Should we abandon rounded corner techniques that require additional HTML markup in favor of emerging CSS techniques that are not yet supported on all browsers? Apparently, the answer is an overwhelming YES.
I’m very fortunate for having attended An Event Apart in Boston this week. I talked to thought leaders, design experts, tech gurus, and many more interesting and successful people during the event. I’m accustomed to being the only “UI guy” in the small organizations that I’ve worked for, so it was nice to meet other people who do what I do. One of the most interesting things that I took away from the event was the fact that design leaders are strongly pushing for web design to move in a new direction (well, I’m not sure if it’s fair to call it “new,” but the push is stronger than ever before).
What I mean when I say this is that many designers are asking you to stop using rounded corner techniques that require extraneous HTML (use the CSS property border-radius instead). Stop using semi-transparent PNG images for background colors (use rgba instead). Stop doing all those little extra things that you do in order to make sure your site is pixel-perfect on every single browser and operating system. Like I said, it’s certainly not a new subject to bring up, but I was surprised at the ferverous support behind it at the conference. I had briefly touched on this subject in my post about CSS rounded corners, where I seem to have agreed with these experts on a number of levels. My opinion is that you should never choose to increase page load times for everyone just simply to ensure that IE users can see rounded corners.
I think Dan Cederholm said it best when he referred to these little design touches (such as rounded corners) as “design rewards” for users who are using the most common OS/browser combinations. Don’t think them as design features, but rather as rewards for users who can utilize them. This makes your site faster, more flexible, and easier to develop. Dan is so passionate about this concept that he created a website to convey his feelings:
Do websites need to look exactly the same in every browser?
I’m not saying we should do away with rounded corner techniques entirely. In fact, I use my own Karate Corners all the time, but I do it when it’s necessary and appropriate. If I’m working on my own project (where I am the primary stakeholder), I will most certainly go the pure CSS route, but as designers, we know that our clients don’t always agree. The most prominent argument against this concept is that our clients don’t really understand the benefits of using pure CSS to create corners and other design “rewards.” The only thing the client can understand is that Mr. CIO uses IE6 on his Windows 2000 desktop and he wants to see the rounded corners that were shown to him in the design concept.
It’s a difficult thing to manage client expectations, and sometimes it’s necessary to bite the bullet and give the client exactly what they want, even if you disagree with them. The best thing we can do is to make the benefits of this technique clear to the client, and hope that they can let go of the aging concept that your website has to be identical on all possible combinations of platform and browser.
There’s certainly a discussion to be had, and I’d love to hear your thoughts on the subject. Let the debate begin…
I believe in the principle of CUSTOMER DELIGHT.
If the client is Mr CIO who has IE6 on Windows 2000, and he wants rounded corners, and it is possible to give him this… then Give It To Him and Everyone Will Be Happy.
Why should he be told to change his Operating System and Browser and what-not? On his side, this implies additional Expense, Effort and Adjustment to his new environment, which definitely DOES NOT equal customer satisfaction.
Instead of making his life easier, you are making it more difficult for him. What’s more, you are painting a picture of HIM as being the “difficult idiot with unreasonable requests” instead of considering and accepting his constraints. And, you are doing this just to make YOUR life easier AND FEEL JUSTIFIED IN DOING SO.
I would create the non-IE6 etc. version – but also create a version for him. At pageload the browser particulars are detected, and this version is served if it happens to be Mr CIO, and the light-weight version for everyone else. Therefore no-one needs to incur bandwidth penalties because of him, and he is also deliriously happy. Now everyone can experience the nice new design, which was designed to be nice so that everyone can experience it that way – especially the customer!
My opinion: Our priority as developers isn’t FIRSTLY to find ways to make things easier FOR US. It is to find solutions for what the client wants. THEN we can see how to do that in the easiest way FOR US. Why? Because that is our job. We are hired/paid to make things easier for our clients.
Sometimes it is necessary to “go the extra mile” to honour this principle. At the end, everybody is happy and we can all go and have a drink in the company lounge without dark glowering looks and enclaved conspiratorial conversations.
Stefan, I definitely agree that there is a discussion to be had about this concept, and you make a lot of good points in support of your argument. I also agree that sometimes you simply have to give the client exactly what it is that they want, whether you agree with them or not. I’m not trying to paint a nasty picture of client expectations, but the reality is that our clients pay us because we are the experts, and we can provide them with the recommendations and best practices that they seek. That means it’s up to us to push this new concept into the forefront of thinking, so that it can be addressed and at least acknowledged.
I have to say that I don’t agree that we are somehow making it “harder” for users (like Mr. CIO) to use the websites that we’re creating by disallowing them the pleasure of rounded corners and other superficial design elements. The fact is, we’re not degrading anyone’s experience by removing these elements of design. The site will appear and function the exact same from user to user; the only caveat is that someone using an older browser may miss out on some amount of polish within the design. You are absolutely right in your suggestion that this is unfortunate, but I think the benefits of designing in this fashion greatly outweigh the disadvantages of choosing a heavy and more cluttered path.
I’m not telling anyone to change their web browser or to upgrade their operating system. In fact, I’m doing just the contrary; I’m making sites that are viewable and completely functional on all systems and browsers. I’m designing for web browsers of the past, present, and future. Furthermore, sites can be designed to be even MORE flexible and compatible if you design them in a simpler, more standards-compliant way. That means there is less of a chance that Mr. CIO will see something he doesn’t like.
In your example, you’re saying that “going the extra mile” makes things harder for us, but easier for the client. I fervently disagree. You’re actually making it harder on your clients by increasing page load times and cluttering your site with HTML markup that isn’t part of the semantic structure of your document. You’re making it harder for your users by designing a site that isn’t structured in a SEO-friendly manner (thus making your site harder to parse and find on search engines). You’re designing your site in an overall manner that actually makes it HARDER on your users. You may not see it, but the effects are there. It’s about the overall user experience, not just the on-the-surface appearance of the site. It’s time to stop thinking about our sites as pixels, and rather thinking about them as an experience for our users.
I will say again that there is a discussion to be had, and I’m sure this opinion will be met with an arduous defense by many designers. I will also say that I think this is a very good thing. This is a conversation that has needed to happen for a long time now, and I’m glad it’s finally out in the open. It’s all about weighing the advantages and disadvantages of each approach, and making an appropriate decision based on what your users (and the primary stakeholders of the project) really want.
Hi Kyle, thank you for your reply. I understand your points. I believe they are all valid – but in a specific context. Just like my points.
What I mean is, all these variables are interrelated and trade-offs are the name of the game. Solutions and suggestions take the form of “Yes you can have rounded corners if you have IE6 and Win 2000 but you will have to payzx amount more, wait y% longer for project completion, and pageload times will be y% slower than if we did it THIS way where everyone else except you can see the corners.”
My main point is, the customer decides what is essential or not. WE may think it is ridicilous to prioritize a design aspect above functional aspects… but just who do we think we are? HE decides what is important for him! Our job is to point out the possibilities and implications.
Which is why I think everything ultimately depends on the initial and final requirement specs.
This is where the client has the responsibility to come up with absolutely crucial requirements and not-so-crucial requirements, and understand what the implications of these are (which is our job to enlighten him about). So that he knows exactly what to expect.
But the point is, the customer ultimately decides what happens – whoever likes it or not, or who thinks it should be different for whatever reason… doesn’t matter.
My sister recently built a new house and I can tell you it was a real battle of wills between the builders and the interior designers. The builders ignored the directive “consult the designers where the light-fittings should be”. When my sister came to check one day, it was found that the design which she had agreed on with the interior decorators, could not be accomplished anymore.
Since the agreed design was the design she wanted, and it is possible to achieve it, since she was willing to pay money for the “irregular” placements and extra material, the result was…
She refused to accept the situation, making herself VERY unpopular with the building crew. She insisted they rip out their work and follow the letter of the agreement. And in my view, rightly so.
They thought she was idiotic, for instance, for insisting that plugs be recessed into a wall so that security monitors are flush with the walls, when this had no impact at all on functionality and created a lot more work and also problems to solve. But I can tell you, little things like that DO make one helluva difference to the end user experience.
You can “live” an experience which isn’t QUITE what you wanted (hmmm…I’m not REALLY delighted but it’s ok I guess…), or …you can get EXACTLY what you wanted – as far as is possible.
Stefan, you’re absolutely right; the customer is the bottom line. If they’re not happy, no one is happy. Tailoring to the customer is priority one. I am very interested in what everyone else has to say…I look forward to further discussion!
Hold your horses… I’ve been pondering, and much to my surprise, I now somehow think… maybe you are right… maybe there is a “best solution”. Maybe it just IS kind of ridiculous to want rounded corners when your page takes an eternity to load. Maybe that IS an aberrated situation which we should not sponsor or be part of. Who says a client’s every abberrated request is the law? Shouldn’t the client’s requirements be compared to the optimal solution? Why should the optimal solution be the thing that suffers, just because the client wants to play God?
But then… how the heck do we recognize this optimal solution…
Well – maybe we can “feel” it when it’s “just right”.
I think it is worth pointing out once again that everything is somehow related to everything else. Whatever we do has some kind of implication.
For example, if you smoke, you probably build-up a residue of tar in your lungs, this no-doubt detracts from the efficiency of your lungs. This in turn spawns a chain reaction of effects in your other organs, attempting to adjust to this change in balance.
The body is remarkable but sometimes the result is that some organs have to start doing more than their share of the work and can become exhausted. Then something else has to start compensating… until one day maybe your big toe suddenly finds itself in the position where it has to act like a heart… because the heart is too busy being a liver… the toe throws in the towel, and you get heart failure. And everyone wonders why the heart medicine you took religiously for years, didn’t work. Well, maybe if you just didn’t smoke without taking any medicine, it would have been the optimal solution, instead of smoking AND taking the heart medicine…
I guess it’s a bit like a tight-rope walker who tries to stay in balance, if one foot loses it and slips off the rope, the weight of the body gets shifted, but he can regain balance by sticking his arms out the other way. So you get two pictures, both representing “balance” – just different solutions to the same problem – one just more aberrated than the other one.
Sorry for hogging this discussion… I will now leave the stage…
[...] all browsers. Whether it’s time to stop using HTML and images for rounded corners is still a matter of debate. Nonetheless, I think every designer finds themselves doing this from time to [...]