Hacker News Reader

Comment by camwest | Hacker News Reader

Hi! I’m the tech lead on AutoCAD web. The engine is C++ which is compiled to Web Assembly, our core team doesn’t actually work in JavaScript day to day.

The browsers are doing a really good job making this a possibility today.

View on HN

The problem is the reliance on the browser. If we look at the mobile market for one, you could do the same thing, but people still want native apps since they are more responsive, lower memory and cpu usage etc …. With web assembly all you are doing is adding more complexity. Keeping things native is preferred.

I would argue that the web is less complex than native, and by quite a bit. You need a client that runs on Windows, Linux (distributed through multiple package managers), and MacOS. Possibly iOS if you want to support the iPad, and particularly the iPad Pro, in any sort of capacity. And if you want to reliably touch Chrome OS, you're looking at web either way. Not everyone computes the same way they did 10 years ago and we have to start accepting that.

You imply that computing of ten years ago was arguably worse than what we have today. In what way?

If I did, I didn't mean to.

Though, using Waze to find the quickest way to work this morning really make my commute much nicer. There was also a new album I'd been meaning to listen to and streaming it on spotify was nice.

Note that the native AutoCAD doesn't seem to be going anywhere.

For now. I guess management is not doing it for fun but to explore profitability and sustainability of this approach. If it works for them, web version might be the only one that gets distributed. We already saw in Adobe how "cloud" is pushed everywhere even if it makes zero sense from the product point of view (uploading a single RAW file from a top-end camera to cloud takes ages and benefit is nearly zero), but makes super sense from economical standpoint.

> If it works for them, web version might be the only one that gets distributed.

How could it work for them? Most decent sized projects would promptly run out of available memory and have nothing to do about it. And before that they have to deal with the ever longer list of restrictions put on web applications because of the trust model.

I assume they are doing it mainly to avoid distribution costs; they simply push a web-app update via a single, well-standardized channel allowing DRM and trusted remote code injection (to prevent pirates from running it), and they don't need to maintain multiple internal tools to do that with multiple attack vectors. So this IMO might be an experiment in this fashion; they are working closely with Google, probably assuming they can't compete with them so it's best if Google's brightest figure out proper ways, and they can focus on their core business.

This is a 35-year-old multi-million-line codebase working in a browser.

Yup. I don't know what to say - I'm impressed by the amount of effort not spent on other things (for your infinite pick of other things).

The future will tell but in the meantime I don't see the problem that was solved by this fix (except for the problem of not running the app in Chrome).

The WASM group spec just added this in the official 1.0 spec. Non-web: https://webassembly.org/docs/non-web/ So this is how you get around cross-platform issues.

Bytecode as portable binary files go back to the first mainframe generations during the 60's and is still in use by IBM and Unisys systems.

Being partially copied by the same companies that are involved on WebAssembly for their own OSes (MSIL/MDIL, DEX, LLVM bitcode).

So nothing really new there for getting around cross-platform issues.

I think WASM is the saving grace here, really. It would be hard to pull of a performant CAD application in JS alone, but with WASM on the rise I don't see why this couldn't work.

>Hi! I’m the tech lead on AutoCAD web.

Hi, what will you do if one nice day Sundarajan Pichai will, for god knows what reason, order web assembly axed out of Chrome and lobby to bury it in W3C?

Reply on news.ycombinator.com