortran because ur not beating the linalg libraries. that's it. the only reason. hundreds of thousands of man hours.

Hi, I'm new to julia and have to run a simulation that was written for julia 0.6. I was hoping that I can use juliaup to do that although as far as I understand 0.7 is the oldest available version. The problem I have right now are the old packages needed for the simulation like Dierckx and ProgressMeter. I just can't find any way to add them to the environment: (v0.7) pkg> add Dierckx ERROR: The following package names could not be resolved: * Dierckx (not found in project, manifest or registry) Please specify by known `name=uuid`. Also downloading old versions is not possible for a lot of packages because they are simply not available anymore. Is there an easy way I just don't know? Would be glad if someone knows an answer. best

Excellent tutorial! Thanks

Make a new programming language with multiple dispatch and borrow checking named Trust. Jokes aside, great presentation! I've downloaded Rust and will be giving it a shot.

where are the code from this channel posted? which github repo or which website?

where is the code?

Thanks for your presentation, i am doing my final year project, "predictive modle for water quality using rheological properties of the water " I found RHEOS ,julia open source can apply for it.i don't have experience about modle creation, but I done through MATLB,can you please give me a guide,or relevant direction,

I am planning to use Javis now, but unfortunately the development has been idle for a couple of years on this project. This seems a lot more intuitive than Manim.

Great talk. For sure Julia is at begin of life (also 5 years in the future, seen from the timeline of the video) and C++ is becoming more end of life (I say this - as a senior C++ specialist - because, whatever is added to C++, it was never designed for the current multi core architectures and HPC we have nowadays). But C++ is still the king here. You can archive the same thing with template specialization. With pro`s en contra`s. The pro is that you do not need dispatching. The abstract types are the templates whereas the instantiated types are the specialized argument types. From the C++ perspective this is a big plus, because there is no runtime overhead. No need to search through functions that match the pattern at runtime. But there are 2 disadvantages which fall in favor of Julia. The first is obvious, template specialization is hard to grasp and - with concepts on C++20 - it becomes even harder. The syntax is complex. Second, even if as good as, or arguable better then ...., it will become hard to convince teams to the common practice in the entire code base (what would be needed to come on the equal level as Julia). So in practice it will not be applied that way. But let finish in favor of C++. I have a choice what to do, static MD with template specialization or runtime polymorphism, and in fact many more options. Anyway, this sheds a new insight and a new light to system software architectures, maybe use of POD structures with template specialization becomes the future for maintainable C++ code bases ;-) . Not so much the end of C++ rather more that of initial Object Oriented intentions.

Why are you frying? To mimic a man???

Koobide? Nice choice

⭐️⭐️📚📚⭐️⭐️📚📚📚

📚📚📚📚📚⭐️⭐️📚📚📚📚📚

ColorQuantization

This seems very promising. Is it possible to use directly the code in an embedded device??? Or it must be transpile to some other language like C. Some experiences using Jetson or similar?? Anyway, the tool looks great.

Could someone please explain the picture reference in the title slide?

Nice

Nice lecture. Please add saddle-node points to your pictures with red and blue arrows.

7:14:40 EditBoundary.jl

Very interesting! 👏🏼👏🏼👏🏼

I see a lot of these macros being thrown about. Is this language becoming like Java where every program is littered with countless "kinda common but not standard" libraries? Code reuse is obviously a useful thing, but having every program entangled with 100 different github projects is not good for maintainability, security or comprehension. Or, maybe I misunderstand?

I'm amazed by your work.

Thanks this looks powerful ! Can trixi also adress simpler models such classic as linear or non linear static?

5:32:33 Hydrological modelling

6:36:45 PythonJL

6:04:00 Julia Kitchen

The github link is a 404. Looks like there are no public repositories on that account.

awesome

Great Tutorial,Keep the fire burning 💕♥️

As awesome as Frames' contribution to Julia is, you might want to change this video's title to match the actual speaker's name.

Man, I wish all tutorials were as clear and concise as this one. Great job! Excited to get to grips with Julia...

In my experience of working with ITensor.jl , albeit it's elegance, it has quite a lot of disadvantages. The indices being separate entities rather than just dummy symbols is a double edged sword. It means you need to change your code significantly whenever index permutations are involved, and then the expressions start to significantly diverge from the easily readable einstein notation. I am also skeptical about it's speed, in my experience, it has always been outperformed by TensorOperations.jl or TensorKit.jl, especially considering the former also supports TBLIS.

Is there a presentation about Julia's state and development?

01:26:15 Accelerate Insights : Code→ Slides, Faster ! 01:36:00 Solving intergral equations with Inti.jl 02:09:20 Building Confidently in Julia with Interface Driven Design 02:38:00 AppBundler.jl - Bundle your Julia GUI Application 04:36:00 Unlocking lightning speed of Julia’s Dict with Robinhood Hashing 04:45:00 Simulation non-particle trajectoires using CMInject.jl 04:55:30 Natrual Science, Computer Modeling and Literate Programming 05:07:00 FUSE.jl : the power of Julia for the fusion industry 05:39:00 Geant4.jl - Particle transport in Julia 06:16:50 Lyotropic Liquid Crystals : Thermodynamics & Numerical methods 06:26:40 Efficient, composable solver for non-equilibrium flows 06:38:20 Jet Reconstruiction in Julia (not in this video) WGPU graphics and compute API in Julia 07:35:20 FastIce.jl : a massively parallel ice flow model running on GPUs