Live data from Hacker News

Introducing Atom-IDE

blog.atom.io

311–320 of 324 posts

Re: Introducing Atom-IDE

#311

Honestly can't see why I'd use this - I have all the professional grade IDEs I need, why do I need one that is written in JS, is slower and consumes 5x more memory? Not to mention IDEs already have their own extension stores too. This seems like an 'IDE-lite', which doesn't make much sense in a world where professional grade IDEs are free.

I concur. Although the thing that troubles me most about them is that for both Atom and VSCode, the binaries that are distributed by their parent companies are not a direct compile of the public source code. They use the MIT license to their advantage, clone the repo to make it private, make changes to that private repo and then distribute the binary compiled from it. In this sense, among the popular options only Vim…

This is demonstrably false. Atom's official distributions are built on public CI platforms (Circle, Travis, and AppVeyor) from 100% public repos.

Re: Introducing Atom-IDE

#312

Honestly can't see why I'd use this - I have all the professional grade IDEs I need, why do I need one that is written in JS, is slower and consumes 5x more memory? Not to mention IDEs already have their own extension stores too. This seems like an 'IDE-lite', which doesn't make much sense in a world where professional grade IDEs are free.

I wish they'd put this sort of manpower towards fixing Electron.

Too bad Mozilla is discontinuing Positron. With Servo coming up and Firefox nightly using Stylo it feels very fast compared to Chrome.

Re: Introducing Atom-IDE

#313

Earlier quoted context omitted.

If I recall the argument against Facebook here, the point is: let's assume that you found a company, file for a patent and implement it using some open source tools/libraries, also backed by Facebook. Now somewhere down the line FB decides that your product is really great and they can make something similar work for their platform, so they pour resources into developing their version of your solution. You are pissed…

Wow. If I understand this correctly, if I use their modules then I can't ever sue the company? :O

That's why some people cannot use React.

Re: Introducing Atom-IDE

#314
post #244

Earlier quoted context omitted.

So it's not a daemon or a server. It's just a process. Running some code in a separate process isn't a 'server' it's just how all programmes operate.

In computing, a server is a computer program or a device that provides functionality for other programs or devices, called "clients".

across a network in modern parlance

Re: Introducing Atom-IDE

#315
post #238

Earlier quoted context omitted.

This is tangential, but the IntelliJ Rust plugin is fantastic, and just got officially supported! ( http://intellij-rust.github.io )

If you check the changelog [1] you'll see the rust plugin has been officially support for a little over a year. 1. https://intellij-rust.github.io/thisweek/

I meant "officially supported by JetBrains" (https://intellij-rust.github.io/2017/07/31/changelog-47.html)

Re: Introducing Atom-IDE

#316
post #311

Earlier quoted context omitted.

I concur. Although the thing that troubles me most about them is that for both Atom and VSCode, the binaries that are distributed by their parent companies are not a direct compile of the public source code. They use the MIT license to their advantage, clone the repo to make it private, make changes to that private repo and then distribute the binary compiled from it. In this sense, among the popular options only Vim…

This is demonstrably false. Atom's official distributions are built on public CI platforms (Circle, Travis, and AppVeyor) from 100% public repos.

Oh sorry, I suppose I misunderstood something. It indeed is made from public repos.

But then again, I researched a bit and found the licenses to be a bit problematic, not that of Atom itself but the additional packages that are bundled in the binary. It includes some packages whose licenses are unknown and others such as Facebook's libraries(a lot of them) which have the non-free BSD+patents license. In totality the license of the binary doesn't seem to be a really good one.

Re: Introducing Atom-IDE

#317
post #104

Earlier quoted context omitted.

So why did the developers of a text editor decide on a web browser stack for their project? Yes, I know that the answer is "electron". They picked the wrong tool for the job.

This is actually backwards: Electron was originally created for the Atom project and even used to be called Atom Shell.

And it was the wrong tool for the job even then. Atom might be Electron's reason for existing, but that just emphasizes that Electron (in its current form) should not exist at all.

I have no problem with the idea of using Javascript for desktop application development. I do have a problem with relying on a memory-hogging framework, which Javascript was originally design to _avoid_ because back in the 90's it was meant to run in the resource-limited web browser. Too bad that Javascript has strayed so far from its roots.

Re: Introducing Atom-IDE

#318

Earlier quoted context omitted.

I love IntelliJ for most languages. When it comes to JS/HTML/CSS though I use VS Code... I find it's IntelliSense and TypeScript support superior. Especially when I'm on a project using webpack/sass/etc. IDEA just seems to get lost. It either can't find a method definition or is constantly complaining about various things.

Did you try WebStorm? Idea is a Java IDE. AFAIK you don't get the same experience from Idea+Plugins that you get from the dedicated version (in this case WebStorm). If you extend Idea with plugins you get what's available for the limited community editon.

I have IDEA Ultimate, which, as you said, allows the same functionality via plugins but isn't the same experience.

WebStorm is somewhat better, but I'm not going to pop for another license when I technically already have that functionality covered.

I haven't tried WebStorm since 2016, so my last experience is dated but I've just gotten to like VS Code for front end work.

Re: Introducing Atom-IDE

#319

Earlier quoted context omitted.

I will regret asking this, but why ? I don't even remember what it was like to code without autosave in IntelliJ. I will review the changeset before creating the PR anyway, so I don't see the downside of autosave.

Close editor by mistake, battery dies, go to lunch and forget what I was doing. What is the upside? One less keystroke..?

The upside of not potentially losing everything you have done since the last time you remembered to save ?

is this a real question ?

Post reply on HN