! README.1ST !! Documentation and command procedure for SCI_APL
!
! @README.1st will COPY the SCI_APL to a tape
! See the end of this file for MORE details
!---------------------------------------------------------end.of.info
!
$ Show time
$ write sys$output "Note.. A tape that will be Initalized, and"
$ write sys$output "       SCI_APL will be written to should be"
$ write sys$output "       on MT:"
$ write sys$output ""
$ inquire what     "INITALIZE and COPY SCI_APL:*.* to MT :default YES"
$
$ what := 'f$extract(0,1,what)
$ if( what.EQS."N" ) then goto enddoc
$
$	Directory SCI_APL:/Totals/Size=all
$	initalize MT:/density=1600 SCIAPL
$	mount MT: SCIAPL
$	Copy SCI_APL:*.* MT:*.*;1		!! SCI_APL
$	Copy SCI_EXE:SPLIT.*,XTOC.* MT:*.*;2	!! Two useful programs
$
$enddoc:
$	write sys$output "Type AAAREADME.1ST for other information"
$	EXIT !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

	This is the latest release of SCI_APL, a resonably complete
version of Iverson's A Programming Language, to DECUS and other
interested persons. As before, this is in the public domain, and will
be supported as time permits. This software is provided by: 

		Mike Liveright		Systems Control Inc. 

		(415) 494-1165		1801 Page Mill Road. 

					Palo Alto, Ca., 94304 

	The following will INITALIZE and COPY SCI_APL:*.* to tape for
SCI_APL. NOTE.. To RUN SCI_APL, Recompile it, etc. it is necessary
that a system name "SCI_APL" be defined that points to the directory
in which SCI_APL resides. The following is in my SYSTARTUP.COM file. 

!		$ DEFINE/SYSTEM SCI_APL DRA2:[SCIAPL] 

	The update history of SCI_APL is contained in the last source
file of APL.FOR, i.e. the SUBROUTINE ZAPL. In general I have tried to
fix bugs as they were found, and remove some of the restrictions.
Thanks are due to the people who tried to use SCI_APL, dispite the
errors, so that these errors could be found. 

	Steve Turley of MIT supplied the basic code for using standard
DEC editors, i.e. EDIT, EDT, and TECO, thanks. I modified it, so he is
not to blame for the errors, but just deserves the credit if it works.

		The files on this tape are:

	README.1ST	.. This file 

	APL.FOR, APL.OBJ, APL.OLB, APL.EXE	.. Obvious 
			.. note.. APL was probably compiled with
				debuging on. To get a faster running
				version, and one that does not check
				array bounds, re-FORTRAN APL.FOR, and
				then re-LINK it.

	APL.Vxx		.. Older versions of SCI_APL

	*.I		.. Include files for SCI_APL compilations 

	*.SCI		.. Data files used in running SCI_APL 

		APLHEL.SCI .. The )help file. This contains 
information about the APL commands, etc. The format of this file is:
header information, followed by a first level title, * in col 1, 
followed by the information about that topic.

		APLMNE.SCI .. The MNEMONICS, e.g. .eq ... for ascii
keyboards, so that the the user can enter FORTRAN type mnemonics for
the operators rather than "strange" apl characters. 

		APLSET.SCI .. The "inital" setting when APL is brought 
up. APL should probably be defined as a system command so that the 
system programmer can define in each user login:

			APL :== $SCI_APL:APL @

The user can then automatically get the inital settings, and append 
these with his own settings, see: SYSSET.SPL.

		APLTRA.SCI .. The translation between the external APL 
keyboard and the internal ASCII characters. After a: )set apl, the 
input keyboard characters are translated into the internal ASCII, and 
inversly on output.

	SPLIT.*		.. A routine for spliting APL.FOR 

	XTOC.*		.. A routine for listing APL.FOR 

	*.R		.. The FALL 1980 DECUS runoff files.
			There may be errors, but some documentation
			is better than none ???

	FOR003.DAT	.. The output of a Calling/Called program run
			against APL.FOR .. This may be out of date, or
			missing, but if it is on the tape it can be
			PRINTED to give an idea of the sturcture of 
			the SCI_APL program.

	*.SPL		.. These separate source files are generated
by $ RUN SPLIT and the giving the source name APL.FOR. The Update
history is contained in the last subroutine ZAPL, of the APL.FOR file.

	There may be one undefined subroutine, i.e. START_TIME. This
is the DECUS program for generating a PC-HISTOGRAM, DECUS, Spring 1979.
The program is not used unless the user enters )pchist. This would
trace the Program Counter for optimizing. Don't worry, or comment out
the call from the program SYSPCH, or write a no-op program to field the
call.