Newsgroups: comp.lang.apl
Path: watmath!ljdickey
From: ljdickey@watmath.waterloo.edu (L. J. Dickey)
Subject: Re: info exchange between APL systems
Message-ID: <1991Dec24.215127.13639@watmath.waterloo.edu>
Organization: University of Waterloo
References: <911224053312_70530.1226_CHC81-2@CompuServe.COM>
Date: Tue, 24 Dec 1991 21:51:27 GMT
Lines: 275

In article <911224053312_70530.1226_CHC81-2@CompuServe.COM>
	70530.1226@CompuServe.COM (Mike Kent) writes:
>For a posssible survey article for Quote Quad, I am soliciting
>descriptions of techniques for exchanging information between APL
>systems, particularly workspace exchange.  Pleas post responses in
>this newsgroup, or send them to me *as mail*.   Thanks.


I agree that something should appear in Quote Quad, just so that as
many APLers as possible find out about what has been done so far, and
about what is about to appear.  The work that Mitloehner and Sirlin are
doing will form a good basis for network communications and I hope that
we will be able to settle on a de facto convention for APLII.  Already
PP is available, and I think that more is coming soon.

Two few months ago, a survey was conducted on this net, and with
the help of many from the net, found about thirty transliteration
schemes.

This list is becoming more complete as time goes on, but I do not have
access to all the data, so much of what I have here is based on
information that has been given to me by others who are "in the know".
I suspect that some of the entries in the table are not really
distinct, because of name changes or company takeovers, or OEM sales.
If you can help me add new entries, or fill in some blanks, or supply a
needed example, or by identifying common forms, please write.


						Sincerely yours,
							Lee Dickey

 ==============================================================================

		     A list of APL transliteration schemes
				   1991-10-10
				    Draft 3




Where known, an example is given showing the form for the statement
"A is assigned the transpose of the 2 by 3 reshape of the first 6 indices".

(1)	STSC keyword scheme
		a #is #transpose 2 3 #reshape #count 6
		a  is  transpose 2 3  reshape  count 6
		a #is #transpose 2 3 #rho #iota 6
		a  is  transpose 2 3  rho  iota 6
	In some instances, either the name of the character or the name
	of the function may be used.  The hash mark, #, is optional.

(2)	The J scheme
		a=.|:2 3$i.6
	Not really a transliteration scheme, because symbols here do
	not stand for other symbols.

(3)	The Budd scheme: (Tim Budd, budd@cs.orst.edu)
		a <- .tr 2 3 .ro .io 6
		a _ .tr 2 3 .ro .io 6
	Used by APLC, the APL compiler.

(3a)	The Budd scheme: revised.
		a <- .trans 2 3 .rho .io 6
	A descendant of (3).

(4)	The Hohti & Kanerva scheme for TeX
		a _ @TR 2 3 @RO @IO 6

(5)	WSIS standard
		(unreadable)

(6)	APL\11 (at least one old version)
		example needed
	Ken Thompson at Bell,
	John Levine et al at Yale.

(6a)	APL\11
		a{\\<bs>O2 3RI6
	The freebie product "APL\11" distributed with BSD Unix.
	Bruner & Reeves, Purdue University
	A sequel to (6) ?

(7)	The Iverson scheme: (Kenneth E. Iverson)
		a@<- @O\ 2 3 @r @i
		A@<- @O\ 2 3 @r @i 6
	Each sequence begins with @ and ends with blank
	APL Quote Quad, 17.4, APL87.
	APL Quote Quad, 18.1, Table 1, A dictionary of APL.


(8)	The Crick logographic scheme   (Michael F. C. Crick)
		A=transpose 2 3 shape iota 6	(?)
	APL Quote Quad 11/1, Sep 1980
	Perhaps this was the motivation for (1).

(9)	The Sirlin scheme (Sam Sirlin, sam@csi.jpl.nasa.gov)
		a.is.tr2 3.ro.io6
	Used for workspace interchange.

(11)	DEC APLSF  (DEC 10, DEC 20)
		A_.TR2 3.RO.IO6

(12) -	IPSA's APL, dictionary scheme.
		A@<- @O\ 2 3 @r @i 6
	Intended to be the same as (7).

(13) -	Honeyapple, Honeywell Bull, GCOS
		a"is "tps 2 3"r "i 6
	Sequence begins with <"> and ends with a blank.

(14) -	VAX APL, by DEC.  TTY mnemonics.
		A_.TR2 3.RO.IO6
		a_.tr2 3.ro.io6
	Every escape sequence begins with a dot and has two chars
	following.  Upper and lower case escapes have same meaning.
	Is (14) an extension of (11) ?

(14a) - VAX-11 APL, by DEC.
		A_.TR2 3.RO.IO6
	A predecessor of (14).

(15) -	The transmission code scheme:
		a[O^H?2 3RI6
	Based on the APL-ASCII typewriter pairing scheme,
	registered character set 68.  <<^H>> stands for backspace.

(16) -	The Selfridge scheme (R.G.Selfridge, selfridg@nervm.bitnet)
		a [ O?2	  3 R I 6
	Every APL symbol uses up two bytes.  If the second char
	of the pair is blank, it is ignored.  If not, the pair is
	resolved to a single composite symbol.

(17) -	The Way scheme (F.T.Way III, fxw3@po.CWRU.Edu)
		a[!O?2 3RI6
	Escape sequence used for composite characters.
	Each such sequence begins with "!", and has two more characters.
	The diaeresis itself is represented by "!! ".

(18) -	APL for CDC NOS/VE
		A$IS$TR2 3$RO$IO6
		a $IS $TP 2 3 $RO $IO 6
	Not sure about case of "a" here.
	Clearly the blanks do no harm, but are they specified?

(18a) - APL2 (APL level 2) for CDC
		A$IS$TP2 3$RO$IO6
	Is this the same as (18) ?

(19) -	Xerox APL, running on Xerox 560 and Sigma 6/7/9 Computers
		A$IS $TPS 2 3$R $I 6
		A _$TPS 2 3 $R $I 6
	It used the dollar sign as escape character.
	Variable length sequences.

(20) -	The Vienna APL2TEX scheme.  Used by Geyer-Schulz and company.
		A{\APLleftarrow}{\APLcircleslope}2 3{\APLrho}{\APLiota}6
	Used for documentation.	 APL source is converted to TeX source.

(21) -	The Mitloehner scheme, (John Mitloehner, <mitloehn@uxf.wu-wien.ac.at>)
		Under development
	This one wants to be "readable".

(22) -	The I-APL scheme.
		A_s2 3ri6
	I-APL has no lower case letters in names.

(23) -	APL/700 (circa 1973) for Burroughs 5000-7000
		example needed
		A<IS><....>2 3<RHO><...>6		(?)
	Just guessing here.

(23a) - APLB (circa 1985) for Unisys A-series.	A descendent of (23).
		example needed
		A<IS><....>2 3<RHO><...>6		(?)

(24) -	IPSA support for Teletype ASR33. (1971)
		example needed
		a_				(?)
	(Robert Bernecky & Roger Moore)

(24a) - IPSA support for ASCII terminals
		example needed
	A descendant of (24) ?

(25) -	Hitachi (IPSA's APL/AF)
		example needed		.tr
	Start with a period, followed by two letters.
	Same as (24)?

(26) -	ISO DIS 10646, a four-octet character coding scheme.
		Unreadable.
	10646 wants to list *every* known character.

(27) -	Unicode, a two-octet character coding scheme.
		Unreadable (for the time being)
	(Maybe 26 and 27 will be merged into one big char set.
	Still under negotiation, October, 1991.)

(28) -	The Harris scheme
		A _ @TR 2 3 @RH @IT 6
	This looks like ???

(29) -	The Hewlett-Packard scheme, APL\3000
		example needed		"IO
	Always three characters, the first was <">.
	Based on some previous scheme ???, but with extensions.

(30) -	Honeywell CP-6 APL
		example needed


(31) -	Xerox CPV APL
		example needed
	Same as (30) ? 

(32)	APL-11, from DEC.
		A_.TR2 3.RO.IO6
	Not the same APL as (6).
	This version of APL is still availble from DECUS.

(33)	U of Maryland APL  (APL/UM ?) on a Univac 1110,
	Format for Punched card input
		example needed
	Done by Pat Haggerty
	Same codes as (11) ?

(34)	IBM 1130, Punched card input
		A#@TRAN 2 3@RHO @IOTA 6

(35)	York APL
		a#[#?&#O2 3#R#I6
	If you knew where the characters were on a 2741 keyboard
	then you York APL by typing # for shift and & for backspace.
	So quote-quad was #L&#K.

Acknowledgements:

I am indebted to many who have helped me start this list.

	Bob Beckwith		<Bob_Beckwith@NeXT.COM>
	Bob Jernigan		<JERN@APLMVS.jhuapl.edu>
	Bruce Frost		<bjf@sv001.sandiego.ncr.com>
	Bruce Nordman		<b_nbca@dante.lbl.gov>
	Christopher Greaves	<christopher.greaves@canrem.uucp>
	Curtis A. Jones		<jonesca@sjevm5.vnet.ibm.com>
	Fraser Jackson		<jackson@matai.vuw.ac.nz>
	Frederick T. Way III	<fxw3@po.CWRU.Edu>
	Greg P. Jaxon		<jaxon@csrd.uiuc.edu>
	Henry M. Cowan		<hmc@ipsaint.ipsa.reuter.COM>
	Johann Mitloehner	<mitloehn@wu-wien.ac.at>
	John R. Levine		<johnl@iecc.cambridge.ma.us>
	Karl Puder		<puder@mltvax.enet.dec.com>
	Leigh Clayton		<loc@yrloc.ipsa.reuter.COM>
	Mark Eklof		<mark@11srus.enet.dec.com>
	Martin Gfeller		<mgf@ipsaint.ipsa.reuter.COM>
	Martin Turner		<mtur@ipsaint.ipsa.reuter.COM>
	Michael Berry		<mjab@Think.COM>
	Michael Rhs		<rys@inf.ethz.ch>
	Morten Kromberg		<insimkr@uts.uni-c.dk>
	Murray Eisenberg	<murray@math.umass.edu>
	Peter Jaspers-Fayer	<sofpjf@vm.uoguelph.ca>
	R. G. Selfridge		<selfridg@nervm.bitnet>
	Richard Tilley		<tilley@ccu.UManitoba.CA>
	Robert Bernecky		<rbe@yrloc.ipsa.reuter.COM>
	Robert J. Frey		<rjfrey@kepler.com>
	Roland H. Pesch		<pesch@cygnus.com>
	Sam Sirlin		<sam@csi.jpl.nasa.gov>
	Seth Breidbart		<sethb@fid.morgan.com>


-- 
Leroy J. (Lee) Dickey, Faculty of Mathematics, U of Waterloo, Canada  N2L 3G1
Internet:      ljdickey@watmath.UWaterloo.ca
UUCP:          ljdickey@watmath.UUCP             ...!uunet!watmath!ljdickey
Obsolescent?:  ljdickey@watmath.waterloo.edu     BITNET/EARN: ljdickey@watdcs
