A few days ago, I was given the task of developing a very quick-n-easy website for a company. They probably wouldn't even have thought they needed one at all, had my employer not convinced them that this internet thing is the way to go forward. So, we're talking about a very basic company presentation, a two-day project, from graphic design, through HTML/CSS-implementation, to final touches and finish.
A dirty job, but someone had to do it. I figured I probably don't have time to develop it using XHTML and CSS, I imagined the client would probably draw a blank stare if I would have mentioned "accessibility", "standards", or "other browsers", writing me off as some kind of perfectionist fool from another dimension (and boy, would he have been wrong or what?).
So back to the old school of table based design it was, this one time, at least. Unfortunately, I was pretty certain that there would be more of these super basic company presentations in the future, requiring me to work perhaps as much with standards-based design as with the slightly more painful table-based variant.
Having done a rather simple, straight forward design, not particularly striking but nothing to be ashamed of either, I started implementing it using tables for layout, and CSS for everything else. I developed it for Internet Explorer only, but since the behaviour of table-based design is more predictable in different browsers than the CSS-based dito is, I figured it would probably mostly look ok in other browsers as well. I'll have a look after I'm done and do some final final touches and make sure it looks representable in Mozilla as well, I thought.
The rest of the day was spent reacquainting myself with quirks-mode. I had the website done just about on time, with perhaps an hour or so left before it was to be presented to the customer. They liked it, I was happy that the pain I had to endure developing it using table-based principles wasn't wasted.
Just a few days later, I got another one of those really basic online company presentations to do, only this was to be even more basic, a one-day project. Just one page or so, maybe two. It ended up being three, or four, then six, maybe eight. But it was still a one-day project.
This time, I had the memory of developing the two-day project fresh in memory, and thought that table-based design didn't actually feel like it took a whole lot less time than CSS-based design does, so I figured I'd try using CSS based layout this time around instead, in spite of the very tight deadline.
Since I began using CSS for layout, instead of tables, I've always thought to myself that developing a website using CSS probably does take a little while longer than using tables, but the extra time spent is worth it many times over due to the manageability of CSS-based layouts.
Today I know better. My recollection of table-based design must have been foggy, because I didn't remember the quirks involved using it, or the many CSS-powered tools and tricks that the table-based counterpart lacks. I re-realized, in particular, that the image-slicing, table-gluing approach common in table-based design (memories of which I had buried in the deepest darkest corners of my mind) is a shitload of more work than doing the same thing using the CSS-based approach.
Table-based layouts do not take shorter time to develop than the CSS-counterpart, it just feels that way because you spend less time on creating the actual layout structure, and more time doing other things that the CSS-based approach gives you more or less for free.
It took me less time to develop the second website using CSS-based layout, for both IE and Firefox, than it took me to develop the first site using the table-based approach for IE only.
Needless to say, it was with a very big smile that I said good-bye to table-based design this week, this time for forever and ever, under any circumstances, regardless of how tight deadlines I may face in the future.
Permanent link
View/add comments (7 comments so far)