The feel of a LispM/List of running machines
Luca Pisati
pisati@nichimen.com
Fri, 02 May 1997 15:34:28 -0700
Alaric B. Williams wrote:
>
> On 1 May 97 at 13:28, Mike McDonald wrote:
>
> > To: alaric@abwillms.demon.co.uk
> > Cc: lispos@math.gatech.edu
> > Subject: Re: The feel of a LispM/List of running machines
> > Date: Thu, 01 May 1997 13:28:13 -0700
> > From: Mike McDonald <mikemac@titian.engr.sgi.com>
>
> > >From: "Alaric B. Williams" <alaric@abwillms.demon.co.uk>
> > >To: Martin Cracauer <cracauer@cons.org>
> > >Date: Thu, 1 May 1997 20:24:48 +0000
> > >Subject: Re: The feel of a LispM/List of running machines
>
> Hey, same mailer :-)
>
> > >I'd be most interested to know how the LispM dealt with:
> > >
> > > - Storage. Files? Persistent objects? A directory structure?
>
> > LispMs had a hierarchical file system. Pathnames looked like
> > titan:>mac>graphics>painter.lisp.37.
>
> So the bit before the colon selects a network machine, like
> a URL? That's pretty handy.
BTW: ANSI-CL pathname structure is modeled on the LispM concept.
An example from our application:
USER(1): #p"ngc:patch;nworld-3-1;nworld-3-1-001.fasl"
#p"ngc:patch;nworld-3-1;nworld-3-1-001.fasl"
USER(2): (describe *)
#p"ngc:patch;nworld-3-1;nworld-3-1-001.fasl" is a structure of type LOGICAL-PATHNAME. It has these slots:
HOST "ngc"
DEVICE :UNSPECIFIC
DIRECTORY (:ABSOLUTE "patch" "nworld-3-1")
NAME "nworld-3-1-001"
TYPE "fasl"
VERSION :UNSPECIFIC
NAMESTRING NIL
HASH 64954
USER(3): (translate-logical-pathname **)
#p"/TREES/devo/nworld-3-1/patch/nworld-3-1/nworld-3-1-001.fasl"
USER(4): (describe *)
#p"/TREES/devo/nworld-3-1/patch/nworld-3-1/nworld-3-1-001.fasl" is a structure of type PATHNAME. It has these slots:
HOST :UNSPECIFIC
DEVICE :UNSPECIFIC
DIRECTORY (:ABSOLUTE "TREES" "devo" "nworld-3-1" "patch" "nworld-3-1")
NAME "nworld-3-1-001"
TYPE "fasl"
VERSION :UNSPECIFIC
NAMESTRING NIL
HASH NIL
Notice as the HOST slot is :UNSPECIFIC in Unix (as well as the VERSION slot)
> > They also supported logical
> > pathnames like sys:color;alu.lisp. A logical pathname would get
> > converted into a physical pathname by the system using info stored in
> > sys.translation files. Physical pathnames were of the form host:<host
> > specific path>. For instance, trantor:/usr/mac/foo.lisp was also a
> > vaild physical pathname, that happens to refer to a file on a Unix
> > host.
>
> So it could transparently use different protocols to access hosts?
> Yet again, pretty handy!
Standard ANSI-CL features.
> > Printers didn't live in a "printer's directory". They were
> > represented by namespace object, along with hosts, users, and sites.
> > Theyre were commands to print files and do screen dumps (anyone else
> > hack LPGs?). Since this was before the days of PostScript, if you had
> > a PS printer hooked up and wanted to generate specific PS for it,
> > you'd have to do that yourself. (You could print text and do screen
> > dump to it, though.)
>
> That seems pretty good, although obviously the ability to speak to
> printers in a high level graphics language instead of characters would
> be nice - a character stream smells of UNIX :-)
>
> > Mike McDonald
> > mikemac@engr.sgi.com
>
> Thanks,
>
> ABW
> --
> Alaric B. Williams (alaric@abwillms.demon.co.uk)
>
> ---<## OpenDOS FAQ ##>---
>
> Plain HTML: http://www.delorie.com/opendos/faq/
> http://www.deltasoft.com/faq.html
>
> Fancy HTML: http://www.deltasoft.com/faq0000.html
--
Luca Pisati Voice: (310) 577-0518
Nichimen Graphics Fax: (310) 577-0577
12555 W. Jefferson #285 EMail: pisati@nichimen.com
Los Angeles, CA 90066 Web: http://www.nichimen.com