Why I support IE6 (and you should too)

UPDATE 14th Jan 2010: While the IE6 usage stats mentioned in this article have dropped, my views about IE6 support remain the same.

Microsoft Internet Explorer 6 (IE6), a web browser released in August 2001, is currently touted as the bane of the Web due to inadequate functionality and poor security.

Many web developers now think support for IE6 should be dropped, citing its lack of features and security problems as good reasons for upgrading. In this article I’ll discuss why I think this is wrong, and why you should continue to support IE6.

Note: My argument focuses on public facing websites. The argument does not apply to sites built for a fixed, known platform within a self-contained intranet unconnected to the web.

Why do we want to be rid of IE6?

IE6 was a reasonable browser when it was released, but Microsoft failed to keep it up-to-date. It suffered from many security issues and as a platform for cutting edge web design it failed, having poor support for advanced CSS and missing features found in other more capable browsers, such as semi-transparent PNGs. Many hacks and work-arounds were found for these shortcomings, but some developers complained that learning and deploying these hacks was time consuming and expensive.

Your users are more important than you

Most problems were fixed in IE7 (Oct 2006) and IE8 (Mar 2009), but despite the upgrades offered by Microsoft, IE6 is still used by anything from 17% to 33% of web users. That’s a pretty big slice of the pie!

By current standards, IE6 is a poor browser. As a web designer I would love to ignore it. But what I wish for as a web designer is secondary to what’s right for my clients and my client’s users.

It’s not my place to tell people what browser they should use and even if I could encourage them to ditch IE6, many are unable to. The largest sector of IE6 users are those in organisations where a company wide browser upgrade is restricted by cost, complexity and lack of business benefit (perceived or real). What company is going to spend money on an upgrade that does not offer a return on investment?

Given that there’s still up to 33% of users running IE6, and will be for some time to come, how should we offer ongoing support for this browser?

What is Support?

To argue supporting an outdated browser, we must define what “support” means. Jeremy Keith discussed it best in his article The IE6 Equation, coming up with a 5 point scale from “No support at all” to “Pixel Perfect across all browsers.” The scale follows:

  1. Block IE6 users from your site
  2. Develop with web standards and don’t spend any development time testing in IE6
  3. Use the Dean Edwards IE7 script to bootstrap CSS support in IE6
  4. Write an IE6 stylesheet to address layout issues
  5. Make your site look exactly the same in IE6 as in any other browser

1. Block IE6 users from your site

I don’t think any responsible developer will argue this is reasonable. Blocking users is simply so wrong that if you think it’s an acceptable way of dealing with the issue, you probably shouldn’t be building web sites.

2. Develop with web standards and don’t spend any development time testing in IE6

This is a good compromise if you’re sure the majority of your audience is using an up-to-date browser. Use your server logs and other research to determine IE6 usage – if it’s below just a few percent then consider this option.

3. Use the Dean Edwards IE7 script to bootstrap CSS support in IE6

I’ve never been happy to take this route: using a large JavaScript library to merely affect the look of a site, at the expense of speed and responsiveness is not something I’ve ever been comfortable with. Your mileage might vary.

4. Write an IE6 stylesheet to address layout issues

Correct the major layout problems, but don’t worry too much that the site will look different in IE6. If it looks more or less right and remains useful, usable and accessible, then that’s good enough. The 24 ways website is a good example – it looks fairly awful in IE6 compared to the expected design, but it remains usable. Paul Boag has a demo on this idea of Graded Browser Support.

5. Make your site look exactly the same in IE6 as in any other browser

This level of support is for the real perfectionists and if you can achieve it – great! But don’t waste a lot of time trying to do so, unless your client is happy to pay a premium for that. Yes, I believe if you’re going to aim for this level of support, then you should explain to your client what it means and should be confident enough to charge more for it.

I’ve chosen option 4 as the level of support that I offer. I think it’s the most appropriate for the kind of sites I design and develop. You’ll need to decide for yourself which level you think is right for you, though if you’re erring towards #1, then read on!

Participation for All

The question remains: why offer support for an outdated browser? For me, it’s simple: I want the sites I build to be used by as many people as possible.

In 1998, Sir Tim Berners-Lee, inventor of the web, wrote a one page article, The World Wide web: A very short personal history, which includes this passage:

The dream behind the web is of a common information space in which we communicate by sharing information. Its universality is essential: the fact that a hypertext link can point to anything, be it personal, local or global, be it draft or highly polished.

Sir Tim believes in the web being an equal playing field for everyone. I believe it too. Everything I’ve ever designed and built for the web has been done with that vision in mind. All websites should therefore be as accessible to as wide an audience as possible. I can’t tolerate any barriers that stop users from engaging in the web.

I choose to support IE6 to a reasonable level since in all good conscience, I must support it. By ensuring that a user can use my sites in IE6 I make sure that the playing field for participation remains equal.

If you’re a developer who advocates out-right blocking of IE6 (or indeed any browser), or building sites that work only in modern browsers, then I think you need to seriously ask why you are developing on the web in the first place. You’re actively discouraging participation and that really goes against the spirit of what the web is all about.

Final words to web developers

Your role as a developer is to make usable sites that work for your users. You’re paid to solve the problem of multiple browser support. If you think it’s too time consuming, then you’re either doing it wrong or need to raise your prices.

As a web designer/developer myself, I feel the pain of supporting IE6 and I can understand why a large number of developers are trying to drop it. But doing so takes your users out of the equation since you are only trying to please yourself.

Web development isn’t about what’s best for you; it’s about what’s best for your users.

32 thoughts on “Why I support IE6 (and you should too)

  1. If you opt for level 4 how do you explain this to a client? I am often fixing things to the nearest pixel because they do not understand…

  2. Hi Zoe, thanks for the question. It can be a tricky one to answer!

    Clients who expect a pixel perfect design across all browsers have been educated to believe that’s what web design is about. Of course it isn’t: it’s about delivering a site that achieves results. The way the site looks is ultimately secondary to the the way it works.

    If you can convince your client of that, you’re half way there. The other half is to explain that not all browsers are equal. Andy Clarke has a useful article where he discusses this, the last section has some tips:

    Time to stop showing clients static design visuals

  3. Great article Matt. I must say I can’t wait until IE6 does call it a day – not only for the ease of my job but so that I can stop getting embroiled in these IE6 discussions ;)

    @zoe – I’ve found that being absolutely clear from day one that their website is not going to look the same in IE6 as it does in a modern browser is met with understanding. I explain it in my proposal and then I stipulate it in my contract.

  4. Pingback: Why You Should Support IE6 | Sharpe's OpinionEdit

  5. Pingback: Frisk design : Web Design & Consultancy in London » Why IE6update is wrongEdit

  6. Personally I never get into the discussion about IE6 with clients; many of them won’t know what I’m talking about, and if at all possible I try to not ‘talk above’ them and make them feel uncomfortable.

    My explanation is that unlike print design, web design will not always look the same. Everybody has different computers, monitors, browsers. Therefore a well designed website will always look ‘right’ no matter what – but it may look slightly different on various computers. Good web design means it will never look ‘broken’ and will always be useable and fully functional.

    Besides, it’s not that hard to account for IE6 anyway. A few extra pixels in your margins to account for the box model, a quick htc fix for transparent pngs if you’re using them. What’s the big deal?

  7. It’s important to remember that yours is likely not the only web-based application that your customers use. And there are still plenty of apps that are only -officially- supported to run on IE 6.

    We have a number of mission-critical, used-daily, web-based apps that only run on IE, either due to actual poorly-written, hard-coded design limitations or in some cases, simply because the vendors won’t certify those apps on any other browser.

    Sure, you can run those apps in another browser but if there’s a problem you need support on – regardless of whether the problem has anything to do with the browser from which it’s being used – you run the very real risk of being denied support from the vendor.

    So, noncompliance with vendor requirements can have very real consequences – the likes of which few businesses can afford (literally) to weather.

    It’s easy enough for end-users to be cavalier about browser (and OS) choice but it’s worth remembering that companies are usually painted into a corner by the applications that they need to do their business and the limitations imposed by those apps & the vendors behind them.

  8. Pingback: Frisk design : Web Design & Consultancy in London » A Web for AllEdit

  9. I think Microsoft’s support for IE6 ended around 10-Oct-2006 (see http://support.microsoft.com/gp/lifesupsps/#Internet_Explorer )? You’re not a mega corporation and can’t test your work incessantly for all browser. Do you define the standards you support in your contract (IE6 == Fail, I believe), and the browsers?

    Unless it hurts you, or more importantly your client, financially I’d love to stop even thinking about it. If you look at your web logs, Matt do you get IE6 visitors between 17% and 33% of the time? I very much doubt it.

    You got the main jist of the article right though. It’s the *users* that count.

  10. Hey Ian, thanks for stopping by and commenting.

    I believe in an open web and that people should have access regardless of the technology they are using. It doesn’t matter to me that Microsoft have stopped supporting IE6, the fact that a large portion of web users still have it encourages me to support them. See my other post “A Web For All.”

    It doesn’t matter whether the stats for my own site don’t have any IE6 users. I build sites for people, not browsers. :-)

  11. I completely agree with with Matt.

    My website is also more or less looks same on all browsers…I did not have to worry about IE6 bugs…

    I have linked to this post on blogs which have hate message on IE6…

    Great post and thoughts…

  12. My employer still only deploys IE6 on all workstations mostly because we have applications that STILL don’t work on IE7. It’s crazy. I’m actually forced to use IE6 to post this comment on your homepage, thankfully I could read the post:-)

  13. Great post. My strategy is to make things work on all the “good” browsers, and then fix IE7 afterwards. If the website is mostly text, IE6 should at least see than then.

  14. There’s an important segment you’re forgetting with the developers “paid to solve the problem of multiple browser support.” comment.

    What about the people who pay those developers? I’m sick and tired of paying a premium just because IE6 is still around. It’s not a moral question, it’s a business question. IE6 eats investment resources like nothing else and is a pointless drain on any web-based app. Let it die.

  15. At our company, security is a top priority, and as such our service does not support IE6. However, we understand that some people at work have no other option other than IE6. As a response, we created a document for them to send to their IT guys which outlines reasons why they should offer an alternative to IE6.


    Your Safe and Secure Online Password Manager

  16. Pardon me for being so bold, Arsen, but that’s kind of a contradiction, isn’t it?

    Your say that your company understands that some users have no other option other than IE6, yet your service is not supported on that browser. And you’ve created a document which outlines reasons why they should offer an alternative to IE6. That doesn’t really seem very understanding at all.

    In fact, you’re kind of putting customers in an impossible situation. I’m not unsympathetic to the costs & issues involved in supporting legacy applications with their inherent glitches & security holes, but it seems as though your customers are being called upon to serve your needs rather than you serving theirs.

    Your numbers seem very flawed – Scridb’s document states that only 10% of Internet users are using IE6 where others have indicated that this number is more like 20-30% – and I suspect that number is low due to hobbyists & home users.

    And your ‘helpful’ document negates the single most compelling reason most corporate environments who still use IE 6 are doing so – because they’re restricted to that browser by vendors whose applications they’re reliant upon. If your company has mission-critical apps that are only supported on IE6, then that’s the browser you’re going to use. That’s not an IT control issue. It’s not a problem of insufficient hardware resources. It’s not a matter of using an illegal version of Windows (which I really doubt many companies are bold enough to do given the severe legal penalties that could result from that).

  17. I think the gist wasn’t that you have to stop using IE 6, but rather to use another browser for general web browsing in addition to IE6.

    Which is probably a good approach given lack of security patches for IE6, it is probably safer not to be browsing the web with it even if you need it for a legacy internal application.

    As for statiastics, I’m sure is varies significantly by market. I’ve seen some widely varying reports. Who knows what they real absolute numbers are.

  18. “It’s important to remember that yours is likely not the only web-based application that your customers use. And there are still plenty of apps that are only -officially- supported to run on IE 6.”

    @Rob O… so Arsen’s company is supposed to bear the burden of other vendors’ lack of development initiative? Wouldn’t the egalitarian weboshpere that you seem to be championing be much better served by those who refuse to change also bearing a portion of the ‘costs & issues’ involved in upgrading their legacy systems to meet some modern standard of compliance? It also seems pretty short sighted of a new company to waste development cycles to support a product that isn’t even supported by its own vendor’s other products (read Sharepoint, Microsoft’s gateway to the future). I also think that you have missed the point of the ‘helpful’ document that Arsen linked to. That article is advocating giving users an additional browser that is modern and web-standards compliant for general web browsing and using IE6 only in the circumstances where the system that you are trying to access requires it. This seems like a good compromise and is the route that many companies (including mine) are taking to deal with IE6 compatibility issues.

    I think that saying “I’ll support IE7/8, Firefox, Chrome, Opera, Safari, etc… but not IE6″ is the more progressive stance. Admonishing webdevelopers that decide not to support IE6 is like scolding them for also not supporting Netscape 4.x or Opera 3, and a ridiculous position to take.

  19. My tone on that response may have seemed a bit harsh and for that, I apologize.

    But I come to this topic as a veteran IT guy who’s spend 20+ in the trenches. I’m no fan of IE6 – heck, I use Firefox at home myself. However, I’m realistic enough to know that in the business world, it’s rarely (almost never) as simple as saying, “I think I’ll use browser X today.”

    We have a half-dozen or more of daily-use, web-based apps in the healthcare setting that I work in that all ‘require’ IE6 – and nothing higher than Java 1.5.1. None of us like that situation, but we can’t dare risk losing vendor support on
    mission-critical patient care applications for the sake of a snazzier browser. It’d be both ethically and fiscally irresponsible for us to disregard the vendors’ regs. Based on conversations with colleagues, I’m confident that our situation is much more the norm than exception.

    I do applaud Scridb’s efforts to education their users and encourage them to migrate to a more secure browser. But I still maintain that their practices seem to demand that the users meet the company’s needs – and that just seems like bad customer service.

  20. So how do your respond to youtube.com disallowing IE 6 users from using it? Do you think this was a bad decision on their part?

  21. @bydesign: Youtube based their decision on analysis of their visitors. They determined that the userbase for IE6 was low enough to now ignore it.

    I understand that thinking but I don’t necessarily agree with it. The site still works in IE6, so why can’t they continue to provide support?

    That said, if big names like YouTube stop supporting IE6, it should certainly help raise awareness of the issue and possibly speed up the process of getting people to upgrade. That still doesn’t help the people who can’t upgrade, however…

  22. After some careful thought, I’ve decided to go with IE6update. As the Brazilians would say, I think it’s kind of “chique” to see that yellow bar at the top of the page, prodding progress in the butt.

    One of your complaints was that it was dishonest… Let me ask you this: What’s on your computer screen? Is it really my site when you’re looking at it through IE6? I don’t think so… IE6 is rendering nothing more than a distortion of what I really intended.

    “Internet Explorer lacks updates to view this site…” Is it a lie, or is it more truth than what most people can handle? I think it’s poetic. I think it’s a message of hope… that’s there’s more to this world than what the corporate giants want to feed us. “If you really want to see this site… If you really want to see the world as it was intended… Just say no to IE6!”

    Poetry, brother!

    You support people who want to see the world in such a distorted manner and those who traffic the tools that empower them. If it was up to you, the Boston Tea Party never would have happened, because you’re the type of person that is happy to let the world run right over you. But even worse than a doormat is an enabler, because not only do you accept, you also support. You are enabling not just the users, but the trafficers as well.

    Think about it. Get off of your silly high-horse and stand up for what is right.

  23. I agree with you, Matt. I think it is a bad business move to completely block users with IE6 as this is just as bad as the sites built FOR IE6 (or earlier) back in the day that block any other browser. It’s frustrating, and some people just don’t have a choice. Whether it’s because they just don’t understand their computer, or they don’t have control over the computer that they use (i.e. a work computer).

    A lot of designers don’t take into consideration the sheer cost and undertaking it would be for a company to upgrade. In most cases it would be somewhat equivalent to switching from Windows-based computers to Mac-based computers. In a large company, this can mean millions of dollars – or more.

    Most companies do not allow their employees to install programs and make changes. This is, of course, because employees cannot be trusted to be smart about what they install, and also can’t be trusted to not install things they shouldn’t be installing and playing around with at work. They also don’t want to allow employees to install programs because most employees at these companies only know what they are trained to know about these computers – which normally is just using one program. This means the company needs to pay their IT department, or pay an IT company to come in, just to upgrade the browser.

    However, the cost doesn’t stop there. The reason that a lot of the companies haven’t upgraded is because something in their infrastructure relies on IE6. There is a program or system or site that they use every day all day, that essentially is part of the backbone of their company, that only supports IE6. This means they would have to pay to switch to a new system – which includes licenses and installation, etc.

    Then comes the training. They need to pay someone to come in and train each and every one of their employees on the new programs. Not only that, but they are also out all of the time that it takes to train each of those employees.

    It can get pretty expensive very fast and waste a lot of production time for many companies. For most company heads it just isn’t worth it – their concern is keeping the company running, not ensuring their employees have a good web browsing experience. And as Rob stated, in a lot of cases it is a matter of the vendor support and regulations and ignoring the risks would be ethically irresponsible.

    That being said, I don’t think that the differences in IE6 in most cases (if you script properly) are bad enough to cause that many usability problems. As long as you don’t completely block the users off, they can deal with the site looking slightly different. Charge extra to clients who want it to look exactly the same in IE6, because of course that is extra time and work for you. I think at most a splash page warning them that the site might not look the same in IE6, but still allowing the user to access the site, is the most extreme most should go. However, if you take the time to set that up, you may want to just take the time to fix any problems on the site.

  24. Hey Sarah, thanks for stopping by and sharing your comments.

    I think you’ve very clearly shown why the IE6 issue is far more complex than it might first appear. Costs — both obvious and hidden — are a huge factor in corporate upgrades and that isn’t going to change any time soon, especially during recession.

  25. Hi, Matt!

    Is the IE6 issue so complex because you’re having a hard time wrapping your mind around the idea that upgrades cost money? Or is it because the mutual praise is causing warm fuzzy feelings? Upgrading is just another one of those costs that are necessary for staying in business.

    You seem to be saying that “Business A” can’t afford to upgrade, but as developers we need to accomodate some piece of crap software, because if we don’t “Business A” may go under. If it can’t upgrade, then it can no longer compete in whatever market it may be in. Your claim has nothing to do with the merits of “Business A.” It may be some scandalous, lowlife sort of business, but as developers we have a moral obligation to do whatever is necessary to keep it afloat.

    From where does this moral obligation arise? It sounds kind of communist to me. Are you trying to put Obama out of a job?

    I have a better idea. As developers let’s move on and support progress. If they can’t keep up, they will either fail as they should, or Ben Bernanke will right them a check.


  26. Hi Matt,
    I came across your website via freelance switch and found this post exceptionally interesting. In my opinion web developers who do not support IE6 are just lazy. I just do not see why as web developers we would create a website that would not work well with IE6 thats like me walking into Tesco and not being able to buy food. Although IE6 use is becoming less and less I still think as a web developers you have an obligation not to restrict business for another company, at the end of the day they are paying you to potentially increase revenue for their company.

    I however, do think that IE6 sucks.

  27. Then what happen if IE6 is no more? Until no one support IE6 this battle will never end.

    All of us must move forward. I dont support IE6 and now im lazy? For as long as the client pay more for fix for every problem they encounter, then im happy to do so.

    You see, if you support ie6 what will happen? 17% will be happy. and if NOT? 17% will just stop using the internet? No, they will not if they are really interested. And they should buy a new products and spend more money and then the whole world economy will be happy. How’s that?

    You cant please everyone. So no! No more support to ie6.