Current issue

Vol.26 No.4

Vol.26 No.4


© 1984-2017
British APL Association
All rights reserved.

Archive articles posted online on request: ask the archivist.


Volume 22, No.4


APL Starmap

From: John R. Thorstensen

It was a surprise and delight to find (in an idle moment of web browsing) that APL Starmap is still alive and kicking on the Web!  I wrote that in the summer of 1973 between my 3rd and 4th years of College, when I happened to find lodging for a summer job in a spare bedroom at Paul and Sarita Berry’s house in Swarthmore, Pennsylvania. Since then I’ve gone on to other things ...

but I still enjoy writing code for the celestial sphere ...

John R. Thorstensen – Professor and Chair
Department of Physics and Astronomy
Dartmouth College
6127 Wilder Laboratory
Hanover, NH 03755

Memories of Ken Iverson and APL

From: Paul Barnetson ( 30 September 2006

It was with great interest, and with many happy memories, that I read the Vol. 22, No 3 edition of Vector, with its theme of “Remembering Kenneth Iverson”.

When I was young, and before I reached the “age of discretion”, I used to irritate my friends and family no end by continually asking them what they meant by the words they used, and pinning them down to get precise definitions. It was to nobody’s particular surprise that mathematics was my favourite subject at school and university, before I joined IBM to work with computers. And back in 1963, that was a burgeoning and very new field!

In 1968 I was sent on assignment to the USA for a couple of years, and while I was there I was introduced to APL, “the language that forces the computer to think in the way that you do, rather than forcing you to think in the way the computer does, like other languages”, as it was described to me.

I fell in love with it then, and on my return to the UK, set out to evangelise about it within IBM-UK. During the 70s I was IBM-UK’s “APL-man”, and some of your older readers might remember me in this job.

During this time Ken wore a beard, many of his US colleagues also did, the marketeer for APL in IBM-UK had one too, and as the technical man I also wore one. With vivid ginger hair at the time, I was the man with “the red one”, and a hirsute appendage was expected of all male APL enthusiasts!

One of my contributions during this time was being secretary to the APL Standards Group, ably led at that time by Dave Ziemann; I think he eventually moved over to a language called Smalltalk. We used to spend hours arguing over whether a particular feature involved standardising the interpreter or the language – not an easy thing to decide – not to mention discussing whether we were standardising what had been done, or whether we were trying to lay down rules for the future.

In the early days, when APL was still only available on 2741 terminals, I would travel to various exhibitions and conferences to organise demonstrations of this marvellous new language. I was always accompanied by my carefully guarded box of APL golf -ball typewriter heads. This box was immediately labelled “Paul’s Balls”, much to the amusement of my mainly female demonstrators – I really don’t know why!

At another APL conference I was chairing a presentation (by Bob Bernecky I think), and a marketing company had given everybody a name badge with a number on. Prizes were being offered for delegates who could find somebody whose badge had the same number as their own. So I defined SET as the collection of all the badge numbers in the auditorium, and I invited everybody to re-organise themselves according to


This way, they would find themselves sitting next to the person with the same number as themselves! If N was the number on my badge (I forget what it was now), I then wrote up


to find there was a match in the auditorium! We later collected our prize together.

I was lucky enough to meet Ken during this period, along with many of the contributors to this particular Vector. Adin Falkoff, Jim Brown, Gitte Christensen, Per Gjerlv, John Scholes, Morten Kromberg, Gerard Langlet, are all people who helped me along the APL way, not that there is any reason why they should remember me. I still correspond with Norman Thomson – he has tried to convert me to J! But you can’t teach an old dog new tricks.

Now I have retired, I still use APL. One application is writing packages for charities; an example is The Charlton Chase, an overnight orienteering event for young people on the South Downs in the middle of winter. My package enables us to calculate when we could expect the teams at their next checkpoint, and so warn us if they are overdue. It also enables the young people to have the final results immediately the last team returns. I wrote an article on this package for Vector, a long time ago now.

As an enthusiast of recreational mathematics all my life, I have also used APL in innumerable problems concerning magic squares, prime numbers, and the like. I find these in New Scientist (Adrian – I also solved Enigma 1368, and in exactly the same way that you did!), and thanks to APL my name has appeared on three occasions as the lucky winner. I have also had similar problems published in Computer Weekly, which carries a “Puzzler”each week.

My only problem is finding a cheap single-user modern APL system; the ones on the market are expensive multi-user systems. Would you believe I’m still using a 1980s version of PC. APL, which somehow came home with me when I left IBM!

My regards to all my old APL friends and colleagues, and – like them – I can truthfully say that APL changed my life.

We’ll miss you, Ken – you changed the world, and not many people do that.

Thank you for the chance to write this letter.

Paul Barnetson
22, Ferndale Rd
W. Sussex  PO19 6QJ


From: Sam Sexton ( 11 October 2006

Although I’ve moved away almost completely from the world of APL at work, I still pick out the interesting (to me!) articles in Vector now and then and really enjoyed reading the tributes to Ken Iverson. If you have a follow-up of further memories, you may wish to add this:

I was at Ken’s presentation in Helsinki in 1985 that Michel mentioned and I recall his “very critical marks”in response to questions about APL2 being put extremely politely along the lines of “well, these are the holes in it, it’s illogical, but if that’s what you want, it may be useful”. However, the stinger was at the end, with his final words on the subject being “I’m just glad they didn’t call it APL!”. Warm applause and cheering ensued. It was an honour to meet him then and on my visits to IPSA in Toronto and I treasure my signed copy of A Programming Language – as it’s red, perhaps it should be referred to as The Thoughts of Chairman Ken...?

Regards, Sam

Sam Sexton
Infrastructure Group
Transactions Group (Sales & Trading)

From: Woody Butler ( 1 October 2006

[dyalogusers] SAGE: A new Web Server for APLers

Greetings, APLers.

Over the past 3 years, Harold Ward from Netwings and I have developed a very robust Web Server engine called SAGE.

SAGE is written in APL2000 (runnnig on Windows) and is designed to interface with any language on the Windows server that supports standard COM methods.

Once an APL workspace is “registered” as a COM object on the server, SAGE can easily communicate with it... sending and receiving GET and POST strings from Internet users browsers and applications.

SAGE uses a back-end SQL SERVER DB for authentication and security control. So, by enrolling users in to tables (Members, Groups, and Web URL Resources) you can easily protect specific URLs. SAGE URL Resources can also be given full Public Access, as well.. so user login is NOT required.

I use SAGE to call DyalogAPL through COM. My DAPL WS gets called upon a URL GET or POST to SAGE, and SAGE passes my ⎕LX autostart program a nested vector with parameter names and values. I run my desired back-end APL program and then return a simple HTML string back to SAGE through COM, and SAGE passes the HTML page back to the user.

We’ve been running SAGE in production for 2 years, and it is rock solid. We would be interested in working with other APLers who would like to write Web Applications using DAPL, APL2000 or any other “COM” supporting language that runs on MS Windows.

Please reply if you are interested, and we can discuss. Since we own a small ISP (, we have 100% control over the servers, network, firewall, DNS, database and development environment.

Woody Butler (
President, TeamBase Corp.

script began 5:49:00
caching off
debug mode off
cache time 3600 sec
indmtime not found in cache
cached index is fresh
recompiling index.xml
index compiled in 0.2965 secs
read index
read issues/index.xml
identified 26 volumes, 101 issues
array (
  'id' => '10011240',
regenerated static HTML
article source is 'HTML'
source file encoding is 'ASCII'
read as 'Windows-1252'
URL: =>
URL: =>
completed in 0.3238 secs