If your web browser formats this page incorrectly, try viewing the page source. In netscape 4, click the view menu, then click page source. The normal way to print in linux is to pipe the data to lpr. lpr communicates with lpd. lpd should have been started from the startup scripts and should be running in the background. lpr/lpd decides what printer to use by checking the lpr command line options, by checking environment parameter PRINTER, or by using the first printer listed in /etc/printcap. lpr/lpd decide whether or not the user is allowed to use the printer. lpr/lpd pipes the data through the filter, which is probably called magic filter or mf. The filter examines the data and determines the data format and decides what conversions to perform. The filter converts the data by piping the data through other filters. Usually the data is in postscript format. If the data is not in postscript format, but is in a recognizable format, the filter will probably convert the data to postscript format. Once the data is in postscript format, the filter uses a printer driver/filter like ghostscript to convert the data to printer format. If the filter does not recognize the data format, then the filter probably does not convert or change the data at all. lpr/lpd stores the printer data in a temporary file in the print spool directory /var/spool/lpd/something. lpr/lpd sends the printer data to the printer port. When you install a linux distribution, it should install lpr/lpd and put the command to run lpd in the startup scripts; it should install standard print filters like ghostscript; and it should install and configure the drivers for the ports which connect to the printer. Assuming lpd, ghostscipt, and the port drivers are ok, to install a printer you need to add the printer data to the lpr/lpd configuration file /etc/printcap and create the spool directory. lpr/lpd needs to know what port device to use to send data to the printer, and which printer format to convert the data to. This information may be stored in /etc/printcap, or it might be stored somewhere else, where it can be accessed by the filter. Try using the printer configuration utility included with your linux distribution. If the computer is only used by you and you never print more than one thing at a time, then you do not need to worry about which users are allowed to use the printer, and you do not need to wait for the previous print job to finish before starting the next print job. Instead of using lpr/lpd, you could create a print script which does something like 'gs -sDEVICE=cdj500 -sOutputFile=/dev/lp0 $1'. The script could prompt for what file conversions to perform. From a program, save to a postscript file instead of printing, then use the print script to print the file. This might be easier than setting up lpr/lpd. However, most linux distributions automatically set up lpr/lpd, so it is probably easier to use lpr/lpd.