Well, I just jumped full time on IronCalc[1] a fully open source, light and fast spreadsheet engine designed and build from the ground up.

I have been working on it as side project for over two years and now, with funding from the EU for the next 2.5 years, I hope I can make of it a real product for everyone to use that can compete with the likes of Excel and Googl;e Sheets.

I can oly say, I am overly, off the Moon excited

[1]: https://www.ironcalc.com

Looking at it and very excited. In unsupported features, charting is mentioned. Could there be any value in not directly implementing the drawing of charts, but tie in to other open source library? Just curious of your thinking.

I think charts is one of those few things I won't implement from scratch, as there are already fantastic libraries out there:

https://github.com/ironcalc/IronCalc/issues/348

We will start working on it by July according to the plan. (This will add a lot if value to the project BTW)

Which EU grant did you receive? Ie. from which fund?

edit: nm, rtfm, it was on the landing page: Horizon Europe programme

Answered to a sibling comment. NLnet and HORIZON.

NLnet is just amazing and can keep you going if you are a student or have some extra sources of income

HORIZON is a huge grant but fairly hard to obtain. Generally related to reasearch grants in academia

Oh, neat. Didn't expect to see IronCalc here for some reason, have been considering it for a side-project. Thanks for the hard work. :)

Interesting, how was the EU funding process?

First we got a grant from the NLnet[1], which I highly recommend as a first step of any project. Single best thing I could have done. That wasn't enough money for me to quit my job. Also I didn't have any _evidence_ that IronCalc was a good idea or that there was a market for it. Then evidence started pouring and I kept working. I started talking to different folks, lots of people many of those were contacts through the NLnet. Then the folks from NextGraph[2] approached us and asked, "Hey do you want to be part of this consortium [3]?". Eventually we got a HORIZON grant after a lot of sweat and paperwork, but NextGraph took the brunt of it.

As you see there is a huge component of sheer luck

[1]: https://nlnet.nl/project/IronCalc/ [2]: https://nextgraph.org/ [3]: https://elfaconsortium.eu/

Congrats! But what is a spreadsheet "engine" as opposed to a spreadsheet program?

The "engine" is the computational part of it. And it is completely separted from the UI. You can use it from Rust, Python, nodejs or from the browser and eventually from a destop app.

The important thing is that is all those cases the engine is the same. I

I'm sure it's a great book :).

I find good popular books on higher mathematics difficult to come by. A nice exception is the trilogy written by Avner Ash and Robert Groß:

Elliptic Tales, Fearless Symmetry and Summing it up (in my order of preference)

Of possible interest

https://mirtitles.org/2024/05/11/little-mathematics-library-...

Mir titles is a feeling. It's nostalgia. It's childhood for many including me. The books are excellent written by some of the finest of their times talking to the lowest highschool level student or even a child and making him or her understand fully what all is going on. Indians love Mir.

My favorite author is Landsberg. He is in Mir titles. He got defeated by our main man C V Raman by 2 weeks to publish the same research (independently) which got C V Raman the only Physics Nobel Prize for India.

Sorry for the stupid question but is Elliptic Tales your favorite or is it Summing it up?

Elliptic Tales, but maybe it is because I read it first

Thanks for the mention! That's indeed the plan

Hi! Congratulations to you and Yousef. And I am lucky enough to be in a position from learning from both of you.

Anyone think what they might about La Suite, but blocknote is a solid product!

Very much appreciated! We put a lot of effort into it!

Related (60 comments, How slide rules work)

https://hackernews.hn/item?id=45985196

Aaron Swartz has a nice blog post about it[1].

It's been discussed several times on HN[2]. I had periods I go through without news. It's been harder to do that lately.

[1]: http://www.aaronsw.com/weblog/hatethenews

[2]: https://hn.algolia.com/?q=i+hate+the+news

I submitted that link. In all fairness the link in this post is more accurate. I'm glad to see people are interested. Let's see where all this goes.

Eli Bendersky's post are always insightful and interesting.

I really would like to see a small language that compiles to wasm in the browser.

Of course you can use things like Lua that has it's own vm also in wasm. Or Rhai with it's own interpreter. But I am looking for a language that compiles to wasm in less than 1Mb of wasm

If you're open to Forth,

https://github.com/remko/waforth

> WAForth is entirely written in (raw) WebAssembly*, and the compiler generates WebAssembly code on the fly.

* https://github.com/remko/waforth/blob/master/src/waforth.wat

Ben Lynn's page https://crypto.stanford.edu/~blynn/compiler/ compiles (a large subset of) Haskell to web assembly (which you can download; a prime number sieve yielded 40KB of code) and runs it in the browser.

That is exactly the kind of thing I was looking for. Thanks you!

WebAssembly Text Format (wat) is fine to use. You declare functions that run imperative code over primitive i32/i64/f32/f64 values, and write to a block of memory. Many algorithms are easy enough to port, and LLMs are pretty great at generating wat now.

I made Orb as a DSL over raw WebAssembly in Elixir. This gives you extract niceties like |> piping, macros so you can add language features like arenas or tuples, and reusability of code in modules (you can even publish to the package manager Hex). By manipulating the raw WebAssembly instructions it lets you compile to kilobytes instead of megabytes. I’m tinkering on the project over at: https://github.com/RoyalIcing/Orb

I'm working on a TypeScript/Swift/Dart style language, and currently this hello-world is 1444 bytes:

    export let main = () => {
      console.log("Hello, World!");
    };

I'm trying to make that smaller. The binary includes the Console class, which is needed (I may be able to tree-shake the non log() methods away), but also the Error and IndexOutOfBoundsError classes which aren't needed because there are no catch() expressions.

I think it really helps to have a language designed from the ground-up to obsess over bytes for WASM. Trying to do that with a familiar high-level language with a rich standard library is tricky.

you can just compile c/c++ to wasm in the browser - there are wasi/emscripten builds of clang itself around (yosys, clang-repl, etc).

Yes, those are fascinating technologies. But way too big to be running in a small app in the browser.

> compiles to wasm in less than 1Mb of wasm

Janet, a Clojure-like Lisp compiles a whole playground of itself and the std lib in 823kb: https://janetdocs.org/playground

https://codeberg.org/veqq/janetdocs/src/branch/master/public...

you can do smaller for other things.

C based Mandelbrot WASM demos can be ~1 KB total. Assuming you mean a simple scripting language though, Assembly Script does exactly that.

AssemblyScript ships its own garbage collector and doesn't seem to making progress on supporting WASM GC.

AssemblyScript in general seems to have stopped making much progress. Unless you benefit greatly from having exact width numeric types it's very difficult to justify using it over normal TS.

Sorry! It was too beautiful not to share it. But I'm sure you can do a show HN once you think is ready.

That's okay, thanks for sharing!

Gossip tense has been discussed here:

(276 points and 255 comments) https://hackernews.hn/item?id=41793485