Goodnight IE6?
After years of pain for developers and end users alike, is it finally time to put Internet Explorer 6 to bed?
Internet Explorer 6 (aka "IE6") is 8 years old — that's like 56 in company dog years. We're talking about an ancient web 1.0 browser here. With the release of 8.0, is it finally time to say "Goodnight IE6"?
A Love Hate Relationship
Let's face it, developers don't exactly love IE 6. How many times have we heard, "well, it works in Firefox." Some folks are just plain angry while some are letting their users know about it.
Fun and games aside, the issues with IE6 are very real. And so is the cost of overcoming them. We're seeing a notable increase in web sites dropping IE6 support and actively encouraging users to upgrade. We're talking big sites like Facebook, YouTube and Gmail.
The Problems with IE 6
Discussing the issues with IE 6 could be an article in itself. But to name a few:- No support for alpha transparency in PNG graphics
- No support for CSS rounded corners
- Drastically incomplete implementations of W3C standards
- CSS and JavaScript bugs galore ("IE6 css bugs" returns 10,300,000 results on google)
- Poor or non-existing support for modern javascript driven applications
These technical particulars may not matter. With enough time and money you can work around most of these issues. What matters is this: If the exact same level of support is required for IE6 as for modern browsers, the development costs start to add up.
Dollars and Cents
The person in charge of the web budget is forced to ask themselves: "Would I rather spend the remainder of my budget getting IE6 to appear to be a modern browser or would the money be better spent adding a feature I know my users will love?"
It can be a hard question to answer. For sure, a brief look through your visitor logs will likely reveal a sizable chunk of IE6 users. Abandoning support entirely isn't an answer. Instead, perhaps it's time to consider what 'support' means.
The Tradeoffs
The first question to ask is: Do websites need to look exactly the same in every browser? The answer is "No".
In fact, no matter how hard you try, no two browsers will render the same web page identically. That's OK because the real questions to ask are a) "Does the site need to work (support business and user objectives) in all browsers?" and b) "Does our brand need to be conveyed in all browsers?" The answer to both of these questions is a most definite "Yes".
Let's lay it out. Let's say it out loud. Supporting IE6 means that the site will work as intended and the key content and brand messages will be conveyed.
In the process of designing a site, we often proscribe and document what we consider to be the best user experience. Sleek drop shadows, sassy bits of AJAX functionality and high touch user feedback all come to mind. For modern browsers, this is all doable in a reasonable amount of time. For IE6, things get a little... whacky. If we choose to spend development budgets on new features instead of trying to make an 8 year old browser play nice, how do we still ensure that IE6 and other older browsers meet our definition of support?
Progressive Enhancement
Starting from the perspective that content is king, we design the baseline user experience ensuring that all content is available to all users. This is the design without bells and whistles. Once that's set, we can sass it up for everyone that has a browser that supports said sass - we progressively enhance the user experience.
Twitter employs a simple bit of progressive enhancement for IE6 users. In order to post status updates, Twitter defaults to soft, square buttons. If you load a Twitter page in a modern browser, all of the buttons display with rounded corners. This is achieved without images using a tiny bit of CSS 3. IE6 doesn't support this CSS so it instead displays the default soft, square corners. Simple. Is the site still supporting IE6 users? Yes. Is Twitter's brand still conveyed to IE6 users? Yes. Did Twitter save a lot of time and money and bandwidth as well as ease the pain of site maintenance by not relying on images? You bet.
One better, bring up twitter.com in a text based cell phone browser. All of your content is available to you. It's not styled and there's no fancy AJAX when posting updates but by employing progressive enhancement at the root of their application, the Twitter team is supporting even the most basic of platforms.
Inside the Numbers
Is it time for your organization to change its idea of "support" for IE6? Most likely. But to what extent? Analytics is key here. Start by looking at your own traffic data. How many of your users are using IE6? More important than today's numbers, take time to look at the overall trends. Which direction are they headed?
Who still uses IE 6?
There are two main groups of people still using IE6. One group is unable to upgrade and the other simply doesn't know better (or care).
Those unable to upgrade are generally made up of either folks on older operating systems (Windows 2000, Windows ME and Windows 98 can not be upgraded to even IE7) or people in locked-down network environments where the cost to deploy updates keeps change slow. There's really not much to do to encourage these users to upgrade. When the economics make sense, they'll upgrade. There is hope that since IE6 is now two versions old it will inspire network admins to finally make the jump, hopefully directly to IE8. If you're looking to convince you're own network admins to upgrade to IE8, google IE6 Security Issues and send them a few links to scare them into upgrading!
Those that don't know to upgrade (or care), are partially our fault as website builders. There is no compelling reason to upgrade if we continue to spend significant time and resources to make IE6 users think they're using a modern browser. Facebook recently took this seriously and now posts a message to all IE6 users asking them to please upgrade. YouTube is more blunt, telling users in a large box on top of the site that their browser is no longer supported and they should upgrade immediately. A number of smaller sites are flat out blocking IE6 users from accessing their site.
Redefining Support
There's no reason to block IE6. Redefining what 'support' means to your organization and employing progressive enhancement techniques during the design and development phases will keep things working well for IE6 and other older browsers for a while. For most sites, it's definitely time to say "Goodnight IE6" but the good night story can be a gentle, pleasant one.
IE6 has single-handedly tacked on HOURS of dev and QA time to the majority of my projects. We've attempted to ignore this browser on some projects - thoroughly informing clients - yet we invariably have to account for it at some point once the site launches. It's expensive, mucks up our code and limits our design options.
Make it go away!!!
I'm right in line for retiring IE6 based on the pain it's caused myself and other developers. Heck, I'll even camp out the night before to get in line, Star Wars style!
My concern is that IE6 still holds a significant market share. This report is showing a foothold of 27%, which is still frustratingly relevant: http://marketshare.hitslink.com/browser-market-share.aspx?qprid=2
I find myself asking myself, "Self, who is your audience here?" If I'm building a marketing website to be viewed by a vast demographic, I should probably consider IE6. If I'm building an web application to be used by people in the technology industry, I might be able to get away with it.
I'd imagine big fish like YouTube who have recently stopped supporting IE6 can justify the cost based on what it takes to support it, but for the smaller fish, it could be a lost conversion.
So much time wasted by so many good people who's talents and creativity could have soared to loftier heights.....had we all not been wasting our precious time phutzin' about - to no gain to anyone - with IE6!
I gave up coding for it about a year ago. A total waste of everybody's unrecoverable time IMHO.
Since Microsoft support for IE6SP1 ended back in 2006 it doesn't seem to make much sense to keep supporting or using it at the workplace. Plus, NO TABS.
Man, I decided to finally cut it recently for any personal sites. I use a tool called "IE6 Update" which puts a notification bar at the top telling them to update. I ensure there aren't any float drops and that the site is usable, but I won't go out of my way for rounded corners, jQuery, small padding issues, etc.
Hopefully with the arrival of Windows 7, people who LOVE using preinstalled browsers like many network admins will at least be using IE8.
a tad extreme maybe but in agreement with your article http://mashable.com/2009/08/18/ie6-offenders/
I've been using something called "Universal Internet Explorer 6 CSS" http://code.google.com/p/universal-ie6-css/ with a few more recent projects. It works like a charm, provides a look and feel that can be easily customized, and saves hours of dev time. Obviously this won't work for everyone -- some clients require that the design remain intact -- but it does give you a great starting point if you *can* avoid designing for IE6. Thanks for the article.
I think we are too hasty; forcing IE6 out is not a good idea yet. Let me tell you why. The group of people using IE6 at work is by far larger than most of us realize. The problem is that people actually do not just use it for "work purposes" but also for their home shopping, home entertainment, etc... you get the picture. For as long as companies do not upgrade their systems (due to costs - I know poor argument) we are stuck developing fixes for IE6.
However, as was pointed out as well, there are many ways to ease the pain of developing for IE6 and for some of us it is hardly a frustration more an acceptation of the day to day reality.
I just happened to be on SkiHood.com and got your message "Hi! Looks like you are using an old browser. Things may look a little off around here. For a better internet experience, consider upgrading your browser or trying a new one." This was with IE7. I probably dislike IE as much as the next person, but I don't know that I would take things this far, especially on a site like skihood.com which is going to have a very mainstream population.
I have just finished building a large website/store with an owner assured me she used Firefox as her main browser. The 4th week into the project we started coming up with problems in the appearance of the site, which I spent hours fixing based upon screenshots the owner emailed me; sure enough they all issued from the owner going back and forth from IE6 to Firefox. We spent the usual hours adapting modern CMS functions to work in IE6
When I discussed the issues with the owner and let her know about display and security issues with IE6 I was assured that she was going to upgrade that day…needless to say that didn’t happen.
The last issue came about when the owner was adding items to her store (using IE6) and created an error in the SSL portion… this time I billed extra for my time. The owner upgraded to IE8 that same day.
Post new comment