Killer Tables - Overview
When Netscape added tables to HTML in 1995, second-generation site design became possible. Although tables weren't meant to contain lots of text, people soon found they could use them to get margins and do better typography. This section discusses the thrills and dangers of using tables to do real typography.
The framers of HTML had in mind a syntax that would let us put up tabular data (see pictures). Tables can be very difficult to build, so the idea was to make tables "smart." A smart table can see everything in it and try to balance things out, spreading them evenly across the page.
While this is good for simplistic schedules and calendars, it isn't very helpful for much of what we want to do. As implemented, tables try hard to help you solve your layout problems. Our goal is to turn off as much of this "intelligent" machinery as possible, because it often works against the page designer. Our second goal is to turn off table borders, because they are extremely ugly and a blight on the Web.
TIP: Use table borders to develop and debug your tables, then turn them off. The best border to use is one pixel because it's the thinnest and causes less movement later when you turn the border to 0. Never leave table borders turned on! Use them only for debugging!
Until we enter the next generation of site design, the TABLE tag will have to be our friend. The more we learn to control tables, the more we will be able to break the grip of the old HTML and make our pages look good. If we have to put all our text into tables -- not really what the framers of HTML had in mind -- we will.
For exact technical information on HTML tables, try the following:
PC Magazine Reference
Microsoft Internet Explorer Tables
Next: Table Sizing
Core | Top | Feedback