Jeff Croft

I’m a product designer in Seattle, WA. I recently worked at Simply Measured, and previously co-founded Lendle.

Some of my past clients include Facebook, Microsoft, Yahoo, and the University of Washington.

I’ve authored two books on web and interactive design and spoken at dozens of conferences around the world.

I’m currently accepting contract work and considering full-time opportunities.

Blog entry // 09.30.2008 // 4:39 PM // 82 Comments

When can we stop talking about “supporting” certain browsers?

Even at a progressive, Web Standards-friendly agency like Blue Flavor, the topic of which browsers to “support” comes up. Clients ask us, “Will our site be supported by IE6?,” for example. And even in the Web Standards community, there’s still a lot of talk about “dropping support” for IE6, and the like.

But doesn’t this whole idea of browser “support” kind of go against what Web Standards is all about in the first place? Because of the way we build sites (and by we, I mean me, Blue Flavor, and most readers of this site), our projects inherently “support” every browser, from Lynx to Mosaic 1.0 to Netscape to IE to Safari to the no-name browser on your crappy flip phone.

And yet, we still talk about browser “support.” What we really mean when we ask if a site will “support,” say, IE6, is “will the site look the same in IE6 as it does in the latest and greatest browser?” But we all know this is a silly question. Of course it won’t. And what’s more, it shouldn’t.

So what can we do about it? How can we move our industry away from this absurd notion of “browser support?” I don’t really have any answers, but I’m interested in starting the discussion.

P.S. I’m also interested to see how someone will twist this into me hating Web Standards.

Comments

  1. 001 // Harris // 09.30.2008 // 5:06 PM

    When I talk about supporting IE6, I’m not talking about whether it will look the same as in modern, more standards-friendly browsers. I’m talking about whether it will be legible and functional, which is certainly a concern with IE6 when building a website. I’ve designed websites in Firefox that had major problems that rendered them unusable in IE6. I wasn’t trying to get it to look the same, just to work.

    It may be that I’m in the minority here, but that’s what I assumed most people mean when they talk about supporting a browser. If I’m right, I don’t think we can move away from that notion of browser support. It’s just a practicality.

  2. 002 // Tom Watson // 09.30.2008 // 5:07 PM

    Couldn’t agree more. Support shouldn’t mean look identical and the sooner the web standards crowd starts talking more along those lines the better.

    The more I think about it this, the more I feel like the standards community has created this problem themselves. Instead of continuing to battle the problem we should just rephrase the problem.

  3. 003 // Jeff Croft // 09.30.2008 // 5:08 PM

    It may be that I’m in the minority here, but that’s what I assumed most people mean when they talk about supporting a browser.

    That’s certainly not what most of our clients mean. :) But yes, if you’re right, then we can’t move away from it, and we probably shouldn’t.

  4. 004 // Chris Pederick // 09.30.2008 // 6:04 PM

    The biggest problem I have is that most clients are using IE6. Often they don’t care how it looks in Firefox or Safari or even IE7 - they just know it doesn’t look perfect in their browser.

    P.S. Why do you hate Web Standards, Jeff?! :)

  5. 005 // Rev Matt // 09.30.2008 // 6:18 PM

    I would like to aim for functional in older browsers (IE6, Netscape) and aesthetically pleasing in current browsers. Our user base are still stuck on IE6, though if they can get special dispensation from the Pope then Firefox 2 is on the approved software list. The pilot for IE7 will probably start in about 6 months, Firefox 3 about 6 months after that. I work for a federal agency, we lag the business community we work with by an average of 2 years on the front end, 4 on the back end. And this sort of thing is why older browsers are still an issue.

    They shouldn’t be. There is no compelling reason for anyone to still be running Netscape of any version or IE 6.

  6. 006 // Anon // 09.30.2008 // 6:24 PM

    OMG stop hatin’ on ‘ma standardz BIATCHEZ!!!!

  7. 007 // Nate W // 09.30.2008 // 6:26 PM

    I agree, the term “support” puts bad spin on what really is the browser’s fault not the web developers.

    Similar to how the “death tax” was turned into the friendlier “excise tax” we should do the reverse, why not turn “browser support” into “browser misrendering” or something with an even less friendly connotation.

    Ideas? “browser defacement”?

  8. 008 // Nate W // 09.30.2008 // 6:26 PM

    I agree, the term “support” puts bad spin on what really is the browser’s fault not the web developers.

    Similar to how the “death tax” was turned into the friendlier “excise tax” we should do the reverse, why not turn “browser support” into “browser misrendering” or something with an even less friendly connotation.

    Ideas? “browser defacement”?

  9. 009 // Nate Klaiber // 09.30.2008 // 6:32 PM

    This is an excellent question. I would agree with Tom, in that I think most clients (at least from my experience), when asking for IE6 support are hanging on to the idea that it will look/function the same as it does in other browsers.

    The question you bring to the table is a good one, and I am not sure exactly how to answer it, as it seems to be different depending on the angle the client is coming from.

    Do you think we have a responsibility, as we do with other areas, to educate the client on what supporting another browser will actually entails? I am sure even this may vary from business to business.

  10. 010 // Justin Lilly // 09.30.2008 // 6:48 PM

    So I hear you hate web standards…

    Aaanyway. I don’t think “functional” is what clients are after either. You could just as easily serve the mobile/iphone stylesheet to all ie users and likely still have the same functionality. ie: getting at content, etc.

    I do agree with Chris, however. Given the clients I’ve worked with, you could almost justify coding for ie6 first because that’s all they ever see it on.

  11. 011 // steven Southard // 09.30.2008 // 7:23 PM

    The overhead and hacks needed to make IE6 look as good as FireFox or Safari just aren’t worth it. If they’re using IE6 they probably don’t care about aesthetics. So things won’t line up the way we intended, so what? As long as they can still read it I doubt they’ll even notice.

  12. 012 // Brandon // 09.30.2008 // 7:36 PM

    Unfortunately, until IE6 users drop below 30% of the market, it’s still a necessary evil to prop up the hideous piece of flop that is IE6. Clients don’t care about the hows and whys of why browser compatibility is such an issue, but IE6 is so bad, and is so time-consuming to support, that a special case has to be made against it.

    For personal projects, I have dropped IE6 support altogether, and charge extra for it :)

    What works best for me is to do my testing in this order:

    Opera FireFox Safari IE7 Camino/Chrome

    If I do have to do something special for IE, I prefer using conditional stylesheets and try extremely hard not to use hacks.

  13. 013 // Nathan Youngman // 09.30.2008 // 7:46 PM

    The designs I work with must be really simple, or just from reading up with CSS Mastery, because while IE6 is problematic, it hasn’t been too incredibly bad. I don’t even have any hacks, just display: inline; for the double margin float bug.

    I agree with the comments though, we could “support’ IE6 the way a site supports Lynx, but clients wouldn’t be happy with that, and the numbers of users is likely still to high for most sites.

  14. 014 // Ben Buchanan // 09.30.2008 // 8:09 PM

    Rough definition: “supported” means if the client rings up and yells about something and you have to fix it, that’s a supported browser ;)

    Tiered support is the way to go - differentiate between “it will look just right” and “it will be functional” and “good luck with that”.

    You’re right of course, we shouldn’t support browsers. We should support standards and let the browsers work it out. But that’s just not the reality. Maybe one day.

  15. 015 // Jeff Croft // 09.30.2008 // 8:09 PM

    Good discussion so far…just a couple of thoughts:

    1. Let’s try not to make this about IE6. The point here is that we write to specs, not the browsers. If we say, “we can’t stop ‘supporting’ IE6 until it drops below 30% of the marketshare,” we just set ourselves up to be dealing this again when we’re all fed up with IE7, or Firefox 4. Instead of focusing on one bad browser, this needs to be about how we get everyone on the same page that it is normal, natural, and right for things to look different from browser to browser.
    2. Would an analogy help? Those who know me know I’m always making analogies. This is a bit like television, to me. TV these days is filmed in HD. That doesn’t mean every TV gets HD content, though. If you have an older TV, you’re going to see it in standard def. Hell, if your TV is old enough, you might even see it in black and white. The same content is rendered differently based on what platform you’re using to view it. The same is true of the audio. Got a nice setup? You’ll get surround sound. An older setup? Stereo. Even older? Mono. Same content, rendered differently.

    One one hand, it seems clear that us web standards-oriented web developers understand this. On the other hand, we’ve got sites and petitions and calls from within our community to “drop support” for IE6 — which shows that maybe they don’t understand this. And I think we can all agree that most of our clients don’t understand this.

    I do think this is about terminology and how we frame it. That’s why I say we need to stop talking about “browser support.” We need another way to frame this besides “support.”

  16. 016 // Jeff Croft // 09.30.2008 // 8:14 PM

    I don’t even have any hacks, just display: inline; for the double margin float bug.

    Heh. That’s one of the most contradictory statements I’ve ever read. :)

    But you’re absolutely right. Making your site work the same way in IE6 that it does in more modern browsers usually isn’t incredibly difficult. But that’s not the point at all. The point is, we shouldn’t be doing it. We should be writing to the spec and letting the browsers render how they will. That’s the Web Standards way. Right?

    And again, this isn’t about IE6 — it’s about the general misunderstanding of what it means to “support” a browser.

    Tiered support is the way to go - differentiate between “it will look just right” and “it will be functional” and “good luck with that”.

    Absolutely, and I advocated this approach in Pro CSS Techniques. But, has anyone had any real success getting clients to buy into it? Sure, we all get the occasional client who really “gets it” (I’ve got a coupe right now, actually), but more often than not, we’re dealing with people who expect their sites to look the same in all browsers. Knowing it’s unrealistic, how do we change this expectation?

  17. 017 // Mike D. // 09.30.2008 // 8:36 PM

    I won’t twist this into you hating web standards since we feel exactly the same way about them, but I do call bullshit on your claim that all sites you develop at Blue Flavor “support” every browser in the world, including Lynx and Mosaic. Your site only “supports” those browsers if you test in them (which I doubt you do… when was the last time you fired up Netscape 4) and if they work similarly well as a site that was specifically designed for them.

    Just because your HTML loads into that browser does not mean your site “supports” it. The fact is that it doesn’t support it at all. It disregards it. Which is perfectly fine. Let’s just not pretend we actually care about browsers that less than 1% of the population use… because in most cases, we don’t. No matter how much we may say we do.

  18. 018 // Jeff Croft // 09.30.2008 // 9:22 PM

    Your site only “supports” those browsers if you test in them (which I doubt you do… when was the last time you fired up Netscape 4) and if they work similarly well as a site that was specifically designed for them…

    You’re totally right, but I think you’re proving my point. The point is: what it means to “support” a browser is very different for different people. What you mean why you say “support” is different form what I mean, which is different what what clients mean, which is probably different from what the “let’s drop support for IE6” guys mean.

    Let’s just not pretend we actually care about browsers that less than 1% of the population use… because in most cases, we don’t. No matter how much we may say we do.

    Absolutely. I don’t pretend to care, and if this made it sound like I do, my apologies. In fact, it’s the exact opposite (as this site is evidence). If I have my way (which I usually don’t, when working on real-world projects), I really only care much about Safari and Firefox. I make sure a site looks good to me in those two, and, as you say, generally disregard the others. But the reason I can do this is that I know my development practices are sound, so I’m reasonably confident my content will be available to virtually anyone in any browser, and that’s good enough for me.

    The real point here is this: what does “support” mean? Obviously, it’s not the same from person to person.

    Good comment, yo.

  19. 019 // Hein Haraldson Berg // 09.30.2008 // 10:36 PM

    We should be writing to the spec and letting the browsers render how they will. That’s the Web Standards way. Right?

    It would be nice to see a realistic approach to this, just once. I see your point, I really do – but despite all frustration it shouldn’t be all that hard to keep real and focus on the realities that surrounds us?

    Have you guys for instance never worked with a site where the design is actually playing the most important role? I work on some of those, and yes – the percentage of whom is visiting the sites with browsers lacking standards support is still ridiculously high. If the design were to break on any of these sites, it would have a direct, major impact on the sales, because the design is often what’s selling.

    All of a sudden the O’ so great web standards isn’t really worth all that much. So, as I understand what you’re saying, my opinion is that it is not possible, in many cases, to play out the standards-compliant-but-not-looking-the-same card.

    I try as hard as I can to combine, and I see nothing wrong in having a site looking the same in both old and newer browsers, if possible.

  20. 020 // Aaron // 09.30.2008 // 10:39 PM

    It’s because the new kid on the town is “progressive enhancement” up until this new term was banded about designers and developers were trying to make websites look the same in all browsers.

    At the end of the day you do testing for IE6 and fix any bugs, or you know from experience the things that IE6 dislikes and work around them when you are coding your site.

    Doing this is supporting IE6

  21. 021 // Jeff Croft // 09.30.2008 // 10:43 PM

    …but despite all frustration it shouldn’t be all that hard to keep real and focus on the realities that surrounds us?

    Have you ever read my site before? Half my posts are railing against the ridiculous ideals of most standards zealots. I’m with you on the real-world front, my friend. :)

    Have you guys for instance never worked with a site where the design is actually playing the most important role? I work on some of those, and yes – the percentage of whom is visiting the sites with browsers lacking standards support is still ridiculously high. If the design were to break on any of these sites, it would have a direct, major impact on the sales, because the design is often what’s selling.

    I work on these sorts of sites every single day. You, my friend, are missing my point. I’m not at all suggesting we shouldn’t strive to have our design “look right” in whatever browsers make sense from a business perspective. Of course we should. All I’m saying is that “support” isn’t a very good word for “look right.” “Support” isn’t a very good work at all, because it’s entirely non-specific about what it means. When a client asks me if to make their site “support” IE6, they mean something very, very different than what web standards advocates mean when they say, “let’s drop support for IE6.” And that means something very different from what Yahoo means when it says “support,” and I know this because Yahoo actually put out a really nice grid explaining what they mean.

    I’m not saying we should make things “look right” in as many browsers as possible, I’m just saying we shouldn’t muddle the waters by calling that “support.”

    I try as hard as I can to combine, and I see nothing wrong in having a site looking the same in both old and newer browsers, if possible.

    Of course. Me too. I hate to say it, but I think you just really missed the point (probably my fault, not yours).

  22. 022 // Jeff Croft // 09.30.2008 // 10:47 PM

    At the end of the day you do testing for IE6 and fix any bug…Doing this is supporting IE6.

    Okay, then. What’s a bug? Is this site not having rounded corners in IE6 a bug? Is it a bug if some content appears in the wrong place in IE6? Is it a bug if a transparent PNG doesn’t display transparently in IE6? IF my content can’t be seen in IE6, is that a bug?

    I would be willing to bet my answers to these are different than yours. So, just saying we both “support” IE6 means nothing. The question is: what does it mean to support something?

  23. 023 // Hein Haraldson Berg // 09.30.2008 // 10:56 PM

    I was maybe expecting a deeper meaning than a rant about human language semantics. Hehe.

    But, if we first get down to that level, the word support has a totally different meaning depending on who is using it. We’ll never get rid off those silly clients with the outrageous vocabulary screaming to get bashed at some random anonymous site, such as notalwaysright.com. It applies to all trades.

    So, I’m just going to treat this blog post as an outburst of frustration.

  24. 024 // Jeff Croft // 09.30.2008 // 11:10 PM

    But, if we first get down to that level, the word support has a totally different meaning depending on who is using it. We’ll never get rid off those silly clients with the outrageous vocabulary screaming to get bashed at some random anonymous site, such as notalwaysright.com. It applies to all trades.

    Sure, but don’t you think we can try to reframe the discussion to be more specific? Instead of just talking about which browsers we “support,” why not publish a tiered document like Yahoo has (for example)?

    This post wasn’t actually out of frustration at all. Rather, I just keep seeing these calls to “drop support for IE6,” and I want to know what that means. Does it mean “stop testing in IE6?” Does it mean, “pop up a message and tell your users to upgrade?” Does it mean “feel free to use CSS properties specific to Safari or Firefox and don’t sweat it if they don’t work in IE6?” I don’t know what it means. And just the same, I don’t know what it means when a client request that we do support IE6 (so I usually ask them).

    We need to stop talking about “supporting” browsers and start talking about how our sites will render in various browsers.

  25. 025 // Victor // 09.30.2008 // 11:13 PM

    Whenever i develop a site i test first on Firefox, then on Safari, and most recenly on Chrome just because i love it.

    But the reason i do this is because generally this browsers render correctly if you code as you should. Wich most of the time i try to do, with the help of a validator.

    For me supporting is that your site looks AND works exactly the same on the browser you’re intending to support, wich in the case of ie6 can be a headache. I only do this when my clients ask me to do so or when i realize a big part of my users will be using ie6.

    You’re right. We should code to the specification, but that does not guarantee that our site will work or look the way we want it on browsers like ie6.

    You pointed out some good analogies, but you can’t expect to play dvd’s on devices that were meant to play vhs.

  26. 026 // Hein Haraldson Berg // 09.30.2008 // 11:34 PM

    Does it mean “stop testing in IE6?” Does it mean, “pop up a message and tell your users to upgrade?” Does it mean “feel free to use CSS properties specific to Safari or Firefox and don’t sweat it if they don’t work in IE6?” I don’t know what it means. And just the same, I don’t know what it means when a client request that we do support IE6 (so I usually ask them).

    The majority of the clients I work for expect that the site will look and work alike in all browsers (we don’t go further back than IE6 and don’t care about Netscape).

    A mechanic wouldn’t spend time wondering what a client means when (s)he says «fix my car». «Did the cusomer mean that I should fix the engine, or would it be sufficient to fix this broken tail light?» For a client, I think your questions and my silly example above, is equally stupid.

    Therefore, I think all of the possible meanings you list above are quite unlikely. Customers in general have no knowledge about that stuff. Supporting a browser for a client simply means «will it look, feel and do the intended stuff in browser x» – not necessarily in comparison to any other browser, since browser x tends to be the one and only browser the client’s got installed.

    Yes, they expect it to look and feel the same in IE6 as it does in Firefox, some webkit browser, Opera etc. when we are coding it all up. I don’t think there’s much more to it.

  27. 027 // Craig C. // 09.30.2008 // 11:49 PM

    It is indeed a matter of poor phrasing (semantics, heh). The question of “supporting browsers” is really a question of whether the browsing software is capable of supporting the languages and techniques employed to build the site. If you build sites according to the specs, and if the browsers all supported those specs fully, correctly, and equally, then we wouldn’t face these issues at all. That’s the ultimate dream of web standards. If everyone follows the rules, everyone gets to play.

    We don’t need to support web browsers. Web browsers need to support us.

    What gives us migraines is a browser’s lack of support for current web standards, or, even worse, bad support. All of the venom I’ve spit at IE6 is inspired by that browser’s failure to properly support the sites I build. To get a site looking decent in IE6, we have to bend over backwards and cast all kinds of hoop-jumping voodoo to accommodate IE’s shortcomings. “To drop support for IE6” really means “to cease catering to the whims of bad software.”

    So let’s not call it “browser support,” let’s call it what it is: browser accommodation.

    I don’t support IE6. But I must begrudgingly continue to accommodate it when and where practical (and it’ll just have to accept a degraded experience). When its userbase finally dwindles low enough, that extra effort will become too costly and I’ll remove IE6 from my QA process. On that day, IE6 can go take its place in the catacombs next to IE5, NN4, and all the other dead browsers.

  28. 028 // Jeff Croft // 09.30.2008 // 11:52 PM

    You pointed out some good analogies, but you can’t expect to play dvd’s on devices that were meant to play vhs.

    Nice one. :)

    Customers in general have no knowledge about that stuff.

    Absolutely, but I wasn’t asking customers, I was asking you, and the web developers that read this site. :)

    Yes, they expect it to look and feel the same in IE6 as it does in Firefox, some webkit browser, Opera etc. when we are coding it all up. I don’t think there’s much more to it.

    Right. I get that. But we know that’s unrealistic. We know that, unless you either code for the lowest common denominator or maintain browser-specific code, a site will not render the same from browser to browser. Hell, even if you do those thing, it still won’t, because every browser display type, images, and such, differently from the next one. The problem with clients is that their expectations are unrealistic. That’s not their fault — they usually don’t know they’re unrealistic. But they are.

    So, do we have a responsibility to help clients understand this? Should we be explaining to them that you “can’t play a DVD on a VCR?” It sounds like you think we shouldn’t be. I think we should be. Because if we don’t, we’ll never get to a point where clients have realistic expectations.

  29. 029 // Hein Haraldson Berg // 10.01.2008 // 12:16 AM

    So, do we have a responsibility to help clients understand this? Should we be explaining to them that you “can’t play a DVD on a VCR?” It sounds like you think we shouldn’t be. I think we should be.

    We take the time to educate one of our design partners, when they report back to us about 1-pixel inconsistencies in Firefox and Safari.

    But educating our clients? Very rarely. Fact is, that in most cases making IE6 read DVD’s isn’t all that difficult. I point it out to the customer when it’s clearly not possible in eg. IE6, or it would mean +10 hours just to make it work in a specific browser.

    Most of the time it’s more hassle trying to educate our clients, than to just do what we know they really mean and in most cases is fairly easy.

  30. 030 // Jeff Croft // 10.01.2008 // 12:50 AM

    Fact is, that in most cases making IE6 read DVD’s isn’t all that difficult.

    Right. For IE6. But what about other browsers? You don’t make Netscape 4 read DVDs, I presume. Do you worry that you’re setting up and expectation that you’ll make IE7 able to read Blu-Ray, too, when it’s on the way out and you’re making awesome shit for Firefox 4.5? If so, are you confident you’ll actually be able to do that?

    Most of the time it’s more hassle trying to educate our clients, than to just do what we know they really mean and in most cases is fairly easy

    That’s what we do, too. And you’re right, most of the time it’s not really a problem. But I worry that we’re just continuing the cycle, rather than trying to put an end to it.

  31. 031 // Marcus Olovsson // 10.01.2008 // 1:15 AM

    I usually tell my clients that I don’t support IE6, but my code is backward-compatible so the website can be used in all browsers anyway.

  32. 032 // Hein Haraldson Berg // 10.01.2008 // 1:30 AM

    Clients tend to have high expectations after being sweet-talked to by some of our salesmen anyway, I don’t think I can do much to change it.

    If I were a designer (I currently do HTML, CSS and some JS) creating a design based on some guidelines and wishes from a client, or maybe to fit into an already existing profile, I would probably be able to control the client’s expectations through my wireframes mock-ups and designs. When a project lands on my desk it’s often too late. The sky high expectations are already there.

    I think you have a good point with the cycle never ending, getting worse and worse, but I am currently not in a position where I feel I can do much about it.

    The situation and reality is also changing, though, and the new browser race we see today will probably make it easier to read them Blu-ray discs in the future. But that doesn’t mean the expectations can’t get too high…

  33. 033 // Pete Eveleigh // 10.01.2008 // 2:13 AM

    I do agree with the sentiment of the post and like the TV analogy. However the crux is that if you serve HD content to an old Black and White TV you still see the picture and hear the sound.

    Visiting a site in IE6 can be more akin to trying to watch that HD TV broadcast on a radio. TV and Radio sort of work in the same way but they are different media and work in different ways.

    To extend and alter the analogy a little I would compare it to the transition between Analogue and Digital Broadcasting. At some point the TV owner is going to have to upgrade their equipment. Broadcasters will be ceasing to ‘support’ Analogue TV. End of story. Sure, it could happen that someone starts broadcasting in analogue to cater for those that don’t have digital - they are the exception though. With IE6 we are constantly trying to avoid it being ‘the exception’ when in reality it probably should be.

    Unfortunately we really are ‘supporting’ IE6 since it can mangle a layout to the point of illegibility unless we pro-actively do something with our HTML and CSS.

    Personally I prefer to use the phrase ‘Cater for IE6’ as this puts the emphasis on it being the browser’s nuances that are the problem. Our work will be sympathetic towards IE6’s foibles, but we aren’t going to bust our guts over it since it’s a dead, outdated piece of software and there are plenty of other viable option. This phrasology does seem to help clients understand too.

    The key is in clear communication with the client and sometimes educating them a little. I, myself, am pushing for making IE6 a low priority browser in which the client should expect his site to be readable and useable but not necessarily in quite the same way as other, more modern, browsers.

    Oftentimes I’ve found that clients ‘stuck’ with IE6 are in that position because of old, legacy software and often this is because of a rut their IT dept. has led them into. The unwillingness to upgrade stems from either an unwillingmess to put in the effort to make things right or a blind denial of the fact they are on a dead end path. Those that aren’t tied to some old piece of junk but just haven’t upgraded due to lack of knowledge or who plain don’t see the need to will often happily upgrade once you tell them why they should.

    The less we ‘cater for’ IE6 the more inclined those people responsible for holding back on upgrading (for whatever reason) might be pressured to upgrade. If we pander to it, it will live on and I don’t think any serious web professional thinks that is a good idea.

    Crikey, my comment is longer than the original post! I should start my own blog!

  34. 034 // Sean // 10.01.2008 // 3:32 AM

    I really like the television analogy.

    I’m new to web dev (handing off my first site to client in a few hours) and my code is valid (XHTML & CSS). I’ve found for the most part things look ok in ie6 and as far as I’m concerned that’s fine with me. I’ve told my client that ie6 is 7 years old and of course it won’t render the same as ie7/8 or ff.

    With future clients I’m going to use the tv analogy and hope that it works - I can’t imagine a better way to get technically challenged clients to understand this.

    We don’t need to support web browsers. Web browsers need to support us.

    I like this also. and I also second Craigs proposal to call it Browswer Accomadation

  35. 035 // bruce // 10.01.2008 // 3:45 AM

    I really liked the television analogy too.

    Here’s what I wrote for the Solicitors Regulation Authotity (UK legal regulator):

    Browser compatibility

    It doesn’t matter how old your browser is—you can use this website. It looks different in some older browsers, and is mostly text in very old browsers (like Netscape 4, or Internet Explorer for Macintosh). But the information is the same, and so are the things you can do.

    http://www.sra.org.uk/sra/access…

    Of course, a public information website has different priorities than a super-sexy design site. But most design-heavy sites still want to reach as many people as possible.

  36. 036 // Matt Wilcox // 10.01.2008 // 5:39 AM

    Firstly, you say “yes, we support IE6” - which puts the client in a good place. Then you clarify what that means along the lines of:

    You’ll be glad to know that we support all browsers, and all of your content is available to all browsers. With old browsers there will of course be certain limitations in the look and feel - a car from 1960 won’t look and feel like a car from 2008, nor will it have traction control and air conditioning. But it’ll drive down every road, and everyone can get from A to B perfectly well with it.

    Expect look and feel differences in older browsers, they are not capable of some fancy new things.”

    And then if need be you can always clarify what /features/ of the site are supported in which browser.

    Talking about browser support in terms of supported design features rather than a blanket statement would add clarity I think.

  37. 037 // Vladimir Carrer // 10.01.2008 // 6:37 AM

    People still using IE6 is about 25%! And that is a lot of people. Does your client care that you hate IE6? Try to explain that he will lose 25% of his visitors.They are only 5% IE6 visitors on my blog but 40% IE6 for our company site (Italy). I feel that if my final product(web site) support more browser is better, I need to write more CSS but this is my job. Personally If is something that I hate more then IE6 is Opera! But I test, test …, this is my Job!

  38. 038 // Andrew Meyer // 10.01.2008 // 7:06 AM

    Great article, Jeff. I really appreciate the way you are able to get people thinking about and rethinking the conventions of our trade. Helpful and important.

    Anyway, here’s my thought: I think the answer to this sort of question can be found in the direction of understanding the goal of a project to begin with. By this I mean, we should look at the success of a given project as being the extent to which it meets it’s goals in a whole host of conditions.

    Regarding “browser support”, I do think we should stop fixating on it, and begin asking ourselves does this site achieve its purpose for all users. If a website’s primary purpose is to communicate information, then ask how well does it communicate? If a given design is showing up differently in different browsers, ask does this lack of consistency really affect the site’s ability to communicate its message? Maybe the answer would be yes, but often, I think, it would be no.

    On a broader-reaching note, if we were able to generally focus more on websites meeting their goals and achieving their purposes rather than on the details of whether or not they validate 100%, or whether they look exactly the same in all browsers, or whether all classes are non-semantic, etc. I feel like we’d make much faster progress in improving users’ experiences. The other things DO matter, but only within a context of communication. I THINK, (this is my humble opinion) that this would be a more profitable direction for us.

  39. 039 // Jason // 10.01.2008 // 7:21 AM

    I think a lot of the support for IE6 has been ingrained into my work flow to where now things usually work first time when testing. It seems like a lot of sites lately have had large image backgrounds and PNGs as backgrounds layered over the large image for transparency. In IE6 they don’t get that sexiness.

    I think the major opinion and the one I subscribe to is, you stop testing in it when the users of the site stop using it. Most sites I’ve done still have around 30% of their users on IE6. Maybe this is lower than the national trend because I’m in a smaller market in Missouri, but I don’t think it is, and in no other industry would creating something 30% of the customers may not be able to use, or may have a negative effect on them, even be a thought.

    I don’t try to make IE6 identical, but I try to get it close - within 10% if something like that is even measurable.

  40. 040 // Corey Dutson // 10.01.2008 // 8:03 AM

    It’s funny that you talk about this, because I had a similar conversation with myself when I was remaking my website.

    I wanted it to look fantastic in every browser - the best laid plans and all that - and set out writing semantic code and clean CSS. The result was, for the most part, a success. The exception of course was ie6.

    I spent an unreasonable time tweaking my layout over and over until things started to look right in ie6 and the others. I eventually got it “pretty good” where I only had one or two possible overlap issues in ie6.

    That’s pretty much when I drew a line in the sand and said “Know what people? If you want this to look super awesome, upgrade your browser.”

    I regret nothing with this decision. I’ve noted time and time again that in an attempt to make ie6 look the exact same as the other browsers, you either eat away your time tweaking, or you downgrade the design to suit browsers like ie6. I don’t believe in stepping back from a good design just because one browser throws a hissy fit.

    The point I’m trying to make is that as far as ie6 support goes, and talk about supporting certain browsers, we are somewhat stuck until the latest browsers support markup and CSS the same. Until that time, we will have to make concessions based on project requirements.

  41. 041 // John Hancock // 10.01.2008 // 8:09 AM

    Do you hate your designers?

    We love ours, and play ball with a wide variety of them, which is why the aim is for sites to be pixel-perfect matches of the layout mocks, whether created in VS, Notepad, Dreamweaver, Illustrator, Indesign, Photoshop or just in the mind.

    To me this isn’t an issue of browser support, but of designer support. Getting sites to render perfectly cross-browser may be challenging, but isn’t the reward in allowing non-technical or less net-savvy users the same experience that the Chrome, Opera, Firefox and Safari users get?

    I accept that your worry is to do with perpetuating a ‘well it should look great in IE6/5/4/3…’ culture and the feeling that it’s not important to have a standards-supporting browser. Commercially however, I have to sit on my standards-based ego and cry inside instead!

  42. 042 // Bruno Vilches // 10.01.2008 // 8:25 AM

    Yes, lovely, what we should do is exactly what you guys are saying. But what we should do lives in a fantasy world and what we must do live in our real world.

    You just can’t say to a client that “your website may just not look the same in IE6”. That’s because the don’t care, they don’t understand.

    The decision must be taken beforehand: either we list the browsers that the site will be “supported” by (maybe including notes about things that will be impossible to do) or we just tell the client that we cannot assure that the website will look the same in all browser and that’s not our fault

  43. 043 // Graham // 10.01.2008 // 9:07 AM

    Having gone through more ie5 box model hacks than there are pixels on the interwebs, I would hesitate to call ie6 a “hideous piece of flop…”. Yeah, its not FF3, but then again, look at how old it is. And sadly, if the client CEO has IE6 on his machine at 800 x 600 resolution, then no amount of statistics and pleading will get him to accept a design based on alpha channel transparent PNGs and 1000px wide layouts.

  44. 044 // Adam @ E-commerce // 10.01.2008 // 9:19 AM

    I’m personally sick of the whole browser support problem that seems to occur quite often when I build a site. They should all display the exact same thing - a basic requirement in terms of the expected web standards.

  45. 045 // Kyle Weems // 10.01.2008 // 9:20 AM

    I think that a lot of standards-aware developers get caught in the “what browsers do we support” trap too easily because certain high usage older browsers (cough IE6 cough) are being used by just enough clients that we often fall into the trap of making a proof, showing it to the client, then finding ourselves unwittingly committed to making the site shine in both the antiquated browser they’re using as well as modern ones.

    It feels to me the issue is one of client education, as well as a wake-up-call for ourselves when we’re in a more idealistic mood. The television analogy is a great one. By putting realistic expectations into clients’ heads from the beginning, we can concentrate on sites that work for all, but don’t have to be screamingly hot for all.

    One way to perhaps head off old-browser clients at the pass is focusing more on showing clients HTML/CSS prototypes, rather than static visuals, like Andy Clarke recently re-championed in http://forabeautifulweb.com/blog…. We can get them to then look at the prototypes in the browsers they use, and thus set their expectations at a level that’s more on par with reality.

  46. 046 // Jason Armstrong // 10.01.2008 // 9:37 AM

    How about calling these issues what they are? They are ‘browser deficiencies’. Most every browser has its own issues (or where it is deficient) where web standards are concerned and overcoming each of these deficiencies is the real task. This is the reality and most likely will be as long as there are multiple browsers created by multiple groups all interpreting standards in their own way.

    I agree that a website should not necessarily look the exact same in every browser, but I think we can all agree that it should look as similarly as possible from one browser to the next.

  47. 047 // Chris Johnson // 10.01.2008 // 2:39 PM

    Fact is, that in most cases making IE6 read DVD’s isn’t all that difficult.

    Right. For IE6. But what about other browsers? You don’t make Netscape 4 read DVDs, I presume. Do you worry that you’re setting up and expectation that you’ll make IE7 able to read Blu-Ray, too, when it’s on the way out and you’re making awesome shit for Firefox 4.5? If so, are you confident you’ll actually be able to do that?

    These analogies are hurting my tired brain. :-)

    My $.02: I don’t think the question is whether or not we should drop the word “support” from our discussions with clients and other developers. We should simply define what “support” means.

    When I say support, I think that means the site looks roughly the same and maintains core functionality for delivering content.

    I think that when Jeff says support he means simply that you can read the text of a page in your browser. I think, in a lot of cases, that’s all you really need.

    As a colleague of mine likes to say, how many NS4 or IE5 users aren’t used to half of the web looking screwed up to them anyways?

  48. 048 // Jeff Croft // 10.01.2008 // 4:23 PM

    As a colleague of mine likes to say, how many NS4 or IE5 users aren’t used to half of the web looking screwed up to them anyways?

    Amen to that. :)

    For what it’s worth, this is also one of my “excuses” for using techniques some people would says “aren’t accessible,” like absolute font sizes. If a person has low vision and needs to resize text, they probably got tired of using a browser that wouldn’t let them long ago, and have since moved to a browser which resizes absolutely-sized text just fine.

    But that’s totally not related. :)

  49. 049 // Ben B // 10.02.2008 // 3:32 AM

    On the question of will it work and will it look the same, it would be good to see what others do.

    We’ve used IE6 as the spoilt child that takes all the attention for minimal gain. If you use realistic standards then it will work.

    In terms of will it look the same part of me is thinking that when I re-design, I should send something close to a print version of CSS to IE6: serialised view, nicer fonts and perhaps some colours; rather than bother with the frustrations of positioning and other tweaks needed just for IE6.

    In designing from scratch, is it worthwhile to do this? Could we get by with two stylesheets — better browsers get the whole lot; but unfriendly browsers get a really cut-down stylesheet? Is this an approach that would avoid additional work on hacks while offering a slightly-better-than-plain-text view to unfriendly browsers?

  50. 050 // Chris Johnson // 10.02.2008 // 7:54 AM

    Could we get by with two stylesheets — better browsers get the whole lot; but unfriendly browsers get a really cut-down stylesheet?

    Ben B, That’s an interesting idea.

    On one hand, I wouldn’t want to carefully craft a downgraded stylesheet or use browser sniffing for browsers that don’t support the IE conditionals.

    On the other hand, if you’re already crafting a mobile or print stylesheet for that project, then it wouldn’t be too much work to force a slightly modified version of that onto unsuspecting IE users.

  51. 051 // Jeff Croft // 10.02.2008 // 8:50 AM

    Ben, I think that approach is sound, as long as the simplistic stylesheet isn’t just as much work as doing the full deal. If you can bang it out in an hour or so, then sure.

  52. 052 // Kristopher // 10.02.2008 // 9:13 AM

    Over the last few months while working for two clients, I’ve decided that if it isn’t for client-work I’m not going to even bother with IE6 and those other browsers. I take more time and effort crafting a stylesheet for IE6 then I do for the rest of the project, and I’m really fed up with it!

    When my client asks me whether it will work in IE6, I say yes. Then they ask, will it look the same? And I tell them, Well…no, it won’t. They get all iffy and start wanting a new design that’s going to work and look the same.

    I have to agree with Ben and Chris, but it’s getting your client to understand and agree that the stuff up-to-date browsers do, IE6 can not.

    IE6 should be banned from all computers, Grah!

  53. 053 // Matías // 10.02.2008 // 8:50 PM

    It could be a problem of supporting being a physical activity, you know, like having a big fat hippo clutching on your back —then the answer would be yes. Let’s stop talking about supporting the hippo and let’s paraphrase it as sustaining the hippo. Sustain feels like we are good with the hippo, his moral ground, and he —the hippo— would feel more than a burden. But I guess the ultimate problem is that you are so against standards that it is not groovy.

  54. 054 // DRoss // 10.04.2008 // 5:53 AM

    Is it really that complicated to have a site you are building look good in IE6? Not really. You may have to add a few IE6 specific styles but other than that, unless you suck, it really shouldn’t be a problem. It still has ~20% share and you know your clients are using it.

    As much as I absolutely hate IE6 there’s no getting around it right now. Maybe in a year we can forget about it?

  55. 055 // DAZ // 10.05.2008 // 8:19 AM

    A great piece Jeff. Love the TV and VHS/DVD/Blu-Ray analogies and Matt’s car analogy.

    I’m not sure that you’re right about web standards though - IE6 gets the CSS spec soooo wrong that you just can’t rely on it working by just sticking to standards. You need to have some sort of awareness of this browser and its rotten deficiencies when you’re coding.

    I’ve been a bit concerned recently with how many people are saying they will ‘drop’ IE6 without them really specifying what this means. I think that if anything this post has got people talking about where they stand on the issue. Fact is that too many people still use it to drop entirely, but the time has definitely come to move on and accept that this browser won’t be able to display sites the same as in other browsers.

    I’d say to all those people that are saying it’s easy to make a site work in IE6 that they maybe aren’t pushing the envelope that far because a lot of the cool stuff just isn’t possible in IE6 without a lot of hacking or resorting to javascript.

    The fact is that IE6 can’t do a lot of the stuff some of the more recent browsers can. Some big changes have happened in the last year or so as webkit and gecko have implemented some CSS3 features. Geez - even IE7 can’t do most of these. Firefox can’t do transitions, but they are cool and this very site is all the better for using them.

    My position is that if you want rounded corners you shouldn’t mess about using multiple divs and background images to get rounded corners anymore - if the browser doesn’t understand border-radius then you you don’t get rounded corners, simple.

    One area that you can’t compromise on is layout - I’ve seen far too many ‘modern’ sites that are just a mangled mess in IE6. You certainly need to find a way of making your site display so that it is legible.

    Surely the choice for clients is - it either looks the same and decidedly average in all browsers, or it works in all browsers, but looks a hell of a lot better in some of them.

    A ray of light has appeared recently in the form of Chrome. Because this is based on webkit it should render all of the nice modern stuff and make websites look fab. I think that clients and customers tend to trust the Big G, so when they see how much better their site looks in Chrome, they might be persuaded to switch.

  56. 056 // Berta Berlin // 10.06.2008 // 6:21 AM

    This is because web pages are not only about content, they are also about marketing. Marketing takes care on looks - so the product has to look like the designer created the style. Obviously (for us non Marketers) it would make more sense to use what the browser can provide (like Andy said in transcending Design). But unfortunately we’re not the one to pay us :-)

    But I am sure that this will change soon.

  57. 057 // jawright // 10.06.2008 // 7:36 AM

    So, do we have a responsibility to help clients understand this? Should we be explaining to them that you “can’t play a DVD on a VCR?” It sounds like you think we shouldn’t be. I think we should be. Because if we don’t, we’ll never get to a point where clients have realistic expectations.

    I’m all about educating users/clients, it seems like people are becoming more sophisticated internet users at a steadily increasing rate. If we can go to these people and show them that coding to the standards looks great ‘on-every-browser-but-this-one’ more and more of them will concede the point…this will hopefully put pressure on browser developers to toe the line and correctly support the standard. And, someday, we won’t need to have these discussions.
    And, we’ll all get ponies for Christmas.

  58. 058 // Bobby // 10.07.2008 // 10:38 PM

    So what can we do about it? How can we move our industry away from this absurd notion of “browser support?”

    My suggestion is to NOT include “browser support” as a standard service, but rather to break it out as a separate offering.

    You can price your development projects to reflect the additional time and cost involved in getting browsers like IE6 to render the site correctly (or as close as possible). Basic example:

    Site w/ valid Markup/CSS - $10,000.00
    Fix IE6/other specific issues - $1500.00 (10hrs @ $150/hr)

    Put the choice of “support” back on the client and have them decide if the additional cost is worth it. Once they see how it affects their bottom line, they will probably pay closer attention to the issue and hopefully come away better informed. If not, at least you’ll be compensated for your time.

  59. 059 // Pim Rakers // 10.08.2008 // 7:43 AM

    I did not read all the comments so this is just my 5 cents. Think about this.

    If I get into a conference room with loads of people from all over the world. We make up these rules that we all talk in english, wait for our turns etc. If some people don’t understand english that’s their problem. Why did we make up these rules after all then? Why did that guy even came to the meeting?

    I hope you get what I mean, but the rules should cover all browsers. And all browsers should obey the rules.

    Get my point? Not an answer, just another opinion, love to hear a response!

  60. 060 // Chopper // 10.08.2008 // 7:53 AM

    What we really mean when we ask if a site will “support,” say, IE6, is “will the site look the same in IE6 as it does in the latest and greatest browser?” But we all know this is a silly question. Of course it won’t. And what’s more, it shouldn’t.

    Seriously? It’s not exactly difficult to iron out the obvious bugs in ie6 and get all browsers to render the same and look identical…

    Is this everybody’s opinion here? I agree, the day ie6 no longer needs to be supported will be a great day. But its not exactly time consuming…

    I can usually code a site using firefox for testing as I build it and 9 times out of 10 it will look the same in all other browsers when i’m finished. I’ve just got used to coding for all browsers and know what I need to do and where as i’m building it.

  61. 061 // Billy R // 10.08.2008 // 8:21 AM

    There should no longer be support for a 7 year old web browser. You want an analogy? Imagine bottle feeding a 7 year old. Whether this is a valid analogy, I’ll leave it up to y’all.

    All this talk about “supporting” browsers like users paid for them. Do you have to pay to go to IE7/opera/firefox, etc? Is there a charge to download or upgrade browsers?

    There should be nothing keeping users from upgrading other than fear of glitches, which have been already worked out through updates. As a tech in my company, I’ve watched bugs come and go. MOST bugs were worked out within weeks, not years. IE7 is 2 years old this month and it doesn’t have nearly as many glitches as it had back then. I’ve had only a few calls on it this year.

    Comments?

  62. 062 // walter // 10.08.2008 // 4:49 PM

    MS should have been mandated to re-call all IE6 installations. It can cause deadly crashing or bad visions.
    PS and we even have to recall IE7.X

  63. 063 // Jerry // 10.08.2008 // 8:50 PM

    It is really our own fault. We developers never assert web standards on our clients. We always accommodate the clients microscopic view of the web and write special “fixes” to make them happy. Until our clients meet web standard we have to realize that what we produce for our clients is not the most efficient, elegant, and cutting edge code. Unfortunately, we’re in the business of selling a product not researching and developing the web languages. :(

  64. 064 // Sean Molyneaux // 10.09.2008 // 2:57 AM

    Even though this article is about the CSS part of standards, the dom element coding is even worse when it comes to IE6. I find it incredibly fustrating when building beautiful frontend interfaces, with just the right amount of DOM coding needed, but when doing fixes for IE6 (infact all of IE). The code becomes a big mess of silly work arounds. I feel that IE needs to really fix its numerous bugs OR stop production of the browser all together as there have been to many hours wasted on a project doing fixes for IE’s stuff ups.

  65. 065 // Steven Southard // 10.11.2008 // 3:15 PM

    One thing I notice about making code that looks more or less the same on Safari, FF, IE6&7 is that by the time I’m done with working out the quarks of IE6&7 I always think the code is better. Safari and FF seem a little more generous and forgiving to me. Mostly it seems to be a difference in how IE measures line height, margins and padding. Of course, there are also some differences in how it understands floats. But without fail, once I’m done I think the code is more clear and better describing what it is that I’m trying to achieve. In some cases I discover Safari and FF were letting me get away with errors and basically sloppy code. Having these extra browsers out there demands that our work is near perfect to render properly. I wonder if sometimes it isn’t a blessing to have these different perspectives on standards.

    The notion of forcing standards on our clients or charging more for it render properly in IE is laughable. Most of my clients could care less about standards and if there was one browser they would cut off the list it would be Safari not IE. Making the sites validate to W3C standards has always been for me not my clients. You guys would really leave a website looking like crappola in IE6 because your client didn’t pay you extra? What if your mom visits that site or god forbid a perspective client? Do you think they’ll understand what a good thing you’ve done for the web as a whole?

    I’d love for IE6 to understand min-height and width or to render PNGs perfect without hacks but it won’t. I’d also love for google to follow an AJAX link or to index Flash but it’s been years and there’s no movement on those fronts either. As developers we have to deal with the landscape in front of us and make choices and be aware of the consequences.

  66. 066 // Jeff Croft // 10.12.2008 // 8:09 AM

    You guys would really leave a website looking like crappola in IE6 because your client didn’t pay you extra?

    I’m not sure who suggested charging extra to make a site look right in IE6, but it certainly wasn’t me.

  67. 067 // Jeff Croft // 10.12.2008 // 8:17 AM

    Even though this article is about the CSS part of standards, the dom element coding is even worse when it comes to IE6.

    Very good point. I don’t write a lot of Javascript myself, which is why this article didn’t mention it, but you’re totally right.

  68. 068 // Don Ulrich // 10.13.2008 // 6:21 AM

    Seems to be a legacy issue more than anything.

    Legacy support is part of business culture. If your audience is a secondary market then it becomes imperative that you support legacy. Most of the use of legacy broswers exist in this environment. The idea that my cool visual won’t work in your old browser is nonsense. It’s like telling the customer he/she is wrong but the awesomeness of my design is right.

  69. 069 // Mark Wunsch // 10.13.2008 // 12:15 PM

    I just recently wrote an article about how we can save cost and protect progressive enhancement by NOT having IE6 look the same as other browers: http://markwunsch.com/blog/archi…

  70. 070 // Don Ulrich // 10.13.2008 // 1:16 PM

    @Mark good article but…

    People who value strong visual design don’t care about the extra divs for rounded corners or the transparent issues with .pngs. They want that visual to be correct in all browsers within the cultural norm.

    From a design aspect it is the scope of the culture you are designing for. It’s no fun to say but the larger the audience the lower the common denominator is. That is true in so many ways with respect to this conversation.

  71. 071 // Michael Person // 10.14.2008 // 2:06 AM

    Nice article and i think web standards is here to support developers with a future standard that makes the future developers to easier make websites for all browsers. IE6 is still today used by some 60% of the internet users global and there is no way we can avoidd making websites to work in this browser. That would kill the whole internet idea.

    Compatible website development with webstandards is not difficult be one has to have experience in how it should be done. many new web designers/developers are starting to develop their websites for Firefox and think that everything is fine, until they open the site in IE6 or IE7 that might not have web standards compability to present a perfectly web standards compatible website.

    Always develop a website for IE6, because of the mass of usage, and then check with the other 10 web standards compatible browsers, and never for get the MAC which is a growing user group of internet.

    Web tandards for future and we are still fighting for things to get better and usable for future.

    Michael Persson web standards developer and online marketing consultant

  72. 072 // Bruce Kroeze // 12.11.2008 // 10:39 PM

    @Steven

    You guys would really leave a website looking like crappola in IE6 because your client didn’t pay you extra?

    @Jeff

    I’m not sure who suggested charging extra to make a site look right in IE6, but it certainly wasn’t me.

    I certainly would. I price in a certain amount of dealing-with-brower-deficiencies into the base bid, but if a client insists on finding places to cut, and chooses to drop this feature after being warned about market share, then so be it.

    By default, I use techniques which are clean and standards based. They are therefore relatively easy for me to come in and tweak for tired old IE6 when I’m being paid to do so. But if I’m not getting paid, I don’t even test it once in that monster. It probably looks “OK” due to my default practices and that’s fine with me. If not, then “oh well.”

    For personal projects, especially free or experimental sites, I simply don’t care, and don’t see why I should.

    What if your mom visits that site or god forbid a perspective client? Do you think they’ll understand what a good thing you’ve done for the web as a whole?

    My lack of care for IE6 on personal projects doesn’t seem to have affected my company’s ability to attract new clients nor to maintain current ones. In my (years of) experience, clients are far more interested in portfolio work than in side projects.

  73. 073 // Tim Clark // 12.12.2008 // 9:35 AM

    I think there is a little too much focus on IE6 here. IE6 will be below 5% within 12 months IMHO - but 4 years ago it had 95%.

    There will always be some “headache” browser, so the focus has to be how do you get the browser builders to follow standards? Like the TV analogy from earlier.

    Seems to me that there has to be a sea change in the way the entire industry works. To spout another analogy, in the early days of railway there were massive differences in the technologies each railway company used. Until commercial forces made it imperative that the companies “got along” with each other, nothing changed.

    The only other option that would work is a big stick! Impose standards, rather than wait for people to adopt them. Completely unworkable I know but effective.

    So, in conclusion, until there are pressing economic reasons for all browsers to stick to the standards this will always be an issue.

    Sorry to sound so downbeat.

  74. 074 // heating it up // 12.22.2008 // 1:39 AM

    I used to be a big supporter of IE, now I’m a glory hog and support FireFox

  75. 075 // Steven Hambleton // 12.28.2008 // 9:36 PM

    Perhaps you should highlight what features from your site each browser supports.

    You have transparent PNGs then IE6 doesn’t support your site fully. Want Text Shadow then Firefox and IE6/IE7 doesn’t support it.

    I guess if you want something then you find the right tool to support it.

  76. 076 // Chris Wallace // 12.31.2008 // 9:29 PM

    I say we all get together for a beer at SxSW and complain about IE6 sucking. Then we can put the discussion to bed and move on with our lives and start worrying about things like whether simple animation should be a part of CSS or how to best control and manipulate multiple background images on one element, stuff like that.

  77. 077 // Matt Chatterley // 01.06.2009 // 4:36 AM

    Good post.

    We also talk about “support” for IE6 - although we interpret this to mean “will the site be functional, and useable in IE6?” - and we do not do this for all sites.

    Only those where there is a clear benefit to the customer in undertaking extra work - for others we encourage the use of a warning message suggesting that visitors upgrade.

    IE6 pages don’t tend to look identical to the “same” page in IE7 or the latest FireFox - but when “supported”, they’ll work, without error, and provide the user with an appropriate experience.

  78. 078 // James Creare // 02.13.2009 // 1:54 AM

    Browser Support is a nightmare, Another consideration that I have to take into account, if a client insists on IE6 Support, is that people who have IE6 almost always have a low screen resolution to go with it. I find myself having to work backwards, making a web-site skinnier, removing CSS3 and other luxuries not supported in IE6. Poo.

  79. 079 // Oliver // 02.19.2009 // 9:17 AM

    As I’m a designer I need to make sure that all the websites I develop work in IE 6+7 and Firefox. There are some people who still use IE 6 so everyone in my team has to make sure that each site looks the same in IE 6 as it does in the other browsers.

    Occasionally we have had problems with getting a site to display correctly in IE 6 but have managed to fix it. Generally people who use IE 6 do have lower screen resolutions but this is rarely a problem for me and sites i have developed have worked in all browsers.

    Sometimes it can be a nightmare because I have to use trial and error with the CSS code but eventually it works.

  80. 080 // tommy // 02.24.2009 // 9:28 PM

    When I first started fiddling around with websites, I made a point of testing them on all major browsers I knew off: Netscape, IE, Opera, Firefox. And yes, it’s a headache getting stuff look right in all of them, which is rediculous, since it’s just for that reason webstandards exist. With experience came an inherint knowledge about how to layout a site, so that there will be no potential for trouble, what extra css lines to use in order to avoid text clipping and the like…

    By now I put my sites together for FF, and check they look okay in IE. If that works, I’m happy.

  81. 081 // Jen H. // 04.11.2009 // 9:05 PM

    What does it mean to support a browser? I’m going to use an analogy… A bra, a crutch, a jockstrap, all are meant to be used as a support of some kind. Is CSS akin to a bra, a crutch or a jockstrap? I think no. CSS, along with HTML, is like the foundation to a house, it’s the concrete that holds the whole thing up. That said, it doesn’t matter how sturdy that concrete is, if the house, in our case, the browser, is poorly built, no amount of concrete is going to keep that abode from leaning or even falling down.

    How we get clients, who are actively browsing the web using a leaning house, to understand that their house is about to fall down around them and all we have is a bit of duct tape to patch it together is the tougher sell that I don’t have an answer for. A pretty spiffy analogy but no answer. :)

  82. 082 // cooper.semantics // 04.19.2009 // 8:23 PM

    Progressive Enhancements :-D

Tags for this entry