Validation of user queries

 

Building user interfaces for the World Wide Web means using the Hypertext Markup Language (HTML) to write the needed WWW Pages (Hyperdocuments). But HTML documents are of a fairly static nature. You can not implement any kind of really interactive interface between the user (client) and the DB System (server) by only using HTML.

The mechanism of encoding `electronic forms' in HTML only permits sending back feedback information in the sense of unvalidated client input. The syntax validation and semantic interpretation of this feedback has to be performed by the server.

This implies that there are at least three interactions between client and server to correct a syntactical or semantical wrong client query:

1.   Server sending the electronic form.
2. Client sending back the user's input for that form.
3. Server validating client's feedback.
4. In case of errors start over at 1.

To overcome this WWW standard problem (i.e. validation of user input) Netscape-- the leading software company for selling WWW browsers-- has invented the supplementary WWW language 'JavaScript' that provides an easy access to HTML forms while being resident on client computers.

With it the interface designer can distribut all input validation mechamisms to the client machine. In an ideal world scenario the feedback will then be 100 % correct when send back to the server.

This reduces the server load in to ways:

1. The server does not have to compute the input validation (appart from some worst case traps).
2. It does not have to send back a correction request (including the interface again).

N.B.: Having gained a wide acceptance 'JavaScript' has addoped many more feature for client-side manipulating HTML documents. Like controlling the document's layout, the browser environment, etc. But all this features are still primarily implemeted to enhance the possibilities of WWW interfaces.