Friday, January 2nd, 2009

“I won’t support IE 6 in 2009”

Category: Browsers, IE

At a New Years Eve party, a friend help up a drink and toasted to his company deciding to discontinue direct support of IE 6 in 2009, and letting users know that the site may work better with IE 7 or another latest browser.

Then, Asa Dotzer puts up a chart of the IE 6 numbers:

Still far too high a percentage and enough to make you grown. Also, the last few pounds are the hardest to lose. The good news for me is that on most Web applications that I personally work, and definitely those that I work on in early 2009 will have much different numbers.

Maybe IE 8, Windows 7, and the great new browser war will help, or maybe some percentage is for lost computer souls.

Posted by Dion Almaer at 10:00 am

4.5 rating from 80 votes


Comments feed TrackBack URI

That chart is a pipe dream for me… Our numbers for our SaaS application (Not generally accessed by the public) is showing still 50-60% IE6 penetration. So I’m forced to still deal with it.

Comment by dgavey — January 2, 2009

The web application I’m building (shipping ’09) is not planning on supporting IE6. It will be public but not free. We have a niche target audience that we just aren’t worried about re requiring a modern browser. I have faith that others will follow that path. Some will continue to support for a while – major public facing sites/low tech sites/web2.0 apps begging for users… will likely support IE6 until 2011, but I expect by the end of 2009, that % will drop below 10% and many new ajax heavy apps will not support it. Additionally, I think there will be a big movement to make sure web sites work in IE6, but don’t have to look as good.

Comment by genericallyloud — January 2, 2009

we’re still facing at least 30-50% due to our heavy corporate clientele. :(

Comment by ilazarte — January 2, 2009

Hey, it’s easy: if you do support it, IE will remain.

If you don’t, it won’t.

If, let’s say, would greet everyone tomorrow with a message saying “Web 3.0 isn’t supported by IE6, please upgrade or stay in the stone age”, IE6-percentage would effectively drop below zero in a day or two.

GMail is already doing this.

How can you help Google in this mission?

Do NOT support IE6.

If none of the major portals do support IE6 anymore – and it’s just a decision, even if it gives a low week or two in traffic – it will phase out IE6, since users will be left with no choice, for everyone’s benefit.

What’s the benefit for a user to upgrade from IE6 if “the internet DOES work with this too”? Nothing. What’s the benefit of dropping IE6 support for a developer? Speeding up development time twice as fast.

So, everyone, just please drop it, and insert a popup (ajax-style, no, saying that you’re outdated, upgrade to IE7, or download firefox/chrome/opera/safari/whatever.

And thanks for GMail team for taking this step at last.

Comment by Aadaam — January 2, 2009

We have a web app deployed to two verticals. Here’s our Dec 2008 numbers:

A) More small/mid
IE7 62%
IE6 14%
FF 22%

B) More enterprise
IE7 47%
IE6 34%
FF 18%

We also get Chrome, Safari, et al, but add them all together and it’s < 2%.

Comment by jadeonly — January 2, 2009

@Aadaam – The corporate world doesn’t move that fast. We have big customers who would like to upgrade, but can’t upgrade ten thousand users from Windows 2000 or rewrite proprietary web apps overnight. Now the recession makes it even harder.

Also, if you’re spending 50% of your development time on IE6 issues then “you’re doing it wrong”

Comment by jadeonly — January 2, 2009

“Still far too high a percentage and enough to make you grown”

That sentence made me groan.

Comment by zhephree — January 2, 2009

I’d just like to give a bit of thanks to teams like ExtJS, who take care of all of this compatibility stuff for me. Most of my users still use IE6 (and will continue using it, I’ve stopped wishing for anything else), but thanks to ExtJS I can code in Firefox, and just occasionally check to see what it looks like in IE as little as possible!

Comment by mdmadph — January 2, 2009

I’ve said this before, but I guess a good thing can’t be said too many times, people should at least warn people about IE like we’re doing at – (hint; visit with any IE browser ;)

Comment by ThomasHansen — January 2, 2009

@ThomasHansen –

The best way to get users to upgrade is NOT to leave them a snarky, condescending message. That’s a great way to get a user to keep using their browser out of spite to all those “elitist internet nerds”.

The way Google’s done it with Gmail is a FAR more appropriate method. They offer a direct benefit (“get faster gmail”), and helpful, non-condescending language.

Comment by jmking — January 2, 2009

>>If, let’s say, would greet everyone tomorrow with a message saying “Web 3.0 isn’t supported by IE6, please upgrade or stay in the stone age”, IE6-percentage would effectively drop below zero in a day or two.
What would drop would be Google’s visitors and ad revenues.

Comment by Nosredna — January 2, 2009

The bad news is the customer we’re working with right now uses IE6 as their standard browser across their thousands of employees, and won’t allow any other browsers to be used. The good news is that they say they’ll transition to IE8 when it comes out. But I’ll believe it when I see it.

Comment by MattCoz — January 2, 2009

I like google’s strategy here. Lots of people are saying they’re blocking IE6 users, but that’s not what they’re doing. They still allow IE6 users, but they inform those users that they’ll get a much better and faster experience on newer browsers. I think this is the best strategy, and everybody should do this. Customers who just aren’t informed get informed, and those who are stuck (for whatever reason) are motivated to move beyond their blockage. At the same time, nobody is actually locked out, so they’re not making their customers angry at them, which is always a good move for a business.

Comment by Joeri — January 2, 2009

You’re actually right :)))

Comment by ThomasHansen — January 2, 2009

@jadeonly said “Also, if you’re spending 50% of your development time on IE6 issues then ‘you’re doing it wrong'”.

If you’re not spending considerable time fighting with IE6, jadeonly, then you’re not doing very interesting work. Sure, if you use the off-the-shelf dojo, jQuery, etc. tools, then IE6 only adds maybe 10-15% more work. As soon as you create custom widgets or push the existing toolkits to the cutting edge, IE 6 adds at least 25% more work.

Comment by unscriptable — January 2, 2009

I visited that site in IE7 and got the message, but the page didn’t look ANY different than it did in Chrome. What’s the point of that? IE7 is actually a pretty decent browser, and will be supported by MS for years to come.

47% of people come up with percentages off the cuff and off the top of their heads. You can’t quantify (across the board) the additional time IE6 development takes except to say that it’s probably in direct proportion to the length of time your project takes.

Comment by commadelimited — January 2, 2009

Anything that doesn’t support W3C standards is not a decent browser.

It would be one thing if IE 6 AND 7 both required the same hacks to work – but both require independent amounts of time to tweak for. IE 6 isn’t as capable as IE 7. Also, just because a product is still supported does NOT mean it is good.

Comment by LJHarb — January 2, 2009

Maybe IE6 users don’t upgrade to IE7 because they simply can’t. Maybe their hardware isn’t up to par. I don’t know, but to simply bail on these users IMHO is wrong.

Comment by skypoet — January 2, 2009

@skypoet: The fact of the matter is that people are upgrading from, or leaving, IE 6 in favor of IE 7 or other browsers. Identifying what percentage of the remaining IE 6 marketshare is comprised of people who “can’t” upgrade is a fool’s errand.

Comment by eyelidlessness — January 2, 2009

@unscriptable: “If you’re not spending considerable time fighting with IE6, jadeonly, then you’re not doing very interesting work. Sure, if you use the off-the-shelf dojo, jQuery, etc. tools, then IE6 only adds maybe 10-15% more work.”

So we agree then that it’s no where near 50%. The original poster said not supporting IE6 would mean he could develop twice as fast. I’m not saying IE6 is insignificant, but I know very well how much time our development team spends on everything and IE6 is just one of several browser headaches and browser headaches are just one part of our development. I look forward to dropping IE6, but saying we’ll develop even 10% faster is far from the truth. I can’t speak for others but if you are spending 50% of your time doing IE6, you are doing something wrong.

We do lots of interesting AJAX and drag & drop and all the latest hotness. Our primary application has several thousand lines of proprietary JavaScript. We’ve replaced all 3rd party code with in-house controls except for mootools, excanvas, and soon Gears. That includes custom validation, combo boxes, date pickers, grid formatting + frozen headers, you name it. You know what I’ve come to hate more than IE6? Mobile browsers.

Comment by jadeonly — January 2, 2009

I’m just curious to see how people will make use of that sudden rush of freedom that will come in the wake of IE6’s absence. Will the common developer even be aware of the possibilities that are finally available or has he/she unlearned the parts of the specs that weren’t supported by IE6 (or just been ignorant of them all along)… Time will tell ;)

Comment by rasmusfl0e — January 2, 2009

For the web app we’re hosting, we’re getting the majority of our customers from Asia and India where the large majority (> 90%) of them use IE6. That’s because in those countries piracy is high and they all run pirated Windows and IE7 won’t install on a pirated copy of Windows.

Comment by Jordan1 — January 3, 2009

You have to understand, that the majority of the IE6 users are corporate, if not all. They are stuck with intranet apps based on IE6 that break in newer browsers. So the main reason we all face IE6 hell, is because back in the 90’s and early 2k we all made crap intranet software.
It’s cheaper to stick with IE6 then to upgrade or rebuild all those outdated intranet applications and that’s the jest of it.
Ironically, plugging newer and better applications and proving cost/time benefits and thus better profit margins to our clients, is the only and best way to get rid of IE6.
5 or 10 years from now, we’ll have the same problem with IE7 and perhaps IE8 unless standards are pushed now and maintained.
And as long as there’s no competition for Microsoft on the intranet market, Microsoft can do what she likes and browser standards will always have to be based on IE, since that’s what the companies are using, which are our direct clients.
Morale of the story, IE6 will be around and mandatory to support for a long while yet… =(

Comment by BenGerrissen — January 3, 2009

The majority of our user community is still on IE6, I’m talking 90+ percent. Thinking of one application in particular that I lead development on… it’s some of the most complex client-side coding I’ve ever seen… I’d be willing to put it up against any other application anyone cares to in terms of complexity and what it asks a client-side application like this to be capable of… I’d say 80% of our UI code IS NOT using a major library, it’s what most people would say is a home-grown framework (we do however have a mix of libraries in play in places including Dojo, DWR, dhtmlx, ActiveWidgets and I think one other that escapes me at the moment).

With all that in mind, we’ve had VERY LITTLE issue with IE6 at all… and no, we ARE NOT targeting IE specifically (our app runs perfectly in Firefox from about version 2 and up and is visually identical too)… all we’ve done is carefully coded to standards and constantly check in IE6, IE7 and FF that things are looking and working as they should and we adjust as necessary when they don’t (yes, I’ll admit, there’s some browser checking in some parts of the code, and yes, there’s some stylesheet hacks here and there, but not nearly as many as you’d expect based on what the application is and does).

I’ve never understood what all the fuss is about with IE6, based on my experience. In fact, performance-wise we see BETTER performance in IE6 in many parts of the application (which is something I’ve never quite understood!). And, all of this was achieved with a fairly junior-level team… aside from myself (who did a lot of the original framework coding, because no one else likely could have pulled it off, and then subsequently moved pretty much to a pure lead role) there were two others with decent knowledge and experience who didn’t really do much coding (one did almost none actually), the rest of the team was relatively new to Ajax and client-side development in general.

IMO, following standards is a great recipe for minimizing cross-browser concerns, whatever versions you have to support (well, within reason… don’t think it’s going to help support Netscape Gold 3.0!). Don’t misunderstand me, I’m not saying IE6 is a saint and that all the problems are imagined… what I *am* saying though is that the work I’ve done would be considered simplistic by no one, yet the issues I’ve seen have been few and far between. The only thing I can reasonably attribute that to is following standards and diligence in confirming things are good to go in various browsers frequently. It won’t eliminate issues, but it sure makes the bumps significantly smaller.

I don’t know, maybe we’ve just gotten lucky for 2+ years, but I seriously doubt it :)

Comment by fzammetti — January 3, 2009

@fzammetti – I think some of the IE6 problems will always be a pain. Like it’s CSS support, lack of proper PNG24 rendering, inferior debugging tools, and also all the hasLayout issues. As you run into more of these problems you know how to avoid them or quickly work around them, so yes, development in IE6 becomes easier. Especially if the interaction designers, and visual designers listen to technical advice. It’s not that IE6 is unprogrammable, it’s just that you have to jump through more hoops to get the same thing done.

Comment by RobRobRob — January 3, 2009

@RobRobRob: Yeah, I think there’s something to that. the part about knowing some of the tricks to avoid things in the first place… certainly it’s true that the framework I put in place early on saved the rest of the team from dealing with some sticky things, and often times when you’ve been doing this stuff long enough you kind of subconsciously account for things that others would have to work to overcome (if that sounds a bit arrogant I don’t mean it to at all)… the part about hasLayout made me think this specifically because I do remember a few times now when I specifically did things a certain way to avoid that problem, but it was a subconscious thing.

And you’re of course 100% right about the tooling… how spoiled has Firebug made all of us?? :) I find I need 3-4 IE tools to get a similar set of tooling, and it’s *still* not the same (and I don’t just mean the convenience of it being all in one place, I mean functionality-wise too).

Comment by fzammetti — January 3, 2009

I would love to stop supporting IE6, though to be honest I don’t think it will be for another 6-12 months yet, and IE7 still has its quirks!

As another point of comparison, here are some stats from our analytics tool for the past 30 days – we are a global financial services provider, I think around 100-125th in the most visited site lists:

IE6 – 22.4%
IE7 – 49.7%
Firefox 3 – 15%
Safari 3 – 5.7%
Chrome – 0.8%
Firefox 2 – 1.8%
IE8 – 1%

The rest is made up of things like AOL’s branded IE and earlier versions of Firefox.

Opera has around 1,500 hits in the weblogs, out of a total of 40 million visitors – not sure what that says!

Comment by ckorhonen — January 4, 2009

These numbers are ridiculous. What they don’t tell you is how many of the IE6 users actually care about the modern functionality you’re trying to force-feed them.
There are exceptions, of course, but in general: If they wanted modern functionality, they wouldn’t be using IE6! Microsoft’s automatic update to IE7 would have gotten to them by now.
For these people, ignore your user interface; just make sure your core service is available to them. At that point you’re probably giving the user what they want. The major RIA apps (e.g. google’s) have figured this out already.

Comment by trav1m — January 4, 2009

The point is to get people to update to something else then IE so that we in the future don’t have to spend so much time supporting a browser which intentionally breaks standards and thereby making it difficult for developers to fully harvest the richness of the possibilities the web has already given us…
Also supporting IE just eats too much of my (and other web developer’s) time, and in a perfect world we wouldn’t need to spend “extra time” supporting something which shouldn’t exist – at least not in its current form…

Comment by ThomasHansen — January 4, 2009

Being at work is no excuse for users with IE6. They can still use Firefox Portable in most companies. They are just bored to or have no idea how to do it.

Personally I just inform them about the issues that IE6- causes to them (primarily, as most people are selfish by nature), the internet and the developers via an intro page that is only displayed once per session in IE6- browsers (they can still procceed in the site, if they want to). I also suggest several other browsers (even Firefox Portable).
The sad fact is that even though they spend an average of 2-3 minutes reading that page, their percentage hasn’t really dropped. :(

Something needs to be done with the IE6 plague though… I don’t know what that should be, but we can’t keep on like that…

Comment by LeaVerou — January 4, 2009

My feeling is that IE6 will be around for a long, long time. As long as your customers ( or your advertisers! ) want you to support it, you probably will. It would be great if everyone switched, but that’s not going to be a reality. I really believe that at least for a couple of years, some people just can’t or won’t upgrade.

Comment by skypoet — January 4, 2009

@fzammetti – I can back you up on your views of IE6. I am also developing a very complex web application, its over 10,000 lines of javascript not including a javascript framework that i’ve written.

I have very few problems with IE6, and it’s performance isn’t so bad. It’s nowhere near chrome of course, but it is usable, and as you said, some things (DOM) actually seem smoother than ‘modern’ browsers.

I’m pretty tired of lazy web developers who expect all browsers to work exactly the same. Yeah, there are standards but all the browser makers have their own custom CSS attributes, all of them do it, not just microsoft. Once a developer can accept this fact, and stop being so annoyed by quirks of one or another browser, then things get a lot easier.

the 20% share that IE 6 has is made up of 90% corporate users who are locked-in to it for one reason or another. This isn’t just some 20% market share of losers, these are people in businesses (with jobs), and they should be valued.

Comment by leptons — January 5, 2009

In my web application, which targets a mass market of home users in a specifical geographical area, the IE 6 numbers already have dropped below 10 %.

Probably I can just skip my IE 6 CSS hacks during the first half of 2009. How great!

Comment by torerik — January 5, 2009

I’m not a developer but just a common internet user (don’t ask how I ended up on this page). I presonally dont like IE7. I downloaded it via windows update and then removed it. If Google was to stop supporting IE6 tommorrow, I would start using Yahoo. If I was looking to make a purchase online and the merchant I wanted to shop with did not support IE6, I would find another merchant. It may be a nightmare for developers but obviously there are those of us out there that like it. If you want to lock us out you can, we’ll find somewhere else to do business other than with YOUR customers. There – 2 cents worth from the common man :).

Comment by lonewolf26 — January 10, 2009

As long as the industry enables the use of IE6, there are going to be those that don’t look for better options. It takes a great deal of time for a developer to adjust for every single browser, and with all of the new entries into the browser war, we could spend half of a project’s budget in testing alone. For those that think we should still be contending with this, take a look at the latest browser list:

Do you want to spend your money on creating a mediocre product that is compatible with IE6 or move forward in the industry with powerful features?

Comment by snoe — January 15, 2009

First as this is my first post I’d like to thank and congratulate all the editors.

I’d like too annonce that thanks to a long lobbying work as a senior developper in our company (one of the major domain name registrar) we decide to not support IE6 anymore. This is unnoficial we didn’t say we don’t support it officialy but we not make test on IE6 anymore. Customer that claims for bugs on it will be invite to upgrade to an earlier IE version or change browser to a recent FF, safari, opera or chrome.

Great day

Comment by botgv — January 21, 2009

I’ll drink to that! Bastard IE6 it’s like a beast that just won’t die! Maybe this year…

Comment by null — January 28, 2009

Even now, in 2012, we’re seeing about 10% of our clients are still using IE6. Some people, like my father, simply ignore any popup window that says “update available” which proves that many users still fear change.

Comment by inyerface — April 3, 2012

Leave a comment

You must be logged in to post a comment.