Hi, Could somebody tell me the difference between client pull and server push? As far as I can figure out client pull is when we send a GET request via a browser to fetch information from a server...but I can't figure out server push.
"Tell me and I forget. Show me and I remember. Involve me and I learn."
A client pull is when a client program solicits data from the server. A server push is when a server sends unsolicited data to a client. An example of a server push would be an application where a client registers interest in a particular data item (lets say a stock symbol) and the server pushes data (such as stock purchase and sell data) as it occurs without the client having to continuously ask for updates.
FYI - note that a form of server push (such as MultipartResponse, from o'reilly) is not supported if you use Internet Explorer browsers as clients. A situation would be when you want to change the header content type twice with one request. For example, you push an application/octet-stream to force a file-save, then you want to display html on the browser afterwards stating "finished!" by switching to text/html... all with one client request. Fails badly in IE. --- [ April 24, 2002: Message edited by: Toby Davis ]
And just to make it formal: under HTTP, you can ONLY do client pull, not serverf push (sneaking headers into the middle of a pull's response stream isn't really a push, since a pull already initiated it).
An IDE is no substitute for an Intelligent Developer.
Originally posted by Tim Holloway: And just to make it formal: under HTTP, you can ONLY do client pull, not serverf push (sneaking headers into the middle of a pull's response stream isn't really a push, since a pull already initiated it).
Correct, if you need to push info to a browser you should use an applet or Flash.
Joined: Mar 05, 2002
Thanks for all the info... So is it possible to incorporate server push using Servlets ? :roll: