The problem he's missing is simple math. We're a small startup making apps for social networks, and about 7% of our customers use IE6. (I keep meaning to check by revenue, I bet it's less then, since presumably a lot of them are using it at work where they can't just whip out their credit card.) Unfortunately, getting our great UI to work in IE6 takes up some percentage of our time. As time goes on, the percentage of users drops and the percentage of our time increases. I'm not sure if they've crossed paths yet, but if not, it's close.

Sure, if you have unlimited resources, develop for IE6, Opera, whatever else. But when you're on a budget, and you have to choose between making your app 20% better for 93% of your users, or making it work for the extra 7%, it's not a no-brainer.

He is not advocating that you get your great UI to work on IE6. He is advocating that you get a lame UI just for IE6 users.

Which takes nearly as much, if not even more effort.

It really doesn't. I'm an experienced front-end coder, and I can tell you right now that the amount of extra work I put into IE6 support is really not prohibitively high. The bugs and issues in IE6 are well-known and well-documented, so it's a simple matter of professionalism to be aware of these things, and account for them, as you code.

Depends on what your frontend looks like. The average run-off-the-mill HTML site can ofcourse be made to work in all browsers with little effort. The real pain starts when you want to do interesting stuff. Advanced AJAX and javascript magic just doesn't work in IE6 without horrible overhead - partly due to the lack of sensible debugging tools.

Or... you could just use a Javascript framework that takes care of that abstraction for you. This is seriously not a hard problem.

Well, frameworks go a long way but they don't help with the pains in the CSS area. When supporting IE6 you're generally limiting yourself to the capabilities of that browser. And for any non-trivial site you will run into hard problems regularly, unless you're leaving HTML behind entirely (ExtJS, Flash).

You can do entirely different fronts for browsers with different capabilities.

Which is hardly a worthwhile option for a reasonably complex website.

If we're talking about designer blogs or minor sites/apps that do not involve a great amount of development, then I absolutely agree with the author.

BUT if we're talking about large web applications that have serious development costs for every "supported" web browser, then his argument breaks down quickly. These costs come in the form of feature development, bug fixing and in customer support for released features. When a single web browser consumes the majority of these costs, yet the browser is continually falling lower and lower in usage, then it's time to put serious thought in dropping support for that browser.

I'm the only developer for a company that has created (and continues to support) some huge sites that require(d) a great amount of development. For the majority of these sites, IE6 is not falling in usage. It has hit - and stuck at - around 30%. Alienating 30% of a client's potential customers is just not an option.

That sounds painful - and interesting.

Do you happen to know what kind of userbase that is?

Since IE7 is pushed pretty aggressively by windows update I would think there must be a few large corporations in there who have rejected the update thus far?

At least I want to think that because IE6 being stuck at 30% gives me the shivers. On our sites IE6 has been constantly dropping since about December and is now at around 18%. That's still too much but at least the tendency is right here...

I can't be too specific but one site in particular that I maintain has a big user base in education (from nurseries to universities) and I think it's very likely that this heavily influences the IE stats.

I know from experience (I've worked as tech support in schools and a college) that these establishments are normally the very last to push out upgrades to things like browsers. There's a big "if it works, don't touch it" attitude.

Which implies that if a sufficient number of web sites cease to work with IE6, such sites will "touch it"?

Not really. Your average IT admin in a school isn't going to care that the students can't get to latestweb2gadget.com

> Since IE7 is pushed pretty aggressively by windows update I would think there must be a few large corporations in there who have rejected the update thus far?

Don't forget that IE7 isn't available on Win2k, which is still being used in academic and government organizations a far bit.

I absolutely agree. You need to be able to look at YOUR numbers and do the math based on your company's needs. Blanketed statements do not work here.

FWIW, most people who are dropping IE6 support are doing so when it hits the 6-8% usage range.

> FWIW, most people who are dropping IE6 support are doing so when it hits the 6-8% usage range.

I know - I want to know where these people get their audience from so that I can swap ;)

Stop doing government work :p

I like to eat - I'll take whatever work comes my way.

Same here. I'm at a huge site, and IE6 is still 30% strong.

There are simply so many browsers and devices being used today, it is hard (impossible?) for a site, particularly a small startup, to support everything. I think its important to figure out who your target is. Do I need support on cell phone browsers? What types of interactions do I need with my users? What groups will be my "bread and butter" so to speak? If I'm selling Macintosh accessories, perhaps I don't need to support IE6.

I think you have to make some decisions about what you want to support. I really don't think complaining to your users about the browser they have chosen helps. If anything, it drives them away. I do think that no matter what you choose, you should always give your users back something. You may not be able to support all browsers, but don't return a page that simply doesn't work and leaves the user wondering why. This may be easier said than done, but I think its a good middle ground between the complexities of trying to support everything and alienating users by only supporting your "chosen path"

What features of IE7 are difficult or impossible to emulate in IE6? In my experience, getting sites working with IE6 is really easy. There are very few things that IE7 can do that IE6 cannot. (It is funny that this argument is usually only made to disparage IE7 but never to support IE6.) To support IE7 you've already given up border-radius, display: table, automatic PNG transparency, SVG, a lot of the DOM, and a lot of the other things that make developing for other browsers so easy. With IE6, you can get a lot of the functionality of IE7 with just a little Javascript (already written for you by others), a few conditional comments, and a little extra CSS. You might have performance issues with adding all these workarounds, but workiing with poor performance is usually better than not working at all.

I'm not sure exactly why there's so much controversy here.

The customers of your website should always come first, and if they use IE6, support IE6. Making them upgrade is a surefire way to alienate them or encourage them to go somewhere else. (not to mention, if you already support IE7, supporting IE6 is not that big of a stretch)

I wonder if putting another browser on a CD and putting it in with a customers purchase would make any impact on the percentage of IE6 browsers visiting that site in the future. It could be a worthwhile investment if IE6 support could be dropped sooner.

For example if a business has many users that are on IE6, but the company put a firefox cd in the box of every order, would the number of users on that website using IE6 drop at substantially faster rate than if they didn't?

I used to get lots of AOL CDs back in the day. It didn't make me in any way tempted to go with AOL.

Some people must have though, or they would have stopped, and I wouldn't have collected a lifetimes supply of coasters.

"you shouldn't even maintain a list of supported browsers in the first place."

Does "any reasonably standards-compliant browser" count as a list of supported browsers?

If you have are a lot of IE 6 customers, it may just be impossible to try to force them to switch browsers without losing a lot of customers. Not sure we really need anyone to explain that.

However, this post seems to go beyond that and basically be quite passionately explaining ways that we can happily accept and just deal with the multitudes of browser incompatibilities and inabilities. Probably, this guy makes his living off of these problems.

This person's attitude is absolutely sickening to me. The situation with browser incompatibilities and deficiencies is out of control. Action must be taken.