EcmaScript Edition 5, formerly known as 3.1: Getting a new JavaScript | Hacker News Reader

EcmaScript Edition 5, formerly known as 3.1: Getting a new JavaScript

ajaxian.com

Even if this is ever rolled out to modern browsers, IE6 is still (according to w3schools) at about 17% of the market.

We'll never lose IE6 entirely, due to the plethora of small/medium/large business apps that only work with it (and won't even work with IE7).

IE6 is trending downward at about half a percent per month. That will likely plateau, but if it does continue we should be IE6 free in three years.

The key, like always, will be to develop two versions of your app: One that's shiny and works in new browsers, reverting to traditional html and anchors for any other browser.

I'm actually seriously considering treating IE6 as if it does not have Javascript support at all, in the future.. and if users want the full experience, they should upgrade their 8 year old browser.

I don't know what this was downvoted. Sounds like a sensible plan to me.

I'm currently in a company that has IE6 as its browser installed by the IT department. Most of the intranet would be broken in IE7 and especially IE8. The current techniques for handling IE6 should be emphasized to any newcomer to web development.

Sad but (if?) true. Another IT department with the philosophy, “No one ever got fired for going with Microsoft.”

If Microsoft is serious about being nice to the internet, they need to release a plugin for IE7 and 8 that makes it act exactly like IE6 for selected sites, and "regular" IE for everything else. Enter some whitelisting of sites, to avoid using the plugin as an attack vector for bad things, and there should be a chance we could all move on.

Isn't that what compatibility mode is all about?

I see a lot of posts like yours on a lot of web development forums, especially on hacker news. "Gotta support ie6, it still has some market share" they always say. But who is running this legacy browser? What kind of person is it? Well it's probably either: A) the office worker stuck in a legacy-IT job, or B) someone who doesn't really know better, and just uses their computer to check their email, maybe.

If you consider this, for the majority of sites, we are already IE6 free. Yes, you could develop 2 versions of the same app, but wouldn't your time be better spent either improving your more functional version of your app, or maybe creating an entirely new app with your free time?

A lot of people only surf the web when at work. If you want to sell to those people then your website has to work on their work computer.

We've been collecting data on visitors for one of the sites I work on, and it's about 20% IE6 still. We haven't yet collated data for what our actual signed-in users typically use, though.

Of course my time would be better spent on not dealing with IE6 irregularities. 80% of our time is spent on making sure 20% of our users can use the site ;)

I'm sick of it.

I'm sick of moans about IE6. IE6 is old and sucks. We get it. Yes, around a fifth of web users use IE6. No, nothing can be done to force users and IT departments to move across. No, it's not Microsoft's fault. If you want to pull support, do it. Good luck to you. But having moan in the comments of tangentially related articles isn't going to help anybody.

It's got little to do with this (admittedly crappy) article. Standards take years to get implemented, if they ever do, but it doesn't mean that standards shouldn't be written for a time when they are supported, because that would mean no progress instead of slow progress, and that would be even worse.

100% of currently deployed browsers won't support ES5 when it's finally available. Why pick on IE6?

Because while 100% of currently deployed browsers won't support it when it's available, nearly 100% of the users of those browsers (sans IE6) will upgrade very soon after a new version is out. IE6 users (especially corp users) will probably not upgrade any time soon. So, while modern browsers might support it, a certain (largish) population will not. It's the NS4 problem. Except worse.

They dropped EcmaScript Edition 4 because of internal fights between Adobe and Microsoft - and I doubt they have resolved their issues as it's a clash of mega corporations. Right now, most of the proposed changes in EcmsScript Edition 5 are minor and there will go a long time before they are used by the mainstream developer. I think most future web development will be focused on HTML 5, CSS3 and improved performance of JavaScript - - which is a shame, since EcmaScript Edition 4 implemented correctly would have made a real revolution.

The first update to JavaScript in nearly a decade and rather being happy that the revision introduces some nice new features, fixes some of the biggest problems with the standard, and has been embraced by all parties so should have a relatively quick introduction, people would rather moan that it's not enough.

This really takes some of the best of Mozilla and PrototypeJS's work (or the ideas from them) and incorporates them into EcmaScript. Function.prototype.bind, JSON object, DOM prototypes, and Mozilla's array helpers. All super great stuff that has been hacked into the language via libraries/frameworks thus far.

Is there a changelog or something similar anywhere? Or better yet a summary, "Here's what developers need to know."

Sure! There is now: http://news.ycombinator.com/item?id=562045

Reply on news.ycombinator.com