“But The Client Wants IE 6 Support” | Hacker News Reader

“But The Client Wants IE 6 Support”

smashingmagazine.com

187 pointsbillpg109 comments

I'm a web developer, whenever a client asks me about IE6 support, i enthusiastically tell them i'll happily support it. I then show them usage statistics for IE6 and tell them IE6 support costs a flat fee of £10,000 and let them think about it. No client has ever taken very long to decide its not worth it and none have ever taken me up on it, although i'm sure 1 or 2 have asked a different developer to make them an IE6 version, but thats fine with me.

this is the correct way to approach the problem, because it puts the client in control. if you tell them flat out that you don't support IE6, they will see that as a deficiency in you or your agency. you need to let the client feel like they are making the decision to not support IE6.

We take a slightly different approach: we offer pricing for both full IE6 support (completely matching the mockup and functionality of other browsers) and doing enough work to make sure IE6 can be used although the styling will not be perfect.

Like Tam's approach the key is to give the client options so they can make the call based on their budget and business needs.

I agree with this. Although I'm young and naive and I rather not pick up a client (because I have that luxury) that wants IE6 support.

I haven't had a request for IE6 support in years, but if I got one I would say something like:

"No, I will not do IE6 support for you. You've trusted me to solve your problem in the best and most efficient way, and supporting IE6 would be a waste of your money (and my time)."

Let me tell you about the UK National Health Service. 1 million employees, they are literally saving people's lives every day, and all they are allowed to use is IE6.

What your statement shows is not a understanding of how to solve the problem for such a client, but a rather a complete lack of understanding. Check the audience metrics, check the real problem being solved, and leave the grandstanding aside.

And for the record, I hate IE (6 or otherwise) and never want to support it. But I am not going let me personal dislike for a technology get in the way of helping a client solve a really important problem.

I'm so glad I've never had a client like that here in Australia. Everyone seems to be OK with the idea of running multiple browsers. We do have clients that use IE6 but we've always managed to convince them to install Chrome or Firefox on their desktops so they can use our software. Some users end up using multiple browsers but that's ok.

I guess we're fortunate in the sense that our software is important enough and expensive enough for top management to make IT allow a new browser just for us.

At the UK NHS, who exactly is the CIO, who exactly is mandating IE6 and only IE6? Can't that person be reasoned with? Overridden for the good of the NSH? Fired even? Just saying 'oh they're only allowed IE6' is giving up too early.

Some historical background, which includes plenty of personal bias:

Governments, for all their posturing, love dealing with oligopolies, or even better, monopolies. They feel safer in making purchases from them. So around 2002, the UK government decided IT was so important, it was going to be done centrally for the whole NHS. They gave the job to Richard Granger, who then proceeded to divide England into five regions, each of which would have one vendor responsible for all software in the NHS for that region.

As part of that deal, they licensed MS Windows and Office. At that point, it became more expensive to use open source software than Microsoft's software. Then they standardised on IE6, and many vendors began using ActiveX.

When I ask departments to upgrade, they say they are going to do so, but first they need every single vendor to have signed off that their software works in the new browser. This will take a really long time.

Some will install a new browser for us, but this will actually result in a worse outcome for the clinicians because they have to use IE6 for most of their existing browsing, and then the other browser for us. They do not understand the point or the difference, and new browsers mean a different user interface, and they want to get on with treating patients rather than choosing between browsers.

The epilogue is that the UK government's national program wasted 12 billion pounds (about 20 billion US, ie the same amount the US government is also identically wasting to subsidise electronic health record purchases). And Richard Granger left the UK in disgrace.

He is now a consultant in... Australia :)

Would they be willing to use Chrome Frame? It seems like the ideal way to target IE6-only customers without "installing" another browser. That way they can use one browser for all their intranet apps.

Hah, after that I expected that he'd be working for Microsoft directly. The way you tell it, he sounds like their best sales guy.

Except certain Australian Federal departments with 70,000+ staff who are stuck on IE6 because their IT boffins only want to use 'secure' web browsers and IT policy doesn't give them any choice in the tools they use.

Can't speak for the internals of that organisation, but the main reason I've come across for corporates still being on IE6 is the very real cost of migrating.

Many legacy web applications were built to work on IE6, and changing to a more modern browser can subtly, or completely in some cases, break these critical apps. It may be a pain for users when their favourite social media site looks rubbish or doesn't even run in IE6, but it's nothing compared to the finance department not being able to run their end of year reports because one screen in their GL system doesn't render properly in Firefox.

And it costs money firstly to test every single function of every single app that your business uses, and secondly to fix those places where it breaks (especially when it's an external vendor's application). That time and money can often be better spent doing one of the thousand other things that are making demands on the IT budget. Most companies are slowly making this transition, but unfortunately it's not just a case of saying "This browser it better/more secure than that one. Let's upgrade".

No see my point is that there is ALWAYS someone above the 'IT boffins' that can force changes in policy like that. You don't fight with the 'IT boffins', you go right to the top. If the CIO or CEO or board of directors decide to use your software, they'll change whatever internal policies they need to make it happen.

The problem is this is piss poor client relations. When you make a declaration like this you seem arrogant. Effectively you are saying that your knowledge of web development trumps any concerns they may have to the point that you don't even need to know about it.

Instead you have to take the time to understand why they are requesting it and then communicate the tradeoffs to them at the appropriate technical level for their expertise.

I'm usually less brash and assertive when I'm not making inebriated offhand comments on the internet. Honesty and confidence can go a long way in building trust if backed with demonstrated knowledge, though.

In my experience, clients are usually pretty good at providing rationale for their requests, so my implied scenario is a bit of hyperbole. Often when the topic of legacy browsers comes up, it's because the client has been in the process of having a website made before, and is relieved to discover that the technology has moved forward since then.

Unfortunately, some (very large) corporations still use IE6 across their organizations and require you to support it.

Perhaps I am naive but this conversation seems easy to resolve with just a few words. Can anyone tell me how a client would respond to this:

You can have a site with newer, nicer features, or you can have a site that supports older browsers. If you want both then you are commissioning two distinct websites. If you want usage statistics for old browsers I have them, and if you want to be economical I can produce the second website for older browsers at a reduced cost with reduced flair, after all only X% of people would see that site and it would still look reasonable, just not amazing. What would you like to do?

To me that's lying to them.

While supporting IE6 is onerous extra work, it's still work that can be done and has little impact on the final product.

Take the other position, that 'Not supporting IE6 will mean I'll charge you 70% of my original quote, only x% of people use it in your target market' or better add it as an extra in the quote in the first place as the article suggests.

Then it's up to them, it's a risk assessment. But be up front in your quotes with it, not an after the quote is won 'I don't know how to do it!'.

The article's incredibly OTT on how much adding IE6 support costs, but I certainly would never go down the route of lying to a customer about it. If he charges double it's probably because he hates doing it and fair game to him.

>and has little impact on the final product.

Depends on the tools you use, etc. It very well may double the time required (as IE 6 support requires tons of testing of all the features, and many software project have huge portions of the projects being testing, not development).

Additionally, you tend to need more experienced engineers, IE6 support is aged out of the current crop of grads.

I think the second legacy browser site is a great solution. Additionally, I think if you're saying "has little impact on the final product", you're not working with clients who open up psds and count pixels on you. Some people are.

I think the article is realistic in terms of IE6 support costs if you're not just building a simple website.

Trying to build a responsive, dynamic webapp, that is identical in IE6 to a modern browser can easily add a lot of time. Just as an example, producing all the graphics for rounded corners, rather than being able to do them in CSS.

I've found that people don't really notice or care when IE shows them square corners.

The thing about IE6/7 support is that it's a complete known quantity at this point. The problems & workarounds are well documented. There's tons of experienced HTML coders out there that know IE backwards and forwards. CSS frameworks like OOCSS include built-in workarounds.

So if IE support doubles the cost, it must either be something very complex, or a real nickle-and-dime project. (Either that or the webdev doesn't really what they're doing.)

That's not just an IE6 problem. IE8 has full CSS 2.1 support but can't do css3 rounded corners, opacity, or shadows. You can sometimes fake it with -ms-filter and CSS expressions (htc files) but those work just as well on 6 and 7.

That said, I'm happy to drop IE 6 where not required by the client - 8 may not have html5 or CSS 3 features, but at least it's relatively bug free.

Rounded corners are already out fashioned. If a client want IE6 support I take it as a good challenge, and do something nice under this heavy constraint. I'll tell them upfront that they won't have cheap drag and drop and that new stuff.

Minor nitpick, but the OP is a she - Lea Verou. To be fair, this feels like such a male-dominated sector that I only picked you up because I had unexpectedly noticed Lea's picture at the bottom of the article, and I suspect you didn't feel the need to check!

If we had genderless pronouns we wouldn't have to care about details like that which don't materially affect the article or its criticisms.

Having said that, fields with a more even gender ratio have had to deal with this for a while - what do they do?

English does have a genderless singular pronoun: They

Just because a book said it was preferred in 1848 to use he when sex is indeterminate (then went on to use both singular he and singular they) doesn't mean that's true today.

The language always used both. Why stop today? Might feel a bit odd at first if you're not used to it, but it's not a bit different than how singular "you" is used ("You are coming today?" uses the plural form of "be").

The book itself used both forms! Shows you that English teachers may have been overly strict with you in the past.

Some more on the topic:

http://en.wikipedia.org/wiki/Singular_they

"You" started out plural. "Thou" was the singular version. Then "you" started being a form of respect, similar to the royal "we" or the Spanish "usted". Finally, we all became "you"s, sort of how we all became "sirs and madams" and "ladies and gentlemen" when those were originally a select class.

Anyway, "They" has been undergoing this process for a few centuries now, and I concur that it's time to embrace it.

>Thackeray's use of both forms demonstrates. "The alternative to the masculine generic with the longest and most distinguished history in English is the third-person plural pronoun. Recognized writers have used they, them, themselves, and their to refer to singular nouns such as one, a person, an individual, and each since the 1300s."

I don't really think it's an evolution, as much as the mid1800's tendency to over classify and striate things going away.

It surprises me that this has to be highlighted. I wouldn't genuinely know if it's a difference between British and American English but I've never encountered a British trend of using gender specific pronouns - or the unilateral usage of feminine pronouns in favour of masculine ones - in place of the neutral forms.

It is one of those niggles that irks my pedantic self though; particularly the politically correct change to using 'she' instead of 'he' (instead of the perfectly neutral 'they', and even 'one').

Does the singular You bother you? Dost thou pine for Thou?

I'm not sure what relevance that has to my dislike of unnecessarily gender specific pronouns.

I misunderstood the antecedent to your "this"

Thank you for this. I had always (only slightly, though, in a grammar nazi way) looked down on people who used "they" for a singular pronoun because I thought that many do so just out of carelessness. But glad there is a formalized reasoning for it.

I've always used "he/she". But that's not quite accurate (or elegant) either.

I tend to just be lazy and put "the OP" instead of he/she/it, although you could nitpick me on that as well since the author of an article posted to HN is often not the "original poster" in question.

"OP" roughly meaning: "walk up the comment tree, including comments and authors, all the way to the article and whoever submitted it, stop at the one I'm talking about, which should be pretty clear from the context of what I'm saying".

At least in theory, "OP" means "original poster", making it unambiguous. Though sometimes people mean the root of a comment thread, rather than the article.

Not to navigate off-topic, but merely as an aside, perhaps we can solicit the W3C to weigh in with guidance for <op>.

http://en.wikipedia.org/wiki/Gender-neutral_pronoun#Summary

This point of view stands so long as the client hasn't heard something else from another vendor. New clients (the ones you'd have to explain this to) are often trying to avoid Caveat emptor situations. These IE development conversations are rarely one-way, regardless of how you sell it.

If they want fancy new features, just use Modernizr with a polyfill. That generally seems like a better solution for the dwindling population of legacy browser users than creating a completely different codebase.

You are still creating another site even when using polyfills. The surface area for display bugs in a polyfill is > 0.

> How many of us actually charge 30-100% extra for this work? I haven’t heard of many who do.

Actually I see the exact opposite around me: developers billing by the hour and warning that IE 6 (or later) will generate extra work (and costs money).

Most people I know just have a look at their site stats and see what is worth it.

Very much this, we informed the client that it was going to cost them more, and we itemized time spent fixing IE6 specific bugs and layout issues so we could offer them an idea of how much it was costing them. They needed this kind of data in order to convince their internal IT to migrate most of their user base off of IE6.

I usually take a much different approach with this issue. I had to support IE 6 for years after anyone could reasonably have been asked to do that. I did it because our customers used it a lot and asking them to upgrade wasn't in the cards for us. We could have lost 30% of our business or just not do all the cool things we wanted to do.

I realize it's not cool but we chose to just not do all those things. IE 6 does support a lot of what you need to do and once you figure that out it becomes easier to make it work without having to go back and fix it later. I'm pretty good at it now so I don't really complain if someone asks me to make something work in IE 6. It's not a huge deal if you know what the issues are.

IE7 only came out 5 years ago, IE8 2 years ago.

Making sure IE6 was supported was still commonplace even last year.

You're not as unique a snowflake as you think you are, we've all been having to do it. It's only becoming viable now to suggest otherwise.

"You're not as unique a snowflake as you think you are"

That's unnecessary. The point, without the invective, please.

Unfortunately IE6 is still very popular in the corporate world of Europe. I'm currently running an internal websurvey for a large car rental company and looking at this weeks http logs shows that IE6 is still getting 61% :(

my logs (for a file sharing solution for businesses in France) show 5.5% for IE6. Maybe because there are more early adopters at the moment? I for sure hope I won't need to spend hundreds of hours supporting IE6...

hi Tommy! My guess is that most "IE6 only" corporations would not allow using a file-sharing site anyway :)

I still maintain one site which is around 70% IE6 (internal stuff in insurance). But one day it will change all at once, for political reasons :)

I'm not sure you will see more than what you currently see on the long run...

oh hi Thibaut :-) good point, I'm not targeting big corporations anyways. I guess the problem could be that even 5% could be a problem, as people sharing files would complain their 5% customers cannot see the files... well, we'll see, no complaints so far ;-)

Large car rental companies are not exactly cutting edge on IT. They tend to have a lot of small locations that's not exactly easy to start upgrading because of the cost of sending people out to do it all.

My local branches look like they had their last refurb in the late 90s for example.

One example does not a sample make.

... They tend to have a lot of small locations ...

These aren't the rental outlets but their main business offices in UK, France, Germany, Spain, Italy, Netherlands, Portugal & Bulgaria!

NB. The Bulgarian results shows 71% using IE6 in their main office(s).

One example does not a sample make

At no point did I say it was relevant outside my sample :)

It is not just car rental companies. Most of the UK's major banks use IE6 still.

This is because most of their intranet apps will only work in IE6, and they have the size to demand that their supplies support IE6 too. Corporate IT policy won't allow browsers apart from IE.

If only Microsoft allowed side-by-side installs of IE6 and a IE8 it might be a bit more bearable (given that IE9 is not available on XP)

> This is because most of their intranet apps will only work in IE6

How did it ever happen that IE6 ended up in such a long lasting and dominating position? Was it (I am just guessing here) the coincidence of tho things: (1) IE6 just happened to be the most recent Microsoft browser around the time when most of bigcorps initially build their inter/intranet apps, and (2) because this was the first wave of building such things, the web coders did not bother to think about a future when IE6 is no longer a modern browser?

Well, part of it was that there weren't other browsers to worry about. At the time there was no Chrome, no Safari (no one used Macs in a corporate environment if there was), and no Firefox. All that you had was IE and Netscape, and Netscape was on kind of shaky ground at that point.

Those two kind of fought with each other by introducing new features that made web pages incompatible with the others. Also, even then corporations weren't really keen on letting you use some other browser, so IE was the go-to since it was already installed. Since the market was so slim, and seemed to be increasingly vanishing, it seemed easier to justify going with IE's extensions.

If Microsoft had kept updating IE and hadn't let it linger for years, we'd probably still be in the situation where they're in control of the de-facto standard. It's only because they sat with their thumb up their ass that things like Firefox and Chrome were able to get footholds and start disrupting the situation. Already being present and practically being named "The Internet" are really big adoption factors.

In addition to what others have said, don't forget that ActiveX also played a strong role. ActiveX was pretty widely used in business-oriented web applications, which hindered business adoption of firefox even among those companies or individuals who wanted to move to it.

There was also a general distrust of open source software during the time that firefox was gaining adoption. With so much malware being spread via the browser (IE6), there was a fear that firefox's open source development model and perceived immaturity were a security and support risk.

It's because IE6 was the last IE release before the end of the first browser war. Previously, Microsoft had been releasing browsers every year, but they stopped after Netscape died. It wasn't until Firefox started to gain market share, almost 5 years later, that IE7 came out. In the intervening years, IE6 had the chance to gain about 99% of the market.

(Not sure if all my figures are accurate, but that's the gist of things).

It happened because short-sighted and sloppy authors failed to use progressive enhancement over a pure HTML version that worked on every browser ever to exist. Instead they picked one browser they liked, wrote bogus code that happened to work on it, and declared "if you aren't running what I'm running, fuck you". In five or ten years we're going to be hearing the same kind of whining about how the Chrome and Firefox holdouts are holding back Web 5.0, blaming user agents when it's their own bad authoring at fault.

I think it's because IE6 is the default browser for Windows XP Installs. I think even with the later Windows XP Service Packs, they still baselined to IE6.

That and there was a big gap in time between IE6 and IE7. Didn't Microsoft once announce that IE6 would be their last browser, as that's all the world needs?

This is because most of their intranet apps will only work in IE6, and they have the size to demand that their supplies support IE6 too. Corporate IT policy won't allow browsers apart from IE.

And this also seems to be the common case when I've done onsite freelancing gigs in non banking (SMB) UK companies over recent times :(

The article mentioned "graceful degradation" but what is really described should coming from the the other direction as "progressive enhancement". As the author mentions, only enable each version of scripts and styles to browsers that support them. For those that don't, the content remains.

Here's an old article that gets the message across in a client-friendly way: http://www.alistapart.com/articles/understandingprogressivee...

Let me preface this with saying I long for the day IE6 is completely irrelevant.

The fact is, when doing client work I have a responsibility to design the site in a way that when their customer visits they get the information they need. If the website is broken, for whatever reason, that looks badly on my client and on me.

When I do a site I code and design for modern browsers. However, I make sure even IE6 users can get the information they need or order products or whatever the case is with the website.

True, the IE6 users are small in number, however in some cases they can make up an older or important demographic to a client.

It all depends on the kind of site you are building and the technologies involved. On some sites it may be ok to deliver a "less beautiful" web but with all the contents, in others it won't. Oftentimes nowadays you'll face a complete rewrite.

In some recent projects I even worked hard to convince the client that it wasn't worth supporting IE8. It's not only that it loads faster with modern techniques (as CSS3 allows awesome effects that would need images/javascript with CSS2), but some functionality built around HTML5 would need a COMPLETE rewrite in Flash.

Of course that means that if they insist on supporting older browsers not only that would mean a 30%-100% increase in price but new features the want afterwards will be more expensive as well (because you're supporting two different projects).

This seems like a relatively simple business proposition to a client. "X% of users on the internet today use IE6. I will be happy to provide pixel perfect support for Y dollars." Adjust Y up and down based on your ability and desire to continue to support web applications that use IE6.

This article is great, but it works on the premise that IE6 support is a purely rational discussion by two knowledgeable parties before a contract is signed.

In reality, you may develop a website, then the client is visiting their uncle bob one weekend who has an old computer with IE6 lying around, client + uncle bob open up the website to have a look, and it is "broken".

Client then sends you a hate mail about broken contracts.

You reply saying that you discussed IE6 lameness before you started the work.

Client says you were talking jargon, client didn't understand, and you need to fix their website so it loads on their uncle bob's computer otherwise they are suing you for not delivering a working website.

Good luck with that :)

As long as the contract clearly mentions that you are not supporting IE6 I wont mind client suing me.

Make sure you have the prevailing party's legal costs as part of the bargain in the eventuality of suit as well: Otherwise wining the lawsuit can be Pyrrhic

Yes , this is a strange phenomenon. People are often more interested in what works on their computers or those of friends,family than they are than looking at it in terms of supporting % of potential customers.

You can bet your ass that if your client suddenly decided to start using lynx exclusively for his browsing that he would suddenly become very stressed that his flash intro doesn't work :)

Reminds me of doing tech support for a web hosting company a few years ago.

Customer: "Our Website is down!" Me: "Ok let me check.... Seems like it is up too me.." Customer: "No it's definately down!"

It then transpires that their internet connection is not working at all and that they already knew this to be the case.

They were still concerned about how embarassing for them it was that their website was down because they had just handed out new business cards with the new URL on them.

And people think that us techies are the borderline autistic ones..

That sounds like an asshole client. Just make sure you have it specified in writing before you start a job the browser that are and aren't provided. Then if they want to talk to a lawyer, they can try.

Best not to take any further work from that client again.

Take that to it's logical conclusion, and you should still be supporting IE5 for mac.

Except it usually isn't "uncle bob", it's their biggest customer who still uses IE6, they're losing business, emergency, etc.

If "IE6 Lameness" was indeed discussed prior to starting work, well that's just bad planning on the clients part. You just want to be sure IE6 support wasn't implied in the contract...

We explicitly don't support IE on our heavy JavaScript web app.

Users get an alert saying, "Use Firefox or Chrome. If you want to use IE and are upset that you can't, let us know."

Nobody has said anything about it.

I think so much depends on your audience. Our audience is writers using their own computer. They're happy to use Firefox and Chrome.

I only see IE6 support as a problem for people on corporate machines who don't control their own software.

Who are these clients and why are you not firing them?

You run a business, they run a business, and everyone is trying to make money. How much money do you think an IE6 user has? the browser is like a decade old. It's why we target iPhone and not Android in the mobile world. ( because iPhone users out spend android users. )

Tell your client the truth, tell them that you keep up to date on the latest trends and if they want to use someone else let them, and then tell them exactly what your work will do.

I would not even offer to support IE6 anymore. Maybe IE7. There are a few reasons.

#1 you can't download IE6 legally any more

#2 you can't install it legally on linux or mac systems to test.

#3 IE7 in quirks mode does not work exactly the same as IE6.

#4 web development for legacy browsers is less about building things, and more about working around problems. and i just don't find that fun.

It is your duty as a web professional to turn down this work. Microsoft does not even support his browser anymore.

Not sure I quite agree here. I know somebody who runs a financial website and IE6 visitors accounted for ~20% of their visitors but had a significantly higher conversion rate for IE6 users than others (no idea why).

#1 & 2 , Microsoft provide a free VM image of windows XP SP2 + IE6 specifically for IE6 testing. It's timebombed though so you do have to occasionally download a new one.

http://www.microsoft.com/download/en/details.aspx?displaylan...

#4 Sometimes work isn't fun, if your in the business to make some money you've got to deal with that now and again.

It's your "duty" to build something that fits the clients requirements relative to the amount they pay you for it unless that involves something illegal or unethical and supporting IE6 is neither.

"Microsoft does not even support his browser anymore."

http://arstechnica.com/microsoft/news/2009/08/microsoft-drop...

Which documents Microsoft's commitment to support IE6 until April 2014.

"How much money do you think an IE6 user has? the browser is like a decade old. It's why we target iPhone and not Android in the mobile world. ( because iPhone users out spend android users. )"

So your solution to the IE6 problem is to ignore it and to repeat the same mistake again for developers to have to deal with again in the future?

We have the IE6 problem because that's where the money was back in it's day. It was the winner of the browser war, and thus became the interface to the substantial source of profit. Very much like your perception of the iPhone.

What you need to consider is not just what's profitable right now, but what will be profitable tomorrow, next week, next month, next year, and next decade. IE6 won the browser war; the iPhone browser is nowhere near that dominant in the mobile world.

Agreed. We did the same with our web dev firm (cut support for old browsers), and nobody seemed to notice. Anyone who "needed" support for IE6 we would have happily given them the names of a few other companies in the area that can help them.

We also were in the very good position of being able to turn down work, though, so it's easy to tell people off. I think it depends on your situation.

Idealistically though, it seems like a huge step backwards to bother supporting IE6. The more people that do, the worse the internet becomes as a whole because of it.

when i was running the site http://unity3d.com less than 1% of our visitors could not properly render an html5 doctype.

Question is, what % of users have the unity plugin installed?

I think it's quite neat that unity let's you embed your game into a web page but I often wonder how many users change their minds once posed with the requirement of a plugin install since this is exactly the kind of thing that IT savvy people advise their less technical friends not to do!

If we choose to make a website pixel-perfect in Internet Explorer 6 to 8, then we are doing up to 100% more work.

Every clients/users should check this once : http://www.ie6countdown.com

As much as I like that site, I can't link people to it because as a Microsoft site it just wants to move people to newer versions of IE, rather than to a decent browser. (Any decent browser, really; I don't care which, just not IE.)

We tried to drop IE6 support for an ecommerce client recently. But looking at analytics, IE6 customers are still doing 5 figures a month in revenue. The client just can't risk the potential lack of conversion in a site with a 'graceful degradation'. The testing that's gone into making it convert as it has must apply across all the major browsers from which they get traffic.

Shame, it's a complete pig to develop for.

edit: They must look and behave more or less the same. Personally I have no qualms about the IE users not seeing rounded corners on some boxes. As long as they're in the same place and the same colour, etc.

Previously submitted here: http://news.ycombinator.com/item?id=3192237

What I find curious is that the de-duper apparently doesn't check for case differences in the URL encoded characters.

http://www.smashingmagazine.com/2011/11/03/%E2%80%9Cbut-the-...

http://www.smashingmagazine.com/2011/11/03/%e2%80%9cbut-the-...

I don't offer IE6 support, offering IE6 support keeps the browser alive, and it has to be killed collectively.

I'm guessing that an example of a client that would ask for IE6 support would be Car Mechanic shops that have CRT monitors and super old computers. And probably a corporation that can't upgrade all people at once and their secretaries don't know what a browser is.

It's interesting that despite all the investment that Google is making in Plus and their desire for a foothold in this market, they flat out refuse IE6: http://i.imgur.com/ONRK4.png

Good for them. There is a trade off to be made, better/quicker features delivered at the expense of supporting an obsolete browser.

I love how it says "Google+ no longer supports your browser" when (as far as I know) it never supported it.

Isn't this just like about anything else? There are still plenty of people who use VCR's. There is still good money to make for people who are willing to support older technology. If you don't do it, someone else will.

Except we stopped making new standalone VCRs years ago, and new movies haven't been published on tape for years as well.

The interesting part of the IE6 debacle isn't the fact that a legacy program is sticking around longer than we had hoped, it's how ridiculously strong the longevity of it is. Most likely IE6 marketshare will stay in the double digits in many countries long after the world has completely forgotten about VCRs.

If any client demands IE6.0 support then I demand 100% more. the reason is not just that involves near 100% more work but also the skills required to do that job will be worthless in future.

That's not unreasonable but the key thing is to break that out in the quote. They need to understand what that support is costing them to allow them to make the choice as to how much they need it.

Maybe I'm missing something, but recently there has been a lot of work in graceful degradation. Such examples include Modernizr (based upon yep-nope) which allow detection of features and asynchronous loading of polyfils for missing functionality. Missing css3 support for IE6-8 comes in a 10 second fix called CSS3PIE. To use it you add 1 css line to apply the htc to the necessary classes (or all with the * selector) and there you have curved corners in IE6-8 with no additional work.

Somehow I had managed to completely miss the term "polyfill" until just now.

Handy!

IE6 Support is currently handled here via a second, reduced functionality site as a standard practice (and charge accordingly).

IE6 support can be made reasonable for most apps as long the client understands that the IE6 support will be usable, decent looking, but not 100% the same look and experience as with more modern browsers. You just need to explain to the client that they don't want to blow their budget wrestling with nit picky white space inconsistencies.

"Even if they don’t care about accessibility, my responsibility is to make the website somewhat accessible." I'm unclear on who the legal requirements for accessibility fall on exactly so I may not understand properly... but I'd think this is a legal requirement to make your website accessible (at least in the US under Section... 508?)

Section 508 only applies to US government organisations and businesses/agencies providing services to those government organisations.

US legal requirements for public sites are probably covered in ADA (Americans with Disabilities Act), under the provision of reasonable accommodations. So far legal rulings are varied and vague, swinging one way then the other about whether ADA applies to websites (despite a DOJ opinion that they do/should).

Some states have their own legislation, like California's Unruh Civil Rights Act which was used against Amazon (Target settled and agreed to make their website accessible).

Deleted comment

>A line item like "IE6 support" is how to generate negative feedback from clients.

Not at all. The few clients I've had who've said anything at all have been overwhelmingly positive about knowing exactly how much IE6 support would cost vs not.

That's one of the reasons we started implementing it as a second, simpler site, much cheaper, allows that number to go down even more.

Fight opinion with facts.

Ask your customers if they got IE6 support as advice from someone.

Show how many people use it now and what that sliver of IE6 users now see with gracefully deprecating css libraries.

"Do what the client needs, not what they want."

Great advice I got once.

In the context of IE6 what's best for the client is the one that best serves their business objectives in regard to their customers. If their target market contains a significant-enough element that uses IE6 then what's best for the client is a website that supports those business objectives with that customer.

Then use chrome frame ;-)

"We have a responsibility to ourselves and to the Web to follow the principle of universality."

Not feeling that at all, so speak for yourself. If I want to target a specific browser or device, I do it when it makes sense. Screw universality. The users install what I tell them to or they don't get the privilege of using my site or product. (Do you think Steve Jobs would disagree?)

Given how much he trumpeted HTML5 and the nixing of Flash from its, yes, I think he would disagree seriously.

People begged Apple for the ability to create real apps. The original plan was to build HTML apps, but comparing a web app to a native app was no comparison at all (and it still isn't; I hate almost all web view"apps").

I would think so, if you are talking about websites at least rather than an entire computing platform.

The entire point of a website at it's core is to send and display pages of information, this isn't really a problem that should require specific versions of special software to solve.

With the exception of requiring quicktime for the videos the apple website has always rendered fine for me on pretty much any browser I have viewed it with.

That sounds like a great recipe to start justifying shaving off fractions of your market until you're down to only reaching 40%. Not saying that's what you're doing, but any kind of "Screw X, I will tell the users what they want" decisions have to be made carefully. It's a lot easier to think you've made a Steve Jobs kind of product than it is to actually make one.

This attitude leads to fragmentation which causes more headache e.g. the web browser market.

Reply on news.ycombinator.com