Time to ditch IE6 support already?
Almost exactly a year ago I was musing about the possibility of dropping support for Internet Explorer 5 when testing websites. Time marches on, and now we are seeing calls to ditch support for IE6 too.
Earlier this week I read on Web Designer Wall a new post titled ‘Trash All IE6 Hacks’. The article’s author, Nick La, argues that as web designers we have been complacent, and need to send a strong message to IE6 users that they should upgrade. Nick proposes that the way to do this is to eliminate all IE6 specific hacks from our workflow (including hiding IE6 specific stylesheets inside conditional comments) and embrace modern browsers’ support for transparency in PNG images. The theory goes that if we reveal IE6’s “ugly side” to its users, they will realize the error of their ways and promptly upgrade their browser. A poll at the end of the article reveals that this is a point of view shared by 85% of Nick’s readers.
Call me old fashioned, but I can’t see the sense of intentionally alienating a whopping 52% of our audience, which is IE6’s current share of the browser market (source: thecounter.com). I realize I take a conservative approach when it comes to browser support (IE5.x usage was less than 2% when I dropped it from my testing suite), but 52%? That’s downright radical!
If web designers took up the “death to IE6” clarion call en masse, the results would be very, very ugly, and I’m not speaking merely of the way web pages would look. Imagine a web where CSS dropdown menus simply don’t work for over half of users and you’ll have an idea of the type of usability and accessibility issues we would be facing. Who would Jane and Joe Average blame for this failure? No doubt the owners and developers of ‘broken’ websites would bear the brunt of their frustration, not Microsoft or IE6.
When it comes to the design of web designers’ personal sites, blogs and so forth, I have no beef with embracing the latest and greatest rendering technologies. Designers’ personal sites provide a perfect opportunity to advance our medium away from the constraints typically associated with clients and ‘average’ users. But when it comes to our mainstream commercial work, I consider such an exclusive stance to be irresponsible in the extreme.
We may curse the popularity of IE6, but for the time being we are stuck with it. I eagerly await the day when IE7 muscles its predecessor out of our testing suites, but until that happens we must continue to exercise patient restraint.
14 thoughts on “Time to ditch IE6 support already?”
Comments are closed.
Yeah, it would seem dangerous for companies with a lot of public exposure to just let their pages fail miserably in IE6. I think a cool way to do it would be to just let the user know that they should update their browser. This can be done subtly, like a little message on a sidebar somewhere, or even more in your face, like a DIV that pops up on page-load for brief amount of time or until the user closes it.
I don’t know… something less… crude. Heh. Good post!
@Alex – I thought of that too. It makes me cringe though, a real throwback to the bad old days of “this site best viewed in Internet Explorer with an 800×600 monitor resolution”! Still, I agree such an approach is far preferable to simply shutting the door in visitor’s faces.
Mmm, I hadn’t thought of that. Yeah, the likes of those days are best left behind. If it were my own homepage, I would try to cut the message with humor or rhetoric (e.g. “Internet Explorer 6 will stunt your growth.”), but that is, of course, silly and not viable in all cases.
I guess we must just be patient or do our best to promote the modern browsers.
Yeah I do not think we should cut out support for IE6 either.
I would rather suggest, “code so you do not need to use hacks” and then you do not have to worry about spending extra time getting it all working right.
I do realise some websites you may need the png or max width hacks but most of the time, code it properly….
@Jermayn – I think that’s a very sensible approach. I have a couple of techniques I use frequently that require workarounds for IE6 (Suckerfish, in particular), but by and large I code in such a way that IE doesn’t cause me many headaches.
When you compare it to IE5 (don’t even mention IE5 Macintosh in my presence), IE6 is really a piece of cake to keep happy.
btw what is a good way to get IE6 to show transparent pngs??? I cannot seem to get anything to work :(
answer here
I found Nick’s “Trash All IE Hacks” to be arrogant and shortsighted. 52% is a majority share. Hell, even his initial 37% is a massive percentage. When our clients and more importantly our customers are still using IE6 (which is a modern browser, still supported by the developer) it is our *job* to support it.
Nice article Jonathan.
While *forcing* people to upgrade to a better browser is a great idea in the imaginary world, in the real world, imagine having to tell your clients that the broken site they are seeing in IE6 really does work, and that furthermore, they are to blame for having an out of date browser is to say the least, a hard sell.
Half my clients don’t even have sufficient access to their systems to upgrade their own browsers, let alone the know-how, not to mention that only a year ago I was having to support IE5.0 for a company whose IT was refusing to update.
Not supporting IE6 would be a quick way for a web developer to lose interest from all current and potential clients. IE6 will hang on for a while for a number of reasons, the ineffectiveness of Vista’s release, people who don’t like change, businesses who upgrade slowly (I know a client of my company still uses IE5, though they are planning to upgrade to IE6 later this year, they may never get to IE7!), and so on.
Blocking out users of certain browsers or bringing back the “This page best viewed in….” messages is also a bad idea, there are no longer so-called browser wars as most browsers are getting closer to the standards. It is possible to deliver a reasonable site in IE6 without too many hacks and I would certainly rather do that than block out half the users of the internet! I do however like to leave adverts to upgrade to Firefox within IE conditional comments, but I know that’s not going to make a huge difference!
Finally, your point about who gets the blame for bad websites is totally right. Your average user doesn’t know about browser bugs or standards support. When they see a broken website they will leave and not return, not something you want to inflict on your clients.
The way I see it is that we providing a service (websites) and it is our job to make sure it works the best no matter what they are viewing (using) the website.
It stinks of slackness and disrespect imo :twisted:
@Nick:
Well put! It’s all fine and well to code first and foremost for modern browsers, but to disregard the other 50% would be, as you say, “arrogant”.
@John:
Good point. It’s safe to assume that in most corporate environments IE is the preferred web browser, and many IT managers are understandably cautious of doing company wide browser upgrades. It is futile to punish end-users for a situation over which they have no control.
@Phil:
My feeling as well. I think it’s possible to design for modern browsers while ensuring pages degrade gracefully in old – yet still relevant – browsers such as IE6. Besides, within a couple of years the majority will have upgraded to Vista, with IE7 pre-installed.
I’ve caught this article a bit late, but doesn’t anyone remember IE5? Times are much easier these days if you are only supporting IE6.
It’s not worth complaining about as most problems with IE6 are solved by using the same solutions. (float:left;, height:1px;. etc).
To be honest I’m kind of worried about what happens when we do drop IE6. People will get a bit complacent when they don’t have to try hard to work on all browsers, resulting in some “lazy” coding. Probably less testing too.
I wouldn’t call it lazy coding, actually it frees up time, which can be better spent on real usability issiues.
Im one of the few, who have ditched support for IE6 and ealier browser versions, without even looking back. The code simply becomes a mess if you must support all these broken browsers!