I think most web developers will agree that it is often easier getting something to work well in most browsers, but not as much in IE.
I was wondering: When designing/developing a website, how important do you think IE-friendliness is? I mean, how worth it is it to bust yourself trying to make a website that works perfectly in major browsers work the same in IE?
If you are talking about anything, that isn't just your small, private programming-related site, then: Yes, you need to support the IE. If you develop a commercial site and it doesn't display in IE, you are going to lose many potential clients.
Furthermore: Really, it isn't that much work to make your designs IE compatible (unless you want to support IE6, which I personally don't do). Your site doesn't have to look exactly the same. But the basic functionality should be there.
Completely dependent on what your target audience is.
On my personal site I may not care if it's very IE friendly. On a public blog, I might make sure all functionality is there and it doesn't look horrible. On a public web site for a client IE friendliness is very important, unless they are specifically only targeting newer browsers with the features they want implemented.
All you have to know is your customer demographics (in terms of browser usage).
IE is quite popular, so, like nikic says, beware not to cut yourself off from a large client base.
Stats speak for themsleves, check out Wikipeadia, some random IE vs Firefox comparison or even Forbes, a pretty reputable source of information. Even with the worst case scenario for Microsoft, they still hold at least one third of the market if not more than half in places.
Use those hard facts to make your own decision as to whether you can afford to lose such a large segment of the market.
First, check what most of your visitors use. Personally, i use Google analytics to keep track of this, as it's easy to set up with tons of data.
Second, decide the level of support you want to provide for IE, based on the data you now have.
In any case, even if IE is < 1%, your basic functionality should work in it. If it doesn't, chances are you're doing something wrong. Furthermore, it is generally a good idea to write the website to work in the major browsers first, then add the hacks to get it to work in IE. If IE is the major browser, I would still design it to work in the other browsers first.
For 3 years now, I've been telling QA and the rest of the org: If there's a bug with Safari/Chrome/Firefox let me know. If there's a bug with IE, use a different browser.
Granted not all business can afford doing this. Or can they? The time I've saved from ignoring IE more than compensated from any loss of revenue from those two users who still use IE.
Your client base may not be a collection of individuals, rather than people from different organisations. As such, they may be subject to company policies. Most of the client base I deal with are stuck on IE6 due to corporate mandate. Even if IE6 was a small fraction of the overall internet traffic, it probably makes up 80% of our client base.
A good framework will abstract out the majority of the differences and make IE* less difficult to work with. We use ExtJS.
I think you're betraying your personal prejudices when you talk about making "a website that works perfectly in major browsers work the same in IE". Much as you might like it to be otherwise, the fact is, IE is not just a major browser, it is the most major browser. Site-by-site, obviously, anything is possible. But across the whole worldwide internet, I've never seen any credible claim that any other browser has gotten close to catching IE's market share yet.
But, four things you want to keep in mind:
As crazy2be said: measure. Especially if, say, you're rebuilding an existing site, you should be able to check out visitor stats and have a good, clear idea of what the browser breakdown is before you get started. If you're building something new, you won't have this opportunity before you start, but keep an eye on the stats and let them give you guidance as you make changes and additions to your site.
If you're building a site for a client, it's not your decision. They will decide what level of cross-browser support is mandatory, and you will have to deal with that - that's what they're paying you for! You should offer them advice, certainly. You should charge them more if they want the site tested and working perfectly on a wider range of browser versions, certainly. But if you're doing work for hire, you gotta produce what the man who's paying the bills asks for.
Don't be thinking about "IE" as some monolithic entity. IE6 causes far, far more pain than all the other versions put together. Abandoning IE6 is a decision which will save you a lot of pain, and may well not hit many of your visitors. Have a look at ie6countdown for some worldwide stats on IE6 usage.
It's really important to triage your support or lack of support into two categories: "make it work properly" and "make it look the same". You can save yourself a lot of effort without losing many visitors, if any, by downgrading inferior browsers to "make it work" level. Some fancy navigation doesn't work in IE? Fix it. Some semi-transparent bit of decoration is displaying solid and look ugly? Screw 'em, if they didn't want ugly they'd use a better browser.
