Fragments of Medusa archeology

Installation, supported revisions
Post Reply
User avatar
Posts: 55
Joined: Fri Nov 16, 2012 6:04 pm
Location: Cambridge, UK

Fragments of Medusa archeology

Post by MikeCauser » Thu Nov 28, 2013 8:34 pm

Poking around in the Medusa startup macros I found some stuff that I
recognised in med2d/m2d/macro/default_view.mac:
  • -- In the following table the first letter is OpSys (S-unOs, V-ax, A-lpha
    -- U-ltrix, HP-UX) and the second is Window Manager (S-unView, O-penWindows,
    -- V-WS, D-ECwindows, M-otif). Each combination has several numbers:
    -- - the depth of the header for a menu window
    -- - the width of other menu borders
    -- - the depth of the header for the "view" window
    -- - the width of the other borders for a "view" window
    -- - flags for whether the co-ordinates are to the inside or outside
    -- - flag for the co-ordinates of the top left corner of the display
    -- None of this can be guaranteed because the border width can be changed
    -- by the user in some W-Ms and the header depth will always vary with
    -- size of text font selected. These numbers assume 12/14 pixel text size.
    let wmlist = "SS SO VV VD VM UD UM HM AM SM LM"
    let mhdr$w = 5 5 6 23 4 24 5 5 5 5 5 -- menu header
    let mbrd$w = 5 5 6 3 4 3 4 5 5 5 5 -- menu border
    let mins$w = 0 0 1 1 1 1 1 0 1 0 0 -- menu co-ord inside (1) or outside (0)
    let vhdr$w = 18 27 21 23 31 24 31 34 34 34 34 -- view header
    let vbrd$w = 5 5 6 3 11 3 10 11 11 11 11 -- view border
    let vins$w = 0 0 0 0 1 0 1 0 1 0 0 -- view co-ord inside (1) or outside (0)
These are the window variations for the supported Medusa workstations
around 1987-90. To get each set of values I had to count pixels on the
screen -- easy if the system had a screen-dump function and a
pixel-editor, but rather hard on the eyes if it didn't... Of all
these I suppose that SunOs+OpenWindows is the only one on which Medusa
might still run. But this old code is still run every time Medusa4
starts. Good job we have >2GHz processors now.

A bit more poking around in the start sequence found some curiosities in
the Bacis1 variables that are left after startup:

RESL$W (U) 1280 1024
Screen resolution: but I have a 1920x1280 display.

Window Manager: Urghh, I disliked Motif 25 years ago, using KDE now.

OS$TYP (U) 'unix'
Well, yes, OS "type" is unix.

M$ARCH (U) 'i386'
Machine Architecture: if it's a 386 how come it cost me so much?
(And, are we ever going to get a 64bit version of Medusa?)

BTNSM$ (S) '@@[ *** Command Aborted ***@@]@@A'
BTNUR$ (S) '@@VCOM$'
BTNUM$ (S) ' '
These are the commands the mouse buttons should execute, S for
with-Shift, U un-shifted, then Right, Middle and Left of course.
But the buttons don't.

REV$ (S) 18 5 1 2
Revision of Medusa. So internally it's really

NOTE none of this matters when Medusa4 is running, it just amuses me to
find it's still there, lying unused for so many years.

I've found quite a few macros still in use that I wrote, or at least
originated but have been modified later. My bracketing and whitespace
style is fairly distinctive. And on paper I have part of a symbol
library generator I wrote in "Bacis1", although we called it
if-then-LCIS at the time, dated 28th November 1984. Happy Birthday....


Post Reply