LibData: Creating a Database Driven Web Site

1
LibData Creating a Database Driven Web Site

• Shane Nackerud
• October 2, 2002

2
Remember 1995?

• Batman Forever was the 1 movie
• Republicans touted their Contract with America
• Gopher pages ruled the Internet
• A little thing called the WWW was picking up speed

3
Why flat HTML stinks

• Too easy to produce bad code
• inconsistencies run rampant
• Pages are entities unto themselves
• links are hard coded into the page
• Flat HTML is very time consuming
• especially for the webmaster
• Flat HTML is not dynamic
• does not change based on user request

4
Why database driven web sites are nifty

• Makes site maintenance much easier
• Changing a link changes it everywhere
• Provides patrons with consistent and up to date
  information
• Allows for multiple ways of displaying
  information
• use the same database to make a MyLibrary site as
  well as drive a list of indexes

5
How?

• Typical database driven website consists of three
  layers
• Interface layer -- the HTML code and graphics the
  system spits out to the user
• Database layer -- stores the information
  pertinent to the library web site
• Integration layer -- makes the contents of the
  database visible to the user
• Perl, ASP, PHP, Java, etc.

6
The U of M Libraries

• 1996 - 1997 -- University Libraries develops
  Research QuickStart
• http//research.lib.umn.edu/
• 1999 -- Libraries begin to reuse QuickStart data
  to drive indexes pages, subject pages, and other
  divisional library pages
• 1999 - 2000 -- Libraries begin developing other
  databases for hours, electronic journals, FAQ,
  CourseLib

7
LibData

• Goal Make the system more flexible and allow
  for more data reuse
• Goal Make the system more user friendly
• Goal Make the system more customizable
• Goal Make the system more integrated