With the life of IE6 coming rather slowly to an end (as discussed in my previous post), the talk of designers and developers dropping support for the browser is rife in blog posts and tweets all over the web.
Some pretty large sites have already got the ball rolling - possibly one of the biggest being FaceBook.
There is also a trend of developers, not dropping support entirely, but only providing it if their client specifically requests and pays for the work necessary to accommodate the browser.
So, where do I stand on IE6 support?
Like most decisions I make in my web development work - at both the planning and design stage - I make my choices from statistical analysis.
This analysis is from:
- the stats from the website in question (if already live and gathering statistical data)
- research into the stats and trends from similar websites
- general global stats (usually from a source such as www.w3schools.com/browsers/browsers_stats.asp)
The stats that can effect your design decisions are often those concerning: screen resolutions, operating systems / platforms (PC / MAC etc.), and - as we are discussing here - browser usage.
Let's have a quick ganders at the numbers ...
Well, according to w3cschools, IE6 usage throughout the web in general is at 17% (March 2009).
That, in my opinion, is a statistic you can not ignore as a professional web designer. Especially if the site is a new build with only general market research to go on.
This isn't about how rubbish the browser in question is. This isn't about how much of an inconvenience it is for the developer. It's a percentage that represents a significant number of people who have come to view and use a website. Would you lock out visitors if that 17% of users were using the Safari browser? Of course not, and when you consider that Safari users currently stand at 3.1% (March 2009), it further hits home what the responsible way of looking at this situation is.
I have designed and developed a number of e-Commerce websites, so I am well accustomed to dealing with the direct link between traffic and sales.
So, let's say that a site has sales of £30k in a year. When you consider that every visit could lead to a sale, that 17% IE6 visitor stat could represent £5100 in annual sales for your client. Not to be sniffed at. That's not an 'optional extra' in my mind.
This is why I feel it is up to the developer to decide whether IE6 support is necessary or not, it is not a decision for the client to be making, and the developer should put aside any opinions on the quality of the browser and go with the reality in which they are faced.
This site here what you is looking at now
You'll notice that this very site displays and functions near enough the same in IE6 as it does on faster, modern browsers. This is because, at the time I started my re-design (December 2007) my stats showed that 17% of visitors were using IE6, and even when I finally got around to launching the new site earlier this year (March 2009) stats showed IE6 users at 10% - still a significant number I think you'll agree.
Don't get me wrong, if you read my 'about this website' page you'll see that I - for the obvious reasons - don't recommend that visitors use IE6, and I provide links to the better alternatives. All developers should do this as part of educating users.
Also, would you say that my design, and its functionality, has suffered from my choice to cater for visitors using IE6? Hmm ...
C'mon people! Remember what 'Accessibility' actually means
I find it baffling when developers quote 'accessibility' as a reason for dropping support:
"IE6 is a rubbish browser in terms of accessibility and web standards, so that is another reason to drop support for it"
What?! Accessibility is about giving the best access to your site - for standard visitors, search engines, other sites and apps, and - of course - for those who may be impaired in some way.
Therefore, if you are purposefully making things difficult for 17% of users who have committed to coming to a site, you are falling at the first hurdle of accessibility.
Are these visitors of less value than others, because they happen to not fully understand the difference between browser types and versions? Would you stop a friend from coming in your house because their trainers aren't the latest Nikes, or the car they have driven to your house in doesn't have a passenger side airbag? No. You let them in (you perhaps make a suggestion about comfier trainers or safer cars). Why do you let them in? Because not to would be just plain snobbery, and would also make you a bit of a dick.
Charging 'extra' for IE6 support
Well, I already do this. I look at the brief, I look at statistics, I do my research, and if the conclusion of this is that a significant number of users will be accessing the site using IE6 (or any other possible method - perhaps the research shows that most users, in this case, will be be accessing the site via their 1998 mobile phone, pressing the buttons with their nose, whilst drunk, on a raft in the ocean, with cramp in their right thigh - my point is you deal with the reality of the situation you are faced) ...
Erm, where was I? Yes, you discover that a "significant number of users will be accessing the site using IE6". It is therefore your responsibility as the designer to cater for this and - of course - work out how long this will take for you to do and, therefore, how much it costs. Nothing new there folks.
Of course, you will also need to try and come up with methods to get the best out of the situation - a design that still looks and works great despite a heavy "17% IE6" anchor hanging around its neck. This may require some compromise and also some "graceful degradation" and "progressive enhancement" where appropriate. Hey, if it was easy it would be boring, right?
When you are asking the client if they "would like you to support IE6", or leaving it as a separate optional cost because "the browser isn't very good and it makes it difficult to do really clever things", what your actually saying, when it comes down to the nitty-gritty, is "would you like to have a possible 17% of your visitors unable to access and use your site properly?".
Just to set the record straight
IE6 is rubbish. If you are able to, use another browser - Firefox, Chrome, Safari, even IE7/8. Don't interpret the above as if I am in someway 'pro' IE6.
I know it's probably not 'cool' to be showing this much respect for a browser that has been such an absolute pain in the arse to developers and users alike for so long, but it seems many have forgotten what 'accessibility' actually means.
I realise that many will disagree, and also that - come 2010 - all this will be 'by the by'. Roll on 2010 I say. But in the meantime: it's raining outside and I have to walk the dog, and - whilst I don't particularly like my waterproof jacket - I'm going to have to put it on, otherwise I'll get wet or the dog won't get her walk.
Yer get meh? :P
COMMENTS
Nice one Kev. I agree in most parts.
In my eyes it's exactly the same set of issues raised by Accessibility. You would not drop support for people with visual impairments because of their 'faulty' or 'ageing' 'equipment'. Why should it be any different for IE6 users? Websites should be accessible to all.
Also a quick note re: W3C schools.
I'm always wary of using the stats provided by W3schools, especially if we are to assume that they gather their stats from their own access logs. It's safe to assume that a majority of people who visit w3cschools won't be using IE6 - as they probably have some active interest in web development. I don't know many (any) web developers who use IE 6 as they're day-to-day browser.
From http://www.w3schools.com/browsers/browsers_stats.asp :
"Statistics Are Often Misleading
You cannot - as a web developer - rely only on statistics. Statistics can often be misleading.
Global averages may not always be relevant to your web site. Different sites attract different audiences. Some web sites attract professional developers using professional hardware, while other sites attract hobbyists using old low spec computers.
Also be aware that many statistics may have an incomplete or faulty browser detection. It is quite common by many web-stats report programs, not to detect the newest browsers.
(The statistics above are extracted from W3Schools' log-files, but we are also monitoring other sources around the Internet to assure the quality of these figures)."
I find the most accurate method of determining the appropriate target browser support is through statistic analysis of the access logs for actual client sites that have been in the wild for a while. If you want something doing properly ...
Get Yuh.
@ Matt : Absolutely mate aye. Actual stats come 'number 1'.
Unfortunately though, we don't have / aren't often supplied that level of focused information. So I find the W3C Schools figures a good starting point - a fall-back if you like. So:
1) Actual statistical data from the site you are working on. If not ...
2) Actual statistical data and market research from a similar website(s) sharing same target audience etc. If not ...
3) Statistical data from the likes of W3C Schools site, and from averages of any other site you are recording data from / have access to - to get a ballpark feel which can often be enough to drop / keep support for a browser, or certain other development decisions.
I have a fair number of sites for a number of clients, and taking a brief look at the numbers (I use Google Analytics, btw) it would seem that, on average, the sites which I would consider to have a non-web-savvy overall demographic have IE6 users at around 15-16%, which does tally with W3C's numbers. My own site and the sites with a web-savvy demographic have the figure of around 8-10% for IE6 users.
Whilst the figures may not, indeed, be 100% accurate, they give a far better justification for making some quite big development decisions than: "I hate IE6, pisses me off, I'm gonna drop support for it", which I think we both agree on.
What's IE6?
Nice article, although I mostly disagree with you. Mostly, I think web developers have a certain responsibility for keeping the web a better place to be. Encouraging users to upgrade from IE6, IMHO, is part of that.
Your bit on accessibility is interesting, but I believe flawed. "Would you stop a friend from coming in your house because their trainers aren't the latest Nikes". Yes, I would stop them if it meant I had to construct special flooring in my house in order to accommodate those Nikes.
@ Mr. Monk : It's version 6 of my "Incredible Elephant".
@ Lance : I completely agree with you.
I haven't at any point said that developers don't have a responsibility to encourage users to upgrade to IE6. They do. Note the word 'encourage' you have used also. Dropping support is not 'encouraging'. It's final. "Use another browser or you're not welcome".
My point is that to ignore 17% of a user market is bad on any level, regardless of which browser we are talking about.
"Accessibility", giving people "Access". Think about what those words mean at their most basic level.
Your "special flooring" point: Whilst I understand my analogies are, at best, flaky, what you are basically saying is that you have to do some 'work' to enable access to your visitor by laying a special floor. So, in saying this, you are making it about you, and not the needs of your visitor.
Anyway, aren't we doing this everyday as developers? Making provisions to help our users gain access? IE7 doesn't support corner radius or other advanced CSS methods - shall I drop support for it because it is 'work' for me to ensure a design works without them? If we're not going by stats and, instead, how good a browser is, shouldn't we drop support for IE7? It's still way behind other browsers ...
I also note in my previous post about the need for a drawn out death for IE6:
"A date for the demise of IE6 was set the moment IE7 was planned, and this date was chosen in order to allow large companies and corporations an appropriate amount of time to update their systems."
Give it a read. Whilst it may frustrate you to learn that this is how long it is taking - and, indeed, has to take, we need to accept that we work in a commercial world, and that IE6, only a few years ago, had 99% of the market share. Practically every corporation would have had it's systems powered, in some form, and to a varying degree, on the IE6 framework and related technologies. You can't just 'pull the plug' on that, even though we are now all in total agreement throughout the web community that it is a 'shit' browser. There is a bigger picture to consider.
Yes, it is an important responsibility for developers to educate users, clients and other developers about the need to adopt more advanced web standards browsers, and to stop using (or upgrade from) IE6. But it is also a developers responsibility to cater for the needs of visitors to a website (especially where statistics show them to be a core and relevant user group), and to do so in the name of 'accessibility' at its most basic and obvious level.
Stumbled. Kinda surprised no one beat me to it :)
No, I think you're wrong. That 17% is not going to shrink as long as web designers pander to the wants of people who are not willing to get with the times. As long as people make sure their websites work with IE6, people will continue to use IE6. If you want that 17% to shrink (and who doesn't?) then you need to take the stick approach, rather than the carrot, and force those people either to upgrade or miss out on what the web really has to offer.
Pandering to the needs of, frankly, stupid or ignorant people is only going to delay how long it takes for new standards to be accepted and available and for the web experience of EVERYONE to be improved. In fact by trying to support IE6 you are essentially reducing the potential quality of the web for those who are using up to date browsers.
@Steve : Thanks for the comment. A few points and questions if I may:
1) Please take a read of my last post (if you would like to of course, I won't force you :P): Why I will miss IE6 which discusses how the 'war' against IE6 is already won (the moment IE7 was penned), and the importance of the controlled and gradual death of IE6 for the long term good of the web (that goes deeper than our own pet hates as developers). Microsoft can't just flick a switch on a technology that, at one stage, had 99.9% of the market share.
2) I don't think it's a good idea to call visitors to a website "stupid" and/or "ignorant", regardless of their knowledge of browsing technologies or, indeed, their access to modern browsing technologies. It is stupid and ignorant to base the value of a percentage of visitors to a website on the fact that "you know the browser they use is shit and they don't".
3) What, in your opinion, is the "stick approach" that you mention? Some kind of support? No support? Links to other browsers? If your pro "graceful degrading / progressive enhancement", and also perhaps some form of appropriate user education method, then you aren't actually disagreeing with me at all. If it is a "forced" method as you seem to be implying, what method would that be?
4) Are you sure, with accessibility in mind, that users should be "forced" as to which browser they use? Where do we draw the line if not with stats? For example: Safari 4 now has some awesome -webkit CSS additions for adding a whole host of effects and control that no other browser supports. Should we "force" people to use that browser only because the user experience is better than that of other browsers - for "the good of the web"?
5) This very site "supports" IE6. In doing so, how am I "reducing the potential quality of the web" exactly? I degrade gracefully, and I shall continue to use progressive enhancement to make use of current and emerging browsing technologies that support those enhancements.
6) I think to pin 'quality' on a website merely based on the fact as to whether it supports IE6 or not is not seeing the whole picture, and - lets face it - there are plenty of websites out there which don't support IE6 which are dreadful on many levels.
Mr. Boag on the subject: http://audioboo.fm/boos/53954
Oh dear. According to the BBC website, Microsoft plan to extend their support for another 4 years!
http://news.bbc.co.uk/1/hi/technology/8196242.stm
Urgh :(
Well let's just hope that stats drop significantly by next year anyhow, to justify chucking our support in the bin.