🚀 go-pugleaf

In article <8849@boring.cwi.nl> dik@cwi.nl (Dik T. Winter) writes:
>[...].  Algol 68 is very nice, but not readily available [...].

	How readily do you need it?  It's available, sort-of, for all the
computers I have had access to since 1972.  For example, it's on the big
ICL mainframes in our computing centre;  it was on their VAXen (VMS) until
they upgraded them and declined to pay for the compiler upgrade;  it was
on our PDP 11/{34,70,44} [a Unix V6 binary (only, sadly), that needed
hand-tweaking to run on V7];  it's on this machine (Piaggio), a Sun 3/260
running SunOS 4.0.3.  The last Algol Bulletin listed lots of compilers
for lots of machines, including IBM and CDC, for example.

	The version I'm currently using is Charles Lindsey's 68S compiler
for the Amsterdam Compiler Kit.  So, if you can run ACK, you just have to
speak nicely [enough] to CHL.  ACK, of course, is very widely ported.  (I
*just* failed to get ACK/68S up on our 11/44 before they came to take it
away [ho, ho, hee, hee, ha, ha, hum, hum] last October;  curses!)  The
Malvern 68RS compiler is also highly portable;  then there's 68C, Flex, etc.

	If CHL is reading this group, perhaps he could supply us with a
more up-to-date listing of availability than was in the last AB?

	Algol 68 arrays are *so* *nice* to *use* compared with those in all
the other popular languages (arbitrary slices, flex, lwb/upb operators,
assignation, mode [type] of element, etc.), without significant loss of
efficiency, that I've never understood why more modern languages didn't
at least start from the same sort of model.

-- 
Andy Walker, Maths Dept., Nott'm Univ., UK.
anw@maths.nott.ac.uk