A while ago Marko wrote about reusability of css based designs. I didn’t agree with him fully but had no idea how to explain/prove it.
In the times of table layouts the process od redesigning a site was actually (at least) three things. First the content was restructured and some of it written again (the sitemap). Then the layout was changed to reflect the new structure and to better fit the content (the wireframes). At the end a façade was made for reflecting the latest trends in ‘webdesign’ (the design – or is there a better word for it?). Even if it was a three way process it was rarely done by three different people – ‘webdesigners’ did almost everything themselves. Since this was quite a lengthy process it was also expensive and some corporate websites changed once every three (or more) years.
Now it seems things are changing for the better. With a DIV layout, CSS based designs and a strict separation of layout and content we can actually split these three into separate stages. The strict division of these stages gives us the posibility to focus our work on what it really has to achieve. It’s true that you cannot use the whole 100% of the other parts but being able to keep 80% is good enough.
About a week ago I found out about two ‘design’ contents and the one Slashdot is holding was interesting because it was about proving a point – you can change the façade without changing the layout or the content. I did the whole thing without changing a bit of HTML (with a bit ugly CSS, would be nicer if a few lines of the code were changed).
I think this way of thinking will settle in and more and more changes will include only one of the three stages. Hopefully they will be based on user testing and/or usage data…
Yes, but you didn’t change HTML because you simply couldn’t.
You’ll obviously work from existing XHTML templates if your client demands that and I absolutely agree that good templates make such work possible. But that wasn’t my point.
What I wanted to know is if you do it otherwise? Do you look at templates first and try to fit design to it when this isn’t expected of you?
I have a sneaky suspicion that you don’t and that not many people do. And that was my point. There are plenty of reasons to do CSS based designs, but reusing existing templates might not be one of them.
On a big site with a fairly complex CSS it can simply take more time to understand previous design (and that’s work providing little fun) than to whip up your own and I don’t see this changing anytime soon.
Of course, there will always be exceptions.
My point is that the technology allows it and more and more companies/people are aware of it which makes it more likely to happen. The thing is that it comes down to the mentality of the caretakers.
When it comes to personal experience my decision is made on the current state of the html code – if it’s clean I just change the styles. It’s way faster. On the backend I strive to make blocks as generic as possible to allow this.
No, it comes down to cost. Unchanged templates aren’t some holy goal we should aim to achieve. I don’t reuse old nails either.
I think keeping templates might make sense only when doing so is cheaper than not to. Certainly by chosing to keep them we add additional limitation to our designs.
I’d like to believe that with each year there are fewer CMS’s around that would make reusing templates a necessary or even appealing idea.
That’s what I was saying – sometimes when a used nail is straight it’s cheaper to reuse it. I think another important thing is that a complete redesign is a ‘huge’ project that companies often find hard to back with arguments.
I could make a parallel with cars – you get a new model and in a few years time you get a ‘cosmetic’ change. You might get a few other changes if the car is still selling. Only after it’s life cycle cannot be prolonged anymore the company decides to kill it and launch another one in its place.