Featured Article

How to kill IE6 and influence people

How to kill IE6 and influence people

A few weeks ago, I thought that I had finally found the perfect design. I spent hours tweaking the XHTML and CSS to get everything to validate as well as look well in Safari, Firefox, and Opera. It was at that point when I moved the page into IE6 to see how it rendered. I expected a few quirks; I was not expecting my entirely valid page to not render at all. I was so mad that instead of accommodating IE, I scrapped and started over.

When I developed Gridlock, the hyalineskies 6 template, I tracked every change in both Firefox and IE6 to make sure the page rendered as intended. When I was done, IE6 rendered everything okay (considering its lack of support for the dotted border-style,) and the page looked great in standards-compliant browsers. I sent it off to a few friends. They all approved. I turned my layout into a WordPress theme later that evening, satisfied.

The next morning, I sent my new work off to a few more people. All said that everything looked fine except for one who replied “I don’t think this works in Internet Explorer.” I was confused; I designed the page step-by-step using IE. I opened IE and it rendered properly. I sent him a screenshot of IE playing nicely, and he replied with a screenshot of his own: nothing but the copyright information and the header graphic was displayed. The page was blank. Sickeningly, he was using the same version of IE that I was. I designed the page with IE in mind and it didn’t work copy-to-copy? I’m done.

Now, resting in JavaScript at the top of every page on this site, I have a small, IE-like warning message telling IE6 users that their browser might not render this site properly. (I’ve also included a link to Firefox.) IE5.5, older Opera, and IE Mac browsers don’t even get that far – JavaScript sends them off to a purgatory page written in old-school HTML 4.01 Transitional. I’m not the only one doing this, either; leading designers and developers such as Michael Heilemann and Khoi Vinh are doing the exact same things with their sites (just visit Binary Bonsai in IE6: what a mess.)

I’m well aware that a key part of the job of a designer is to build accessible pages. I probably wouldn’t be such a browser Nazi on a contracted project, but there comes a point, however, when we need to demonstrate to the obsolete our power as designers as well as tell the unaware obsolete that they are. I don’t have a problem shipping IE users to the obsolescence factory on this site; the majority of its readers use “real” browsers that conform to web standards anyway.

It seems contrary to our democratic society for me to take this dictatorial edge and tell the majority of the web – all of the IE6 userbase – that they are wrong; after all, to reach the maximum amount of people, we should be designing pages for IE6 and then cross-checking them in compliant browsers like Safari, not vice-versa. This is what Microsoft would want us to do. This is what we would theoretically do in most day-to-day circumstances, but we know better than the average user. We all need to grow up, act like the parents of the web, and tell the users that IE can’t render CSS well; we would be doing just that if the users were running Netscape Navigator. Unfortunately, for PC novices, their computers come only with Internet Explorer. The more people that harass IE users, the quicker they’ll become and migrate to something better.

This is exactly where the second part of my point in all of this comes in: we create the Web. If we, as designers, developers, users and webmasters, join forces against IE, gradually all sites would be replaced with purgatory pages for IE and more people would be coerced into using standards-compliant browsers. After a while, people would realise that it’s not the page that’s the problem, it’s their own software.

Obviously, it’s rather impractical to start this on sites that have a primarily IE-centric user base; on our personal sites, however, we really have no reason not to tell IE users what’s wrong. Some of the IE users won’t switch out of the fear of new technology or general stubbornness; they’ll end up with a terrible user experience and will probably ignore us to read something somewhere else. That small, Luddite section of the demographic is probably not worth our personal time, anyway. The people using IE that do value our content, though, will accelerate their switch as more and more related sites stop supporting IE6. Just like we have made the web unfriendly for IE4 and Netscape, making the web unfriendly for IE6 will make those people upgrade, too. After all, technology is about progress. If we’re progressing, it only makes sense to take our users with us.

Article Abstract

Posted 14 November 2005. Approx. 827 words.

Internet Explorer 6 is useless. Through the power of the blogosphere, we can move the majority of the userbase to something better.

Comments

Comments RSS Feed for “this post”

Haha! I love it! I really hope more folks start doing this. Heaven knows how badly the internet needs some accountability; and it would make our lives so much easier.

Great implementation! This would be an interesting movement if, suddenly, IE users started to find that a significant portion of the sites they visited targeted them specifically with “error messages” like yours!

I can only imagine that the overall effect would be stronger if the anti-IE message was uniform — repeated stimuli and all that. Have you considered CC licensing your code and writing up quick install directions for less-savvy designers to do something similar? I hate to speak in terms of blog memetics, but this could go places!

Right on. They call them standards for a reason.

Damn the Man!!! Save the Empire!

Eston,

Based on your post, I gather that your site generally renders properly in IE6, except in one instance installed on one friend’s computer. Is this correct?

As I see it, the cause of the problem could be:

1. Your code
2. IE
3. A specific problem with a specific instance of IE on one computer which may have been corrupted by a number of factors including the user, spyware/malware, or other (and could’ve happened to any browser installed on any OS)
4. Other

I’d suggest identifying the cause before pointing the finger at IE6 for not being standards-compliant.

And if/when you do identify the cause, and if it is a bug in IE, I’d be happy to forward it on to the IE team here at Microsoft. :-)

- Adam
This posting is provided “AS IS” with no warranties, and confers no rights.

Adam,

Thanks for your defense of IE, but we’ve found that the source of the bug seems to be two-fold.

1. It has something to do with his IE. His seems to render the page improperly from all other versions of IE we’ve used, even at high resolutions.

2. Another form of the same bug can still be replicated at low resolutions (800×600 or very small IE windows) on multiple copies of IE6. The source of the bug is actually rather well known and is known by the IE team already- it’s the famous IE Peekaboo bug, and it’s actually pretty easy for me to fix. The very reason I don’t, though, is because it’s a bug that shouldn’t be happening, yet has happened continuously without repair from Microsoft for over three years. This bug is left there for the very reason that Microsoft has all but abandoned IE6, and due to this, it is – and should be regarded as – obsolete. We have shuttled IE4 and IE5, as well as non-Microsoft browsers such as Netscape, off into the realm of obsolescence, and whether IE7 is ready or not, IE6’s obsolescence date is long overdue.

If IE’s CSS rendering worked properly and played well with standards, this problem wouldn’t exist. However, this bug has been around for a long, long while, and to date Microsoft has done nothing at all to fix it other than point to IE7, which still hasn’t materialised on any widespread scale.

IE7 may be a wonderful fix to these problems and will – if Microsoft is being truthful with their marketing speak – be a godsend to the world of web design. In the meantime, however, they have known about this specific bug as well as a whole slew of CSS bugs that not only have been brought to Microsoft’s attention on numerous occasions, but have been left unrepaired in IE6 for years now.

As far as I’m concerned, Microsoft – as a developer of what is the most widely distributed browser on the planet – has a responsibility to make their software play nice with what the W3C dictates as standards, and not attempt to strongarm the rest of the world into having to hack the hell out of their stylesheets to fix the problems in obscure ways because they are forcing these issue resolutions and CSS bug fixes into IE7 as opposed to releasing patches and updates for IE6 like every other browser company on earth is doing.

(Also, what’s with your comment statement under your signature? Are you posting from work?)

First off, I never set out to defend IE6. :-)

Second, I don’t disagree with your decision to design a standards-compliant page, and notify IE6 users that the content may not render properly (though you may want to specifically call out “Internet Explorer 6″ in your warning since IE7 is in Beta and on the way).

That said, it seems you worked hard to ensure cross-browser compatibility, then were (justifiably) upset when your friend couldn’t render the content properly. The cause is still unknown, right? If you find the cause, and it is in fact a problem with IE6 code/design, please let me know!

Regarding Peekaboo, and other CSS rendering bugs, the IE team has definitely owned up to these. The problem, as Chris alludes to in that blog post, is that releasing a patch now that changes the way IE6 renders may break a bunch of sites that expect IE6 to render a certain way (even if it is the wrong way).

Assuming web developers are checking browser versions, and not just browser names (which you in your code, except when displaying that IE warning ;-)), shipping these fixes with IE7 will fix the problems without causing new ones.

Also, sometimes these issues are caused by poor standards, not poor standards-compliance. For an example, and a call to action on moving away from CSS hacks in IE7, check out this post.

As for the comment at the end of my post, just a good habit when posting in some capacity related to my job. Anyhow, I understand the IE6 frustrations, and hope this helps. When the first (officially) public beta of IE7 is available, I hope you’ll give it a try and help us make it better.

- Adam

Hey, I remembered this post from a few weeks back today after seeing a similar idea put into effect elsewhere. This application, though, can get a bit more ornery with Luddite visitors, if you’ll let it: varying degrees of “seriousness” can show IE users a big ol’ message bar above the page or actually block all content completely. It’s from Downhill Battle, so it’s gotta be good, right?

Unfortunately, I’ve yet to see an accessible execution of this idea that allows viewers to disable the bar for good (likely because both obvious ways of doing it — user accounts and cookies — require server-side processes). The office guy who loves my blog but operates under the a capriciously restrictive IT department might not like to be reminded of his plight every time he pays me a visit. So among web publishers worried about visitor counts (which is to say, all of us), no invasive technique like this may ever take off. I for one will be interested to watch it.

I saw that yesterday as well and was actually going to write an update to this entry about that. Supposedly, it’s been around since before I wrote this entry and I just didn’t know about it.

There is one show-once-and-disable way to do it, but you’d have to do it by IP and record all IPs that have accessed your site in a database, which would be huge and unwieldy. At the same time, messages like mine (those not as invasive as ExplorerDestroyer) work much like IE’s warnings do: they never go away.

Looks like your “oldbrowser.html” page isn’t there anymore. That’s extra mean to anyone using an old browser! Redirecting them to a broken link, you bastard… =)

hi
1o81cm22caoe15fm
good luck

Write Your Own Comment

To preserve legitimate discussion, please use your real name and email address.
Your email address will not be published. Derogatory comments will be deleted.

External Discussion

No trackbacks have been posted to this article yet.