top | item 7462348

You appear to be advocating a new IDE. Here's why it will not succeed

93 points| munificent | 12 years ago |gist.github.com | reply

66 comments

order
[+] habosa|12 years ago|reply
Yeah! People should never try to build new things that fit their needs/philosophy. I'll stick with Eclipse 1.0 thankyouverymuch.

Or not. Discouraging people who build or discuss interesting things without actually offering a better solution is petty and a waste of everyone's time. if someone thinks that building a new IDE where you type upside-down that only supports Arc, good for them. One of those people might come up with Visual Studio or intelliJ, and then save tons of people countless hours.

[+] aaron695|12 years ago|reply
> Yeah! People should never try to build new things that fit their needs/philosophy.

Correct. Certainly in the software world.

Software for young white males is done to death. And although this article is pure humour it makes an important point.

As if the IDE is going to be fixable on a whim. It's used and thought about by every person who writes software most days, as if there are any magic bullets left to implement.

And if you get put off by a humours article which you don't get the points on, you're doomed to failure anyway.

Go off and try something different. If people can't think of reasons why you shouldn't do it, then there's a chance it's a good idea that might have some lowing hanging fruit on it.

[+] ZenoArrow|12 years ago|reply
I understand the frustration of the author of this list, but do still believe there's room for improvement in the field of IDEs. Light Table is the most promising new IDE that I know of, do think it has a decent chance of bringing benefits to coding, partly because it seems to focus on the one important area that I still feel lacks any real polish in many current IDEs, which is in debugging.
[+] gkoberger|12 years ago|reply
I got the impression that the author wasn't frustrated, but rather mocking the common responses new IDEs like Light Table or Atom inevitably get.

EDIT: To reiterate what leoc posted below, the format this is presented in is a well-used "meme" in the tech world.

[+] munificent|12 years ago|reply
The great thing about this list (and the original programming language one that I stole from, of course) is that it mocks itself.

It is true that almost all IDEs are doomed to fail and the reasons why are easily listed. But it is also true that the critics of those are giving so little insight that all of their criticisms often boil down to a few things you could put on a checklist.

[+] gojomo|12 years ago|reply
You think you're being clever with your checklist dismissiveness. But your smug cynicism actually makes you one of history's greatest monsters, because...

...

[+] _prometheus|12 years ago|reply
Versions of this have always applied to creating editors -- or for that matter, any new {language, system, product}. And yet, new ones are created all the time, some even become widely successful. Your finality is drastically overstated, and fundamentally negative in value.
[+] lotsofmangos|12 years ago|reply
You know, I get the feeling that you are just not taking this seriously enough.
[+] pearjuice|12 years ago|reply
Atom could have been Github his expansion. Its final monopoly settlement. A web based IDE so modernly written it would only work on Chromium nightly (for few months). A business model in which your IDE is in the cloud backed by Github. One private repository for free, public unlimited and after that maybe $0.99 per private repository/IDE-project or something. Synced all across the cloud, responsive, adaptive, buzzwords alike.

You log in to atom.io and boom there is your IDE. Partner with Google for Chrome OS. But no, instead we get some poorly optimized, hacky IDE with half of a Sublime Text look-and-feel. In Javascript. On the desktop.

This could have been it. They had all the ingredients. Too bad. Maybe in the future.

[+] shadowmint|12 years ago|reply
nope.

See, web ides have two fundamental fails that no one has figured out yet.

1) your browser context is not isolated; you have other tabs that kill performance, you have browsers you need to restart periodically (dns cache), or die because of dev javascripty things (eg webgl).

Restarting your ide all day? yeah, no.

2) no access to local files without a stupid plugin.

enough said.

There are some not terrible run-your-own-local web ides, but how is that any better than a fat client?

Its not a wasted chance; its that what you want currently cant exist in a form that isnt worse than the current solution.

[+] jmhamel|12 years ago|reply
Hi! Perhaps the people of HN can offer some advice:

I’ve happily been using Emacs for about a year now, with loads of customizations from where I worked. Now that I no longer work there, the customizations don’t really make sense for my own projects. I started from scratch today, and now I totally hate emacs! I want to just start coding, and not spend a whole day setting up my IDE. Do you have any suggestions for other IDEs/reasons for me stick it out with Emacs? (I’m on a mac, doing mostly rails and node development btw)

[+] dntrkv|12 years ago|reply
Intellij IDEA. Hands down my favorite piece of dev software I use.
[+] Nursie|12 years ago|reply
Become agnostic/independent.

I use nano, gedit, eclipse, slick, sublime, whatever's on the machine I use at the place I work and available in the situation.

I don't think a good IDE holds you back, but if you can learn to live without one then you're probably ahead of the game.

That said, I'm just an old C hacker, so what would I know...

[+] dredmorbius|12 years ago|reply
Vim user, but in general: the advantage of text-configurable editors is that you can carry your configs around with you.

If you find you've got customizations you don't need, organizing them so that they relate to specific functionality which you can activate / deactivate as needed is a Good Thing.

The advantage of vi/vim or emacs over other alternatives is that you very nearly always have them available. A significant downside of any given IDE is when it's not there for you.

[+] collyw|12 years ago|reply
I usually use Aptana (Eclipse based). It still takes me the best part of a day to set everything up the way I like it so I don't know if you will gain anything.
[+] malkia|12 years ago|reply
Add to this list - good window docking system, and by good I mean Visual Studio, because what's built-in Qt is not enough, and while QT Creator is great, and it's very well suited for laptop use, I can't use all my monitors at work as much as I want to.
[+] codeka|12 years ago|reply
Is this in response to anything in particular?
[+] dredmorbius|12 years ago|reply
You appear to be asking a rhetorical question on HN. Here is why it will not succeed ....
[+] wmf|12 years ago|reply
Probably Atom, Nitrous, Light Table, and all the other Web IDEs being created in recent years.
[+] cheepin|12 years ago|reply
[ ] No one really believes that your IDE is faster than: [ ] Eclipse [ ] You have reinvented Eclipse but worse

I doubt it

[+] romanovcode|12 years ago|reply
I don't think there is an IDE slower than Eclipse.

Even NetBeans is faster.

[+] shadowmint|12 years ago|reply
amusing.

I particularly liked 'no one believes your new IDE is any faster than any of the other successful new IDEs other people have made'.