So

VirtualMatrix

basically generates values on-demand using that supplier function? Interesting...

Is Python really this slow compared to Kotlin/Java unless you bring in library support?

Here is an article describing plotly.kt and my approach to use dynamic structures inside statically-typed wrappers: https://medium.com/@altavir/plotly-kt-a-dynamic-approach-to-visualization-in-kotlin-38e4feaf61f7

:trollface: https://hackernoon.com/why-python-is-an-excellent-choice-for-machine-learning-71acfa03f567

I am writing a conference paper about kotlin applications in particle physics, and I desperately need some references: • Some recent study about C++ applications. If someone knows an article (preferably with DOI), please give me a link. • Some examples of scientific work in kotlin. @thomasnield, do you have articles or books, I can reference? • References to kotlin (besides Kotlin in action) and kotlin libraries. I can add links to github, but it would be better to have DOI links like the ones, generated by https://zenodo.org/ (@elizarov is it possible to establish this integration?)

@altavir no articles or written content yet about Scientifik. I only have a handful of toy code examples that I still need to blog about

@altavir @Thomas Legrand my link was more focused on a comment where somebody was predicting Swift as a replacement for R and Python. But yes, I agree with both of you regarding R having little future. Python is definitely the only incumbent but it does need to be replaced by something

I was struggling finding a nice solution for plotting

@Zelenyi I've just pushed a commit to plotly.kt which works around the JavaFx dynamic update problem. It introduces a way to update plots without websockets. The client just requests the plot with regular interval. The reconcile is done on client side so the amount of data being sent is much higher, but on the other hand we do not work with websockets. I think in future I can make those pull updates as effecttive as push updates.

@dalexander sorry I've been off the grid lately. I've had some work keeping me busy and my career is possibly at an inflection point. I am teaching a class at University of Southern California next year and started a company on the side. I'll try to re-engage. https://www.nieldconsultinggroup.com/

Hello friends, this Netflix project may be interesting to you, especially if they can be convinced to add support for Kotlin https://medium.com/netflix-techblog/open-sourcing-polynote-an-ide-inspired-polyglot-notebook-7f929d3f447

By the way, in terms of referencing the code, HEP comunity decided to use https://zenodo.org/, it is pretty easy to setup and it automatically captures new releases.

The other part of science with kotlin: https://matrix.org/_matrix/media/r0/download/matrix.org/mWqvshReBiwJnkNPaSimbswf

It is not hard to add new instruments in TANGO. The code base is not quite the state-of-the-art, but it is working properly from the user point of view.

I'm using a lot of RDF/JSON-LD for other projects. That may be an useful way to communicate capabilities (and fallback if one UI system doesn't know about a specific device, but still knows how to handle a number of that type)

Are you using google docs? What's your preferred communication way for that kind of things?

@altavir I think it will be interesting for you https://github.com/Kotlin/kotlin-numpy

what I want is either the GUI knows how to deal with that device and display something like a graphical representation of states etc, or it doesn't know

videos are out ?

I switched my project to a MPP, ready to make the JS part so I don't have to describe the formats twice

I am thinking ubout upper level package name for scientifik applications. Right now I use

scientifik

, but it sounds odd. How about

SciKiT

? It sounds close to SciPy, has KT in it and sounds good.

it is extremely painful, the doc on kotlin-js is all messed up with the old way to handle things (that have been removed)

I think I will just scrape all the kotlinjs stuff

A small announce. Iv'e implemented a demo full-stack application for the project we have been working last four months under JetBrains Research banner (it is not the only one project, but the one we promised to finish before the end of January). The library is a 3d visualization (and not only 3d) engine for so-called event display (particle interaction visualization in accelerator and non-accelerator particle physics experiments). It will be used in accelerator experiments on NICA in Dubna and probably other experiments as well. Our lab will work a bit on documentation and then publish it. I expect to write a peer-reviewed article later. Many thanks to @elizarov and @olonho for helping with JBR funding for the project.

would that use SHM?