You may not think about it but by NOT supporting those 5% or telling them they have to upgrade (which is unbelievably easy) saves much, MUCH more than 25,000. Hell, my friends dev team wasted thrice that just adding support for a heavy JS web app.
They complain about the cost and later said it'd cost a 1/10 to just get their admins off their fat hairy arse and update their computers. If not that, you could even use Chrome frame!
Read the article. It pushes Progressive Enhancement, so if you are building an app that relies on JS your only choice is to either strip back the JS or ditch older browsers. My beef is with normal websites that don't support legacy browsers.
Regardless, you don't get a choice in what your users use, and more often than not they don't have a choice in what they can use or install. Chrome Frame is a good option, but the average user isn't going to jump through hoops to view your site and if the average marketing manager looked through a site the requirement for an external tool before a user can use the site would be the first issue flagged.
As I said above, it's not hard to get a typical website working in IE6-8.
How many times are we going to have the same argument?!?!?!?
There is always a choice. IT departments can, and do, install modern browsers on people's desktops. IE6 can run along side Chrome, for example.
The guy sitting in the cubicle doesn't have much say when the upgrade happens. When the guy in the corner office starts to feel the pain, a few phone calls are made and something gets done. If you keep pushing progressive enhancement then we can support IE7 and IE8 for another decade. Of course, the amount of Javascript that you can use on your site will be limited.
As many times as it takes for you to realise that you don't have power over everyone's IT department.
Yes, installing a browser is trivial, but people are at this offices to do work. The argument is that these people shouldn't be browsing the Internet in the first place.
The only thing keeping these dead browsers alive is legacy operating systems, and Microsoft needs to find a way to get people off of XP. If Microsoft can kill XP and get people on Windows 7 then IE6-8 are gone from the equation. This is the real battle, not whether you can be a lazy developer.
Everyone already realizes that. Now when you realize that we do realize then we can make some progress.
Once it becomes uncomfortable enough for enough people, or the right people, then IT departments will respond. That's how many departments work. Push change out as far as possible.
As I have already explained, there's an opportunity cost for supporting legacy. Companies like Apple are quick to drop legacy, and move forward. It works quite well for them. Google doesn't support less than IE8 for their apps (are they being lazy?). They claim to only support the last two browsers. It'll be interesting to see what happens when IE10 ships.
One thing that I think would be useful is to tell people that your site no longer "officially support" browser X before you actually stop supporting it.
They complain about the cost and later said it'd cost a 1/10 to just get their admins off their fat hairy arse and update their computers. If not that, you could even use Chrome frame!