home *** CD-ROM | disk | FTP | other *** search
- Submitted-by: peter@ferranti.com (peter da silva)
-
- I FTPed the Palladium papers. I haven't looked beyond the introduction to
- the design document (I don't have a workstation at my desk and it's a LOT
- of paper to pump through our lone LN03R) and the Usenix paper. The thing is
- very much like lpr on steroids, with all the dubious bells and whistles
- and hardcoded magic names and numbers (sgtty flags! dvi filters!) that lpr
- suffers from. It's *not* a generic queuing system (as someone implied) and
- its main advantage over BSD is the centralised administration. Compared to
- the rather general System V mechanism (where ephemera such as particular
- file formats are handled in a general mechanism), it's a step backwards.
-
- The shortcomings Palladium is intended to address are:
-
- 1. Spool and administration files are duplicated at each
- print location. This increases the administrative load
- geometrically as the number of clients and servers grow.
-
- 2. Authentication and access controls are very primitive.
-
- 3. Communication between printers and clients is one-way.
-
- The shortcomings that remain are:
-
- 1. Details of operating system and application file formats
- remain embedded in the administration database.
-
- 2. Destination services are still assumed to be printers. For
- example, the return channel is assumed to be narrow (mail,
- zephyr) and human-readable.
-
- It's an advance over lpr, and the administration files are likely to be
- pretty familiar to people used to printcap. But I don't see that it's
- enough of an advance to justify using it as the standard.
- --
- Peter da Silva `-_-'
- Ferranti International Controls Corporation 'U`
- Sugar Land, TX 77487-5012 USA
- +1 713 274 5180 "Zure otsoa besarkatu al duzu gaur?"
-
-
- Volume-Number: Volume 30, Number 54
-
-