I originally started this post in May of 2006, intending to use the libraries to demonstrate how HTML tables can contain multi-dimensional data, and use the seldom-used HTML elements like
TFOOT to generate aggregate data about the table. This was going to be the follow-up to my tables and data with CSS post. I had a rough draft sketched out somewhere: a table full of numbers, dates, currencies and strings. A drop-down menu and a “format paintbrush” would let you reformat it all on the fly, and it would all be generated from semantic information attached to the table cells, not hard-coded into the page.
This was only practical because of the efficiency of my libraries; to reformat entire date regions in the table in real-time, for example, you’d need to parse the value as a date in one format, then reformat it for output in another. It was to be a showcase of how much efficiency matters for some things.
Tangent: I suppose it’s less important for people who aren’t still running 500MHz laptops these days, but efficiency really matters for me; a lot of these flashy sites these days simply take too much CPU for my little old computer to run well. I stubbornly resist getting a new computer because I cringe at the thought of the environmental cost, but I’m slowly breaking down; it’s gotten to the point my battery won’t charge, and Dell doesn’t even have a record of my service tag anymore. Spare parts for these things are long since unavailable.
Who knows – someday I may end up building some browser GUI systems again and use these. In the meantime it’s encouraging that they remain useful to people.
Note: This episode is pre-recorded. I’m taking a short hiatus from blogging and will respond to your comments when I return.