Loading Pages Automatically
Using Client-Pull 


Definition
Client-Pull refers to a technique that enables a browser/client to request (or "pull") data from the server automatically without the user-intervention.  

The key property here is "without user-intervention."  If not for  this attribute, then the term is meaningless because it will be no different than when an user enters an URL to request a page.

With this capability, it is possible to have a web page that refreshes/reload itself; or even a web page that loads another page, without the user doing anything.

How To
Client pull can be done via cgi, by outputting this http header 

Refresh: numOfsecond; URL=someURL

But the good news is that it can also be accomplished by using META tag within a html page.  The syntax is as follows: 

<META HTTP-EQUIV="Refresh" 
CONTENT="X[; URL=http://whatever.com/something.html]">

where:

  • X is the number of seconds before the reload/refresh; and
  • URL is the URL of the page to be loaded.  Note that according to Netscape, this should be a fully qualified URL (complete with the site's domain name); although I've seen it work with relative URL.  Also, this doesn't actually have to be a html page, it can be an image, or whatever.

Note that the URL field is optional.  

Example 1:
This html code below will cause the current page to reload itself after 10 seconds.  It reload itself because no URL is specified.

<HEAD>
<META HTTP-EQUIV="Refresh" CONTENT="2"> 
<TITLE>Page</TITLE>
</HEAD>

Example 2:
This html code below will cause page2.html to be loaded after 2 seconds.  Note that there are only 2 quote signs in 
CONTENT="2; URL=http://whatever.com/something.html"
This is not a typo.  It's the way it is specified in the http specification.

<HEAD>
<META HTTP-EQUIV="Refresh" 
CONTENT="2; URL=http://www.permadi.com/page2.html"> 
<TITLE>Page</TITLE>
</HEAD>
 
 
Example 3:
In the following concrete example, client-pull is used to create splash screens effect.  Three html pages will be loaded subsequently after 2 second.

Why would anyone want to use this?
At the present time, client pull has been supplanted by newer and more powerful technologies such as "animated-gifs," Flash, or "dynamic html."  However, on its days, it has been used for:

  • Web-cam applications, such as web page that shows a real time image every some minutes.  (This is perhaps one client-pull application that is still relevant today.)  
  • Automatic slideshows
  • Splash pages
  • Gimmick, novelty

For further reference, consult Netscape documentation at http://home.netscape.com/assist/net_sites/pushpull.html

<< INDEX >>

    
(C) F. Permadi

10/01