On Sunday 11 April 2004 04:28, John Andersen wrote:
On Saturday 10 April 2004 23:28, Curtis Rey wrote:
The client (in X) is "the remote machine where the application is running", and the server is the local machine that displays output and accepts user input. Counterintuitive on most counts - since everywhere else in the IT world the "server" runs the app and the "client" is where user input and output is generally done.
I really think this is why so many companies avoid Linux/'nix on the consumer end - it runs bass ackwards from a programmers sense.
Add to this all the nebulous layers and arbitrary divisions of labor (X windows, Display managers, widget sets, GTK, KDE, Gnome, fwvm, xfce... It goes on and on.
Remember these "nebulous layers" are what make the system stable and robust and give you the choice to use what you like. This is how things are separated in process space as well as divisions of labor for building them. This is why a misbehaving application can't take out the entire OS or even the entire GUI. As for the backwards portion that Curtis wrote about, who cares which is the server and which is the client when the system is working? It is a display provider and an application provider. It doesn't matter what they are called, end users would never call them the right thing anyway. They could never tell you that the app server was down or the database server was down, they can just say that my "xyz" isn't working today. Or my drive "w" isn't there. -- Kelly L. Fulks Home Account near Huntsville, AL