Why we don't support Internet Explorer (2009)
engineering.silkapp.com
engineering.silkapp.com
> Google Wave team also chose to drop Internet Explorer support
3 years is a significantly long period of time in the browser world and Chrome was a 1 yr old browser and there was no IE9. Also, there was no windows phone 7.
Once in a few months, an article pops up telling how a startup saved $100k by not supporting IE or how hundreds of developer hours were saved by not caring about the things that IE failed to render properly. It narrows down to the number of users an app or service has and also the demographics of the users. If you are building a hipster social network or a video post processing tool, ignoring IE is going to save you significant time since it can be assumed that a good percentage of the users are going to be on macs using the latest browsers. On the other hand, if you are catering to e-commerce, finance or any other common industry even thinking about ignoring IE isn't wise at this point. With jQuery and graceful degradation, fixing the major issues shouldn't be really hard.
PS: Can just-launched-starup ignore IE for their app? Probably.
Can Pinterest ignore IE8? Of course not. Even if it is 10% of their user base (I bet it will way more), they will be ignoring a few hundred thousand or may be a few million users!
Unless you need webgl, appcache or contenteditable there's no good reason not to support at least IE9. The effort to support it is negligible, if you're already doing opera, firefox and webkit. And if you're doing contenteditable on all three of those, you might as well do IE as well and make those gray hairs actually mean something.
There are large populations of users who are forced to use IE6. Still. In 2012. They are not allowed to upgrade their work computers even though they may hate IE6 and use a modern browser at home.
The reason is usually that their corporation, hospital, organization, agency, etc. paid lots and lots of money many years ago for custom software that cannot run on newer versions of IE, or is not certified to be reliable/secure/accurate/etc. on newer browsers, and so those organizations must either force their users to use IE6, or pay lots and lots of money, again, to replace/re-certify custom software that is still working perfectly well.
So the business case for forcing your employees to stick with IE6 is still very rational in some places, even today.
If you're developing a new way to tweet facebook likes or something of that ilk, that's just fine; anyone using IE6 in your userbase should be encouraged to stop.
But if your users might include any of the sorry souls who are forced to use IE6, at the very least you should be kind when you tell them they aren't going to be able to use your site.
Yeah, out of context but I think this is telling. I know a lot of devs who ignore one browser or another or ignore mile or ignore desktop or only do responsive design as opposed to building out a mobile only site and vice versa. In most cases they do some mental gymnastics to convince themselves they don't need to worry about a certain segment of the market. I think is is dangerous in that it can cost you a lot of money. Many times these decisions are made because they don't want to pay for supporting X or don't want to learn how. Then they justify it by saying that market segment X is too small to support. Instead of being based on reality it instead is just a self-fulfilling prophecy. Market segment X doesn't use site Y not because they're so small but simply because X is not supported. Imagine HN launched today. Today is HN's first day and for some reason PG said we won't support IE because HNers are all using "better" "modern" browsers. Then he checks the logs and sees a very small portion of IE users that steadily shrinks. Is the portion small because Hackers who use IE are few and far between? Maybe. Or maybe it's because people who visited HN on IE couldn't access the site and so they didn't spread word to their friends who may also use IE and of course they never return because the site never worked for them to begin with and so the number of IE users remains small enough to be ignored. But in reality we know that there are definitely enough HN users on IE to justify supporting it. I know because I see my own logs from HN referrals from posts I've made in the past or people reading my profile. The percentage is lower than other browsers but definitely more than a trickle and surely not low enough to justify not supporting IE if I ever had to make that decision.
My whole point is that I think it's dangerous to make assumptions about your audience because they often become self-fufilling prophecies. You can't say a certain platform isn't worth supporting until you have data to show it and you can't get that data (at least not accurate data) unless you support the platform in question to begin with.
I do agree with you about Windows phone though. At this point, at least in the mobile space, I think companies would be smart to make their mobile browsers compatible with what's out there now rather than releasing a mobile browser and expecting developers to cater to it. I wonder if the iPhone would've taken off like it did had they decided to release a very stripped down version of Safari and expect devs to modify their sites to support it.
If the latter, give me a break. By my count, over 50% of our traffic comes from IE.
Extra development to support IE7 and above last year: 2 hours.
You're telling me it's not worth 2 hours per year to stop yourself from alienating 50% of your market?
Plus, IE8 and IE9 are pretty compatible in that whatever works in Webkit/Firefox will pretty much work every time in IE8+.
So this tells me that explicitly testing in IE8+ is a waste of time, then.
If you'd rather have it "definitely work" for those users, then test it.
Lemme also say that I'm not a Windows user, Chrome is my fav, etc. But from the testing I've done and people I've talked to, that is what I base my assessment on.
That's certainly not true, to my profound sorrow; they're both solidly on my support matrix, because quite a few of our users are obligated to use IE6.
See my top-level comment for more details.
Now I have to say I was wrong about IE. It's not the awful monstrosity of a browser I've learned (or rather been taught as the case many times was) to hate. There's no disputing the fact that Internet Explorer from v6 to v8 got stale, held the web back, and was a pain to work with. The IE team got so used to creating the rules rather than playing by them that they seemed to think their reign would last forever and they seemed to ignore all standards for a time (in the case of the Trident CSS box model I have to admit I think that makes more sense). However this is not then. Today's Internet Explorer has a bigger problem with prejudice and it's reputation than anything technically inferior.
There's still a lot left to be desired from IE like a shorter release cycle that adopts new specs and standards like the Webkit and Gecko browsers are doing but despite that there's still not very much you can do in Firefox, Chrome, Safari, or Opera that can't be done with roughly equal effort in IE. And with IE10 the gap is closing even more. Sure, Microsoft has made some intentional decisions to support or not support certain features or purposely implement them differently but both Firefox and Chrome make similar decisions and are applauded for them or at least don't get chewed out by the techie crowd for it. I've learned that IE users as a whole are actually a different kind of user altogether with different needs and competency levels than are users of other browsers and I'm talking about those users as a whole, not just the ones working on corporate networks. That considered, I think IE has done a great job of walking the line between keeping up with the pace of the web and serving those users.
This is all coming from a reformed IE hater. I learned that in most cases the browser wasn't the problem, it was my laziness or lack of skill that was. That was hard to admit. To all the people thinking of dropping support for IE or hating on it as its so easy and trendy to do, I'd tell you to really give it a shot and learn how to support it. You'll be a better developer for it and you'll find it's not as bad as it was when you swore it off all those years ago.
Edit: Spelling and brevity (tried my best to shorten this, sorry)