It all depends on your end goals and your audience. I don't have #'s off the top of my head, but let's say IE8 accounted for 40% of browser usage across your visitors. But IE8 is old. Are you willing to say to 40% of your potentially paying customers "Sorry, but your browser sucks. We're not supporting you."?
If you are, great. No coddling necessary. But for other people, they may not be comfortable with the percentages yet, and I think that's perfectly understandable.
You wouldn't say it like that, I hope. You would say, “Here's a link to a browser upgrade that will allow you to enjoy the best product we can offer you. It's completely free, and takes only a few minutes to install!” Also, IE8 accounts for less than 10% of browser usage worldwide now, so the odds of a company having it be 40% are rather slim, unless they specifically target enterprise in which case they dug their own grave and I have no sympathy for them.
Oh no, of course not. I'm just trying to make the point that some companies may not be comfortable with whatever IE8 is at right now. I think you can substitute "IE8" for any browser, and "40%" for whatever that browser's current market share is.
Coincidentally, I work in the multi-family housing industry, and one of the industry titans requires IE to use their product--which everyone uses--so we have ~82% IE share on our own product (which is supplementary to the titan). Of that, IE8 stands at ~31% of our total site traffic, which is compelling enough for our company to need to support it. (IE9 is only ~39%, with IE6/7/10 making up the remaining 12%.)
So until that industry titan changes their way, we're stuck in an IE rut.
This is exactly right for multi-family housing. I can't see any sort of startup or disruption coming in anytime soon without taking years of hard work to reach 50% feature parity to meet what the industry expects are standard features. (Or maybe I've been working here too long to get past that notion.)
What exactly is shitty about it? Old browsers can still display text, images and hyperlinks. There are millions of use-cases perfectly covered by them.
For starters, users expect modern-looking websites. That includes users on old browsers, thanks to the amount of bending over backwards done in the web industry to provide those users with as much of the modern browsing experience as possible. If you serve up a page looking like the c2.com wiki to a non-technical user, they're not going to be too happy about it.
Far more so than that, though, what's shitty about having to support IE 6-8 is the complete lack of consistency. Even if you aren't doing anything fancy, you have to spend a hell of a lot of dev time if you want things to look the same across browsers when they're included. Hours and hours of wasted time is pretty shitty, especially when the work involved is so aggravating and, from a creative point of view, utterly pointless.
In effect, your question is like asking what would be shitty about developing apps for a 10-year-old Palm Pilot. There are millions of use-cases perfectly covered by those too, but that doesn't mean an iPhone app developer is going to want to spend time porting his apps to one.
Even if the users expect "modern-looking" websites - it is not something that cannot be done with IE6.
Secondly, you don't have to design 1:1 pixel-perfect copy of your website that looks the same on all browsers.
The main thing - if your developers cannot degrade the website functionality - they are seriously doing something wrong - get more experienced
You can as well design Flash site and call it "progress" and tell the rest of our users to go somewhere else. Now, how is it different from other "modern" technologies? Its W3C support?
"but that doesn't mean an iPhone app developer is going to want to spend time porting his apps to one."
If your customers still use Palm Pilot, then why would you design your site ONLY FOR IPHONE??????
The developers always can degrade the functionality; whether they can is not the point. The point is how long it takes them to, and how much they could be doing for other users with that time. Getting more experienced at making products that work on broken platforms is not a good thing. That is not good experience. That is a symptom of a broken industry and it makes your developers worse, not better. It's like people who say PHP isn't that bad, you just have to learn all of its quirks. Learning its quirks doesn't make you a better programmer, it makes you a programmer who is able to use a specific poorly-designed tool. That experience and knowledge is not portable and is not valuable outside of a specific arena, and is not the kind of experience and knowledge that developers should be working on.
If your customers still use Palm Pilot, you've chosen the wrong market segment. Yes, make something people want, but you're allowed to be selective about who those people are. If you're a hacker, it's very likely because you enjoy creating and working on interesting and fun things. Making websites work in IE6 is not a fun thing. There is enough value waiting to be created that we don't have to waste our talents and energy on this kind of nonsense.
By not supporting old browsers with all of their inconsistencies, a team can save a significant amount of development time. That development time can then go directly towards new features, bug fixes, etc. It's not so much that the latest browsers can do so many new things (though in some cases it's that as well), it's that you can actually develop for them efficiently without diving into a rabbit hole of IE compatibility issues every time you add a feature to the site.
Of course, it is easier to support only 1 mature platform , 1 database, 1 application server, 1 library of whatever. However we don't have this luxury even on the backend, and you cannot seriously expect it to be different for browsers.
And if you don't develop cross-platform because of lack of time this is fine. Please, don't cover it with insults about dumb CEOs and ignorant lame users - bad excuse for bad programming habits (not personal, just generalisation).
Again, there is no technical reason not to give workable version of the website to all major browser users. It doesn't have to be a pixel-perfect copy. It just has to work.
I'm not asking to support one platform or one browser or one anything. I'm asking to support the standards. Industries have standards for a reason.
I didn't insult anyone, CEOs or users, so your generalization is a very poor one.
The technical reason for not giving a workable version to IE6 is that it can take as long to make a workable version for IE6 as for all other browsers combined. IE8 is not that bad, but it's similar in principle. If I have to spend 50% of my time on 10% of my users, I'm better off doing twice as much for the other 90%.
That all changes if your client uses browsers that are more then 2-3 years old, and you are financially dependent on your client being served properly.
You can tell your client to use a new browser. It's all in how you phrase it. Everyone in this thread seems so damn scared of their clients. You have a way to improve their experience of using the internet and using your product. That's an opportunity to embrace, not something to run away from.
Naturally that depends on your target audience. I haven't checked the version, but we're stuck on an older IE at work. Possibly 9, but I think it's 8. No choice in the matter.
Does someone have a gun the CEO's head stopping him? He could tell everyone to install latest Chrome or Firefox and have everyone upgraded in 10 minutes.
These decisions typically have much wider reaching ramifications, like an existing tech stack which relies on some quirk of IE6, or activeX. Yes, it's simple to upgrade a browser, but the potential impact must be evaluated, and this takes time (more than 10 mins to install a new browser).
I think this really depends on your target market. Big corps such as banks are typically slow to move and I know of at least one which is only just moving from IE6 now.
