IE6: To support or not to support.

Question

As most Web developers know, IE6 is a pain to support when it comes to making a website look and feel just as in the other major browsers (Firefox and IE7). I'd like to know what are the feelings of other developers toward supporting IE6 on their websites.

Of course if your main user base uses IE6 or if you're working for a client that requires you to make it work in IE6, you don't even ask, but what about a web blog for example?

---

An interesting idea to help converting users: http://www.savethedevelopers.org/

Articles about the subject:

• Why I Am No Longer Supporting IE6
• Is it Time to Ditch IE6?

Answer 1

As nice as it sounds to not have to support IE6 anymore, the fact of the matter is a large percent of internet users still use IE6. Many of these people are business users in offices where they don't have any control over what software is installed on their PC, so no amount of convincing or reasoning with them is going to change that.

Bottom line: unless you want to risk alienating a large percent of your potential user base, you have to still support IE6.

Even with something like a personal blog, unless you know that your audience is going to be highly technical, and thus much more likely to use a browser other than IE6, it's still a smart decision to support it.

Answer 2

Take a look at http://www.w3schools.com/browsers/browsers_stats.asp, IE 6 is still ~25% of visitors to w3schools which almost certainly has a more tech savvy set of users than usual.

Answer 3

some companies or gov agencies still use ie6 as their standard browser because a lot of legacy web apps were written for ie6. they can't afford the money to upgrade those apps, so they're stuck.

depends on the nature of the site you're building.. for public sites, i say to make sure the site isn't unusable in ie6.after that, i wouldn't worry about the fancier stuff other browsers support that breaks in ie6.

funny timing, i just blogged about ie6 not too long ago. http://www.8164.org/dear-ie6/

Answer 4

I think we're stuck with IE for some time. Looking over my stats, about 80% of our traffic comes from IE, only about 20% of that is IE6, but our stats are heavily biased by our intranet running IE7.

That being said I still test in IE6. I won't jump through hoops if something looks a different than it should but I do make sure that all the core functions of the site work.

Answer 5

Good question, and something that concerns me the more I get in to web development..

I would personally probably mirror other comments here, I wouldnt jump through major hoops, but I would ensure that my app still functions.. If it looks a little crappy because of IE6, just display a small message to let them know they are using a crappy browser and consider upgrading to a newer browser.

If your building applications/sites for a client though, that may well be a different matter. If I paid for a site, I would expect for it to work on all major browsers as expected (unless agreed otherwise).

Answer 6

I would suggest monitoring the W3Counter Global Web Stats which are published on a monthly basis.

As I write this, it shows that Internet Explorer 7 has the biggest market share, with IE6 coming second and Firefox 2.0 coming third.

This clearly shows how a particular version is being adopted by the masses and will give you a very good idea when to not support a particular version of a browser.

EDIT:

Stats at September 29, 2010

1    Internet Explorer 8    26.66%
2    Firefox 3.6            23.98%
3    Internet Explorer 7    10.29%
4    Chrome 5               9.93%
5    Internet Explorer 6    6.27%

Answer 7

You might want to use IE7-js. IE7 is a JavaScript library to make Microsoft Internet Explorer behave like a standards-compliant browser. It fixes many HTML and CSS issues and makes transparent PNG work correctly under IE5 and IE6.

Answer 8

As much as it pains me, I still support MSIE6 on all of my websites. One day we'll all be free of this monstrosity, but for now it is still far too common to ignore.

Answer 9

One of my websites (targeted to gamers) has 48% Firefox, 45% Internet Explorer and 7% "others". One third of the IE users still use IE6, which means a total of 15% IE6 Users.

Ergo: I'd still make sure that IE6 users can use the page without any hassle, but I'd probably not put too much effort in it if "tiny" problems occur (i.e. picture one pixel too far to the left).

If you need a way to check if the page works with various Versions of IE, you should give IETester a try.

Answer 10

I don't think you can ignore IE6 at this point. I would suggest, however, that any fixes you use in CSS etc. that specifically target IE6 (the * html filter for instance) be contained in a separate stylesheet. This can then be included in the page using the IE conditional statement in your HTML.

Answer 11

So long as it functionally works in IE6. I wouldn't worry about styles/layout being broken so much as whether or not they are able to functionally use the website. Once it functionally works, then worry about the design looking correctly in your free time.

Answer 12

At my day job IE7 is our standard, although I use FF for development...

Basically: Code using FF, tweak for IE7, then spend a few hours getting it somewhere close to the same in IE6...

We've just bought in some back office software from a 3rd party company and were told they would support either IE6 or IE7. While this might work for internal apps I reckon we should still be supporting IE6 on public sites for a while yet.

Answer 13

I agree that IE6 should never be totally crossed out, but depending on the conditions, I think it might be ok not to worry about it.

• Globally, any public web site targeted to a standard public audience should work with IE6, no debate here.

• If a client still uses IE6 (internally or externally), then there is no debate either, you should support IE6.

• For an internal application, IE6 can be ruled out depending on the client's demands.

• For a public web site targeted at a technical audience, I think I might not care about IE6 even if it means losing a few users/readers. I'd do the minimum, but I would not lose valuable time and valuable functionality that does not work properly in an outdated browser.

Answer 14

For now you're better off supporting it. Use a javascript library like jquery to minimize the impact of the browser on your code.

Answer 15

It depends on your target demographic, but generaly it's a bad idea to drop support for IE6 at this time, since as pointed out above, IE6 is still the biggest browser out there.

Answer 16

Here's an interesting article about the subject.

Why I Am No Longer Supporting IE6

Answer 17

I hate IE6 as much as anybody, but the real problem is that the people that are on IE6 are the same people that do not understand that you're site doesn't work because they're the type of people that stay on IE6.

So, you're likely to be punishing your least tech savvy visitors, and they have a hard enough time as is.

Answer 18

Thanks for the link to my post :) I do want to make sure one point is very clear:

It's not about making your site inaccessible to IE6 users at all. It's not about even making the experience for IE6 users even sub-optimal (well, maybe a little).

It's all about allowing the site to be decent for IE6, just not pixel perfect.

At some point we'll need to stop supporting IE6. I mean, we don't support IE4, Netscape2, etc anymore do we? For me, I can make this decision because I have a very clear target audience for my sites. The same would be said for StackOverflow.com. I think it would be perfectly acceptable for StackOverflow.com to not fully support IE6. Not saying it wouldn't be usable for IE6 users, they would maybe just see some PNGs with the gray backgrounds where there should be transparancy, or something that is spaced weird or something. For my sites I can make this decision because they are targeted to developers, IT professionals, and those looking for someone that deals in technology. I don't think the web development community should be required to support users who don't upgrade because 1) They are using an illegal copy of windows 2) They cannot upgrade due to being on Win98 or something 3) They just don't know how to upgrade. People who are forced to use IE6 due to IT restrictions, that is different. They should still be able to use our sites - but it just might not look as good as if they were on a newer browser.

BTW, if you're going to allow the site to look less than perfect for IE6 users, I am all for letting the user know. The SaveTheDevelopers.org site mentioned is great, but someone mentioned a better one in the comments of that post which I like better (I think it is less confusing). It is http://www.pushuptheweb.com/. The script you can add to your page on this site makes things a bit more clear (although to be honest could also be mistaken for a malicious link as well "you have an important upgrade available - click here")

-Ryan

Answer 19

The answer is, it varies depending on your situation.

Check your logs, analytics etc, see what percentage comes from what user agent. From that you should be able to work out when it's worth supporting a browser or not.

Update - I wrote a blog post illustrating how I use analytics to shape our browser support policy

For example, My employer has several websites, our own promotional site and then some ecommerce sites.

If you check google analytics, our promotional site receives a clear majority of visitors from Firefox, the majority of which are Firefox 3. The rest is mainly IE7.

On our busiest ecommerce site, we get the most visitors from IE7, then IE6, then FF. Interestingly, on the ecommerce site we get more visits from IE6 than all versions of FF combined.

So there will never be a day when we can all turn around together and say let's not support IE6. You will have to work out when is best for you.

Answer 20

If a client insists on IE6-support you should emphasize that it will come at an extra development cost if the site in question needs to work on modern browsers.

I usually code using standards (FF), tweak for IE7, do just enough for IE6 to be bearable.

I would even go so far as to have a "Please update your browser"-banner" at the top of all pages of a site when people use IE6 to view it.

Answer 21

As others have pointed out, it isn't a binary question of whether or not you support it, just to what extent. If you're considering going back to the Dark Ages of a "browser gate" where you disallow IE6 users, I would advise against that.

Salesforce.com, which caters exclusively to the business community, has chosen to drop IE6 support for new features.

Salesforce and Internet Explorer 6

Depending on the complexity of your site/application, IE6 support may not be all that difficult. Our HTML/CSS needs are basic, and as such we have a very small, conditionally-included IE6 stylesheet, included last, that tweaks the settings defined for FF3/IE7.

So before making your decision I would consider several things:

1. What/who is my intended audience? A site like w3schools or stackoverflow stands a much better chance of having users who are on the latest browsers.
2. What level of support is adequate? What (if any) features will IE6 users be missing out on? Can we gracefully degrade?
3. How difficult will it be to support? A few CSS tweaks or an entirely separate set of templates? If you need 50 lines to have terrific IE6 support, why not?

It's all about balance :)

What I would advise against is making a soapbox stance, and I'm little concerned about the word "feeling" in your question. It's simply a business tradeoff, as the SF.com blogger pointed out - either you spend time creating more new features in FF3/IE7 more quickly, or you spend time getting those features working in IE6 (thus slowing down your delivery of new features).

I would also suggest choosing your statistics carefully. While the original response pointed to w3schools, it should be noted the audience there is highly technical and I'm surprised any of them are on IE6 at all.

Answer 22

Well, since beginning of this year, I decided to drop support of IE6 in my projects. I mean, I try to make it WORK in IE6, but if there are small bugs in rendering -- I don't fix them anymore for IE6.

Answer 23

HEre are some arguments against supporting it; http://ie6.forteller.net/index.php?title=Main_Page

And yes, I am in favor of not supporting IE6...

Answer 24

Businesses who stay with IE6 do so because there is no business case to upgrade. Why would they? To get make non-business related websites work better? Yeah, right.

I also believe that the end result is that the present levels of IE6 usage are very near to what they'll ALWAYS be, if there continues to be no business case to change. That's why we MUST stop supporting IE6, all of us. Stop listening to your IE6 stats, because those users aren't listening to you.

Unfortunately, the most important software sector for this fight is also the least likely to change: business applications. People who sell shrink-wrapped software to the kinds of businesses with entrenched IE6 apps should NOT support IE6. That takes serious balls, but it's a huge step to getting out of this mess. I applaud 37signals for taking that step. It shows long-term vision for reducing coding costs for themselves, as well as for freeing the web for everyone else on the planet.

Answer 25

Browsers and Screen Size on Stack Overflow:

Here’s a quick look at our Analytics data from June 1st – August 19th.

Firefox             50.53%
Internet Explorer     29.12%
Chrome               9.04%
Safari               7.33%
Opera                 2.44%
Mozilla             1.15%
Konqueror             0.10%

Note that within Internet Explorer, the breakdown is 45% IE7, 29% IE8, and — this is depressing — 24% IE6. Which means that about 7% of our overall audience is still on creaky, broken, ancient old IE6. The Firefox breakdown is primarily 3.0 and 3.5, with a smattering of older versions.

Answer 26

Consider adding an unobtrusive javascript-based notification bar to your websites. That's provided by this website: Browser-Update.org
As they say:

"This service is an opportunity to inform your visitors unobtrusively to switch to a newer browser. "

Surely, that has not to be necessarily done with browser-update.org, it may be a similar service.

Answer 27

Apparently, IE6's funeral has been scheduled at www.ie6funeral.com, so the answer would be: 'No' :)

Answer 28

I see a lot of people here quoting statistics and target groups etcetera. In the end when developing applications for clients, it is simply about business, and statistics really don't matter that much (apart from helping to give advice to clients).

In my company in offers we explicitly mention that we do not support IE6. If clients specifically request IE6 support, it will cost us extra, so we charge them extra. To us it really doesn't matter, because the extra development time gets paid anyway.