The History of Din Timelines Backend

Here is the big picture. At its inception timelines had basic html pages, each page with the items directly on it, including all the links in the text (inline links) which were hand written. Next came a perl/ssi version that lasted several months, each inline link was changed to an SSI in the shtml pages that called a little perl script and an flatfile database. This worked fine until my host gave me a friendly CPU usage warning, which was caused by the banner script I was running, but I thought it best to change all those perl calls.

This brings us to the stage where I wrote a perl script to run at home, revamped the database structure, added an editor and did all sorts of fancy things. I had templates of all the pages, each page with all its items with markers for the inline links in them. Id run the script and it would run the templates through the parser and replace the markers with the links and then I would upload these finished pages. Additions to this included building a system to place ads in the side menu based on the page size.

Then two years ago I had a major harddrive crash and lost all of everything. So i found a newer version of hte home server I had, with php and started rewriting everything in php. it wasn't long before I installed mySQL and started playing with it. placing the inline links in mySQL and the actual timelines in an XML file that would be parsed into the template, which is now just formatting without any information except for what information needs to go into it.

From this point on everything will be in mySQL except for the smaller flatfile DBs with the navigation information between timelines. In the future I;m looking at further abstraction between content and formatting.