I’m considering a bit of writing about what it means for [professional] designers to ‘go with the grain’ of CSS layout. Why? there’s an interesting bit of aesthetic/technical thinking that goes on in the mind of someone (say, me) who intends to create a layout for a customer using CSS, and who has some idea of the things that will be successful and straightforward to achieve in a sensible amount of time and the things that won’t. This inevitable bit of thinking happens during the phase when the designing arguably should best be done with a pencil and against the advice of those who feel that technological constraints are best ignored at the first design stage. They are right, firstly because designers should keep their horizons broad even at work and secondly because no customer wants to know that their uniquely exciting new site has been tramelled into a humdrum and conventional design straightjacket because of mere technology. A design of that kind is (hopefully) hard to slip past internal QA.
Some of the things that work well in CSS are visually successful and some aren’t. This is very evident in the layouts typically used by the web illuminati and generally restricted to blogs and social software – the cool end of the web. You can see the strengths exploited and the weaknesses sidestepped. There is an equivalent to the ‘grid’ of paper design[
0], a semi-formal rubric that gets the designer started with reliable use of space, colour, size and weight contrasts in an environment where 1. there is no ultimate control over any visual design parameter and 2. concepts like ‘parallel’ and ‘column’ become expressions of hope; and this grid-equivalent, usually either with confident, strong colour contrasts or very subtle colour, is fundamental to most such designs.
These layouts start from a position of knowledge; the designers (sometimes working for the community, sometimes for relatively savvy social software houses) have realistic expectations and, paradoxically, the desire to push the parameters as far as possible. Probably this is because CSS, with its restricted set of tools and parameters, sets the bar for anyone to create designs democratically low, but by the same token for anyone to create something visually attractive with this simplistic equipment the bar is high. Basically, it is online stretchy Lego; and this is reinforced by the list of things that unaided CSS is unable to do, such as anything that is not rectangular, straight-cornered and parallel to the lines of pixels on the monitor. To be honest, real Lego is a lot more capable than that.
There is also the fact that to get something really good from the limited resources of CSS is a serious challenge and success is something to show off about. The adoption to genuinely useful inline dynamic replacement of content (‘AJAX’), which works with the grain of the DOM and CSS, considerably increases the opportunities. This though forces the discussion from static visual considerations to the whole user interface, which has change-over-time elements that you can’t really put in a grid, at least not a Cartesian one.
I’m sure all this is analysed elsewhere (I do mean analysis, not complaint, and not the CSS lists) but I haven’t found that place. It doesn’t seem to be anywhere in, for example, Eye magazine, which is the kind of place it should be. Any pointers, anyone, or should I keep going? Computer Arts? Academia? the year 2000? One obvious starting point would be this from 2001:
http://www.alistapart.com/articles/journey/.
Or this, from 2002:
http://www.alistapart.com/articles/flexiblelayouts/. And how bittersweet A List Apart articles look a few years down the line! From the second of those: ‘Think of these [proposed design methods] as a bridge to that point two years from now when all designers are trained to think natively in CSS.’ What happened in 2004?
Footnote [
0]. A List Apart is fertile territory for harvesting information about what web designers think, of course. In 2005 this article appeared:
http://www.alistapart.com/articles/outsidethegrid/. But as comments revealed,
http://www.alistapart.com/comments/outsidethegrid/?page=7, there was an interesting tension between the code view of the world (broadly, ‘designers are using the grid because html tables made them do so’, the author’s contention) and the design view (‘designers are using the grid because it is a generalised and effective tool they brought with them from print’). And of course though there is plenty of potential for ‘deconstructed grids’ it will be up to the designer’s clients to decide whether they become a common sight. We’ve been there before.
