Ask HN: What is your policy on IE6 development?

I think that you have to look at the client's visitor profile before you make this decision. If a client's site sees 10%+ IE 6 visitors (and many of mine do), you can't simply say we're not going to support this technology.

Often I tell them that we will review the site in IE 6 and make sure that it's "functional" but we don't guarantee it will be pixel perfect unless they pay for the extra work.

Also I don't like the IE banner / alert option. Maybe you want to try to proselytize the death of IE6, but I don't think it's appropriate to force your paying clients to do the same.

This is the same policy we keep at my firm. We support IE6 on all sites, I don't see it as an option.

Part of being an expert is making things look easy. IE6 is frustrating, but its our problem.

The exception is web applications that aren't public facing. In those cases I think its reasonable to specify that the app will be run in a specific browser, and to charge more if it needs to run in many.

Being an expert is about knowing how to do something not about making things look easy. In fact, things that aren't easy take more time so the last thing you want to do is make them look easy because then your customers will have unreal expectations as far as how fast the work can be done and how much they should be charged.

There is a legitimate argument for charging more for IE6 compatibility and I don't see what's wrong with telling a customer that. If it were me I'd offer a basic option for free (no graphics with a link suggesting the user upgrade their browser) and then offer full compatibility for an additional charge.

That's fine, definitely two ways to look at the problem. We charge a higher rate than some shops, because I think our work is worth it, and it includes things like IE6 compatibility. But these are just differentiation points, and there's room in the market for both approaches.

Regarding making things look easy, I've always enjoyed Law 30 of the 48 Laws of Power:

Make your Accomplishments Seem Effortless: Your actions must seem natural and executed with ease. All the toil and practice that go into them, and also all the clever tricks, must be concealed. When you act, act effortlessly, as if you could do much more. Avoid the temptation of revealing how hard you work - it only raises questions.

To each his own.

I think you are spot on. But maybe you should not look at the visitor profile but at the revenue profile. Maybe the typical IE6 user has a higher probability of clicking on your client's advertising.

After all, if your gain from supporting IE6 is higher than the cost of supporting it, it makes no sense to abandon IE6.

This is a good point. I work for a non-profit and it's my job to maintain the web site. We notice users who click on our banner ads the most are IE6 users and AOL users (who use a newer browser but who you can identify by the IP).

The less savvy users are the more likely they are to click on ads in my experience

After all, if your gain from supporting IE6 is higher than the cost of supporting it, it makes no sense to abandon IE6.

Although is you abandon full support in IE6 and users want to stay and so use another browser then you've kept their continued custom and saved your extra dev costs. That is, it's not a given that removing full support will impact your profit.

Also, the improved design that could be purchased with the extra dev costs could net you more customers using a browser made less than 8 years ago or indeed convert more of the same.

People can have a non-profit motive too.

I simply don't cater for IE6 anymore. Not for contract work or my own sites (IE6 use for my sites hovers around 3% and it wasn't until it hit 10% where I decided to not care for the browser). Hell, even my corporate dayjob is deprecating IE6 development (we're in the health insurance industry).

If I'm doing contract work I usually spit it out early that I won't be catering for IE6. It is up to them to figure out whether the lack of an IE6-friendly frontend would harm them or not (I deal mostly with the same clients I've had for years, so they trust their metrics and me by now).

It hasn't happened yet, but if a client demanded I had to cater a project for IE6, I won't be afraid of firing them.

Well, #1 and #2 aren't mutually exclusive.

In our experience, IE6 users are simply unwilling -- or more to the point, unable -- to upgrade. The people who come to mind are parents whose kids have moved away, office workers, unfortunate souls in school systems or other non-profits that don't have a decent IT budget, etc.

My shop leans towards #1 and will include IE6 "compatibility" separately. This works well for our clients since IE6 is rarely being used by the people that our clients' sites are built for.

Unless we're explicitly asked not to, we use a banner-style alert as a general practice in order to encourage upgrades. We use IE6Update:

  http://ie6update.com/

On a side note, Microsoft's "Browser for the Better" campaign was actually pretty helpful because we could say that upgrading to IE8 was feeding hungry families:

  http://browserforthebetter.com/

The solution is very simple. Pricing.

Give your client two prices: one with IE6 support, the other without. Give honest prices based on the genuine overhead of working with IE 6.

If the client is willing to pay for IE6 support, fine, do the work you are being paid for. Let the cash decide the future of IE6

...but this is another example of developers making decisions that they should be making. For example, a software developer that decides to build a system with full database independance (hibernate for example) COULD be doing the right thing, but could simply be driving up project costs for requirements that never existed.

Developers - DONT make these decisions, price the options and ask the client.

The problem with this approach is that it opens up a good angle for your competitors to attack you on compatibility grounds. They can tell your client that they don't charge anything extra for IE6 compatibility, leaving you explaining that either they are charging them but that it's a hidden cost, or that their sites are so crappy and outdated that they work fine in IE6 from the get-go.

Guestlist (http://www.guestlistapp.com) does not support IE6 even in the slightest (except for the public page where you buy tickets, that works in IE6). We support IE7 and higher, and that will probably stay that way for the foreseeable future.

As for contract work, I do whatever the client asks. I do let them know that IE6 support will probably cost them an additional 20% in my time, and it will not be identical to modern browsers. It will work and that is about it. If they want to spend the money it is up to them. They do know their customer base better than I do.

Only 20%?

Wow we charge more like 75-125% more for that browser

This would be HTML + CSS only. I don't do Javascript and if I did, then yes, it would be a lot more.

Ahh, jQuery helps, but still, IE6 burns.

Let your competitors eat the cost of IE6 dev then

Always charge as LEAST as much as is required to make you either not care or actively want the client to buy a feature

So if IE6 makes the site 2x as difficult, charge them for the site using modern browsers, and line item (basic IE6 testing for at least limited functionality) at whatever price makes you not care.

Charge for them to maintain an outdated site, charge a lot. Then you won't care if they want IE6 support or not

We're a product company, so it's a slightly different stance. However, we made a conscious decision early on not to support IE6 for the following reasons:

1. Our app is very interactive, and we didn't want to have to deal with incompatibilities.

2. Our target market is (primarily) software developers working for small companies. The only people that use IE6 are people who don't know any better or who are forced to do so, and there are few of those in our target market.

We know that we're probably losing some sales due to our lack of IE6 support, but for the reasons stated above, we're going on the assumption that the additional sales wouldn't offset the cost of development and testing.

I think for a consulting company, it depends on the project. If you're writing something for general consumption or enterprisey software shops, it's much more important (unfortunately) to support IE6.

For me it just depends upon the type of site I'm working on.

In these competitive times you need to support as close to 100% of the sites user base as possible. If the site drops 10-12% to a rival because it lacks ie6 support thats not going to go down well at all.

I do like the car analogy, but if you have a 25 year old car & you can't get a newer one, it's rather harsh. You can get away with it as long as the percentage is low enough.

On backends (cms/admin areas etc) then this is were you can specify what browser can be used and ignore ie6 as long as you can explain why.

For my work projects ie6 is a must as it still registers around 10% of visitors. Personal projects tend to be more techy with virtually 0% ie6, so no need to support it.

I look forward to when I can ignore ie6 completely!

I think I'm in the minority, but I will fully support IE6 until it's less than about 5% of traffic to a website I'm working on. Once you know IE6's quirks, it's not that hard to work around them. I think the benefit of an additional 10% or 15% or web traffic outweighs a few extra hours of development. Again, it sounds like most people disagree, but I developed a somewhat complex ajax app back when there was only Firefox 2 and IE6 and it wasn't that bad. I'm not in charge of a big design studio, but that's my policy for side projects and part-time clients that I work for. They shouldn't have to care about IE6. It should just work for them.

For my in-the-works site, I'm going with displaying a notification to IE6 users notifying them that the site may not work as intended on their browser and asking them to upgrade to Firefox/Chrome/Safari/IE8.

We're down to 37% IE these days. Only 8% are IE6, for a total of under 5% IE6 usage.

I don't really know what the difference is between our site and the sites that say they have 65% IE still... it's not at all oriented towards techy type people. We have 45% FF and 15% Safari.

It's a significant enough number of people to test new features with IE6, but not enough to go wild trying to make it perfect. If I was developing sites for clients, I'd go for option #1.

NN 4.x stayed around forever BECAUSE we supported it, and IE6 will do the same. We have met the enemy, and they are us.

Maybe another way of looking at the problem: I think a bigger challenge would be supporting mobile devices.

Whatever your site does, you might need to also do it in a stripped down no-frills way that will render well on a tiny screen.

If you can manage that, IE6 support should be no problem. Worst case, push IE6 users to your mobile site. :)

If your website will have a tilt to "Enterprise Customers" (worker bees in Fortune 500s) in any way then you must support IE 6. You have no option.

Otherwise forget it for now. "Early adopters" are on the cutting edge (browser wise) anyway. You can always accommodate IE 6 bugs later on if the site "makes it".

The most professional option, in my opinion, is to make sure your designs are at least basically functional in IE6, but put making IE6 a design priority as a separate line item with it's own cost. Doing extra work deserves extra billing, and you owe it to your clients to give them a choice.

I will create a site that is designed for newer browsers. I am not going to forbid IE6 usage like I have seen other sites do. "Your browser is not welcome here, please upgrade." Instead, if IE6 works, it works. If not, I am not going to worry about it.

I agree that without a very frank up-front discussion about IE6 limitations, you'll find yourself redoing PNGs as GIFs and writing some conditional CSS hacks the night of launch when the client's sister boots up her Compaq to check out the cool new site.

I operate an English language learning site, and we have one demographic that has a staggering amount of IE6 usage: 66% of our Chinese user base is using IE6. Needless to say, we continue to support IE6.

I generally just make sure that the website is functional in IE6. If an element is bumped/ aligned or distorted in some way (and is still usable) it doesn't get fixed.

I am deprecating it quite aggressively now.

every advertiser i've worked with requires IE6 compatibility. so, theres that to keep in mind.

Not supported.

My policy for IE6 is: don't.

DO NOT WANT!