British-made programming language

[SN]

Anyone remember CHiP?

[rurwin]

Ferranti Autocode was British, and my first introduction to programming languages courtesy of a book in the school library way back before anyone invented the micro.

Then you'll like this: MicroAutocode
That should be the same language; Manchester University were designing the Ferranti machines.

All my own work; nobody else was daft enough. (esr thought it crazy enough to list it on his museum page.) But it's a living language; one of the Leicestershire Amateur Radio beacons is programmed in it.

[antiloquax]

There's also Miranda. A functional programming language that influenced Haskell. I haven't actually used it ...

http://miranda.org.uk/

[aaronsloman]

"Are there any programming languages invented and maintained in the UK or by a company/organisation that is based in the UK? If so, what are they?"

I have only just noticed this question. One of the respondents mentioned Poplog. Poplog is an interactive learning and development environment developed at Sussex University, then marketed by SDL, then ISL until ISL was bought by SPSS. It contains incremental compilers for not one but four languages: Pop-11, Prolog, Common Lisp, and Standard ML (Poplog version is PML).

Of these Pop-11, the core language, used to implement all the others, was designed and implemented in the UK, building on ideas in POP-2, developed at the University of Edinburgh in the early 1970s for AI research, and used in the pioneering Freddy Robot, first demonstrated in 1973. (There are wikipedia and other online information sites for each of the languages, for Poplog and for Freddy.)

After SPSS bought ISL (in order to get the Clementine data-mining systems, implemented using Poplog -- mainly Pop-11, and now part of IBM's business software), Sussex agreed that Poplog should be hosted as free and open source at the University of Birmingham, where it has been available since 1999. See http://www.cs.bham.ac.uk/research/proje ... oplog.html

It has been used there and elsewhere for research and development, including for example AI/Cognitive science research using the SimAgent Toolkit http://www.cs.bham.ac.uk/research/proje ... agent.html which makes considerable use of Pop-11's language extension facilities, based on its incremental compiler available at run time (compiling to machine code).

(For information on ISL and Clementine see http://www.cs.bham.ac.uk/research/proje ... /isl-docs/)

The current linux version is available here http://www.cs.bham.ac.uk/research/proje ... installing

A port to Raspberry Pi would require porting the linux vesion to the ARM cpu, a non-trivial task because most of the Poplog source code, including the code for the incremental compiler, is in Pop-11, requiring a boot-strapping process.

Some teaching examples using Pop-11 are here: http://www.cs.bham.ac.uk/research/proje ... /examples/ and first draft set of teaching videos (to be improved/extended) here: http://www.cs.bham.ac.uk/research/proje ... rials.html

There are examples of Pop-11 code which can be compared with formulations in other languages, on the Rosettacode web site http://rosettacode.org/wiki/Category:Pop11

If anyone is interested in helping with or discussing uses of a port of Poplog to ARM+linux (which could add a powerful interactive AI language with a very small footprint, for teaching or applications, to the R-Pi), or merely wishes to know more, please email me at a.sloman[AT]cs.bham.ac.uk (school of computer science, University of Birmingham)

Thanks
Aaron Sloman
http://www.cs.bham.ac.uk/~axs/

[mikio]

CESIL anyone?
I have an awk script that will interpret CESIL code if anyone wants it?

I know it's an old post but i would still like to know: what license is your awk version of cesil (gpl ?) - i love it and would love to spread it around if i could? Great fun!

[mikio]

Ok - I'll have it! I've worked my way through my old school textbook writing out the cesil programs on paper. I'll run them all through the awk-cesil .

[flapjackboy]

lol I thought nothing could get any weirder than Prolog.

Richard S.

Befunge
Brainfuck
Whitespace
Chef
INTERCAL
Shakespeare
FALSE
LOLCODE

It gets a hell of a lot weirder than PROLOG.

[mikio]

"Are there any programming languages invented and maintained in the UK or by a company/organisation that is based in the UK? If so, what are they?"

I have only just noticed this question. One of the respondents mentioned Poplog. Poplog is an interactive learning and development environment developed at Sussex University, then marketed by SDL, then ISL until ISL was bought by SPSS. It contains incremental compilers for not one but four languages: Pop-11, Prolog, Common Lisp, and Standard ML (Poplog version is PML).

----------------------------------------------------snip---------------------------------------------------------------------------------

The current linux version is available here http://www.cs.bham.ac.uk/research/proje ... installing

A port to Raspberry Pi would require porting the linux vesion to the ARM cpu, a non-trivial task because most of the Poplog source code, including the code for the incremental compiler, is in Pop-11, requiring a boot-strapping process.

Some teaching examples using Pop-11 are here: http://www.cs.bham.ac.uk/research/proje ... /examples/ and first draft set of teaching videos (to be improved/extended) here: http://www.cs.bham.ac.uk/research/proje ... rials.html

There are examples of Pop-11 code which can be compared with formulations in other languages, on the Rosettacode web site http://rosettacode.org/wiki/Category:Pop11

If anyone is interested in helping with or discussing uses of a port of Poplog to ARM+linux (which could add a powerful interactive AI language with a very small footprint, for teaching or applications, to the R-Pi), or merely wishes to know more, please email me at a.sloman[AT]cs.bham.ac.uk (school of computer science, University of Birmingham)

Thanks
Aaron Sloman
http://www.cs.bham.ac.uk/~axs/

I read the examples page of using poplog to teach computer science: there is a bit of magic sprinkled throughout those words! I hope the poplog system will get ported to the arm platform one day, it would be a real asset on a raspi! Me, i am am still wrestling with cesil (and ansi basic) .

[SN]

CESIL anyone?
I have an awk script that will interpret CESIL code if anyone wants it?

I know it's an old post but i would still like to know: what license is your awk version of cesil (gpl ?) - i love it and would love to spread it around if i could? Great fun!

Only four years later...

I have updated the source cesil.awk file I pasted into this thread four and a half years ago to include a GPLv3 licence for reuse

You can download this version at

http://stevenicklin.com/downloads/cesil.awk

[Dub]

occam

[jamesh]

CESIL anyone?
I have an awk script that will interpret CESIL code if anyone wants it?

I know it's an old post but i would still like to know: what license is your awk version of cesil (gpl ?) - i love it and would love to spread it around if i could? Great fun!

Only four years later...

I have updated the source cesil.awk file I pasted into this thread four and a half years ago to include a GPLv3 licence for reuse

You can download this version at

http://stevenicklin.com/downloads/cesil.awk

I suspect you might win the prize for the oldest resurrected thread, but it's taken away again because your post is actually relevant and hence disqualified.

[gordon@drogon.net]

occam

Nah- I mentioned that over 4 years ago. Do keep up at the back

-Gordon
(Who has written OCCAM programs, also Transputer assembler)

[timrowledge]

Let us not forget BLACKPOOL - British Local Area Context-free Knowledge Processing Object Oriented Language, developed at the 1987(?) ESUG meeting in Leicester, UK.

[gtoal]

Well, I was reading through this thread and thinking I'ld like to add some info, but had decided not to when I saw the 2012 date - but since some brave soul has already resurrected it... here goes...

You've got the main ones. Pleased to see that Gordon mentioned Imp - though to clarify the primary language there is really Atlas Autocode (invented by Tony Brooker as a better alternative to Algol) and its later derivations, Imp9, Imp80 & Imp77 - collectively just "Imp".

I see Aaron also mentioned the POP2 family (and {Prolog/Poplog), and of course ML. Edinburgh University also turned out a bunch of minor languages that no-one will likely have heard of - for example HAL - a High-Level Assembly Language from Hamish Dewar with ports to four or five architectures that I can thijnk of. And LC - a Low-level Compiler by Rainer Thonnes that is similar to HAL but approaching the language design from the other side of the Macro language <-> compiled language spectrum.

Talking of macro languages - there are a bunch of UK-designed turing-complete macro languages (GPM, ML-1 etc), though I'm not sure if they would pass your criteria of being true programming languages

Oh - I almost forgot the Edinburgh dialect of LOGO - "AI2LOGO".

Anyway... back to Imp... Peter Robertson designed the portable "Imp77" variant, and when I worked at Acorn we hired his company, 3L, to build some compilers for us. They did code generators for the 32016 and the ARM, with back-ends for Pascal and also a C that they wrote for us specifically. And because all of their tools were written in Imp77, that meant that there was an Imp77 compiler for the ARM - the early version before Thumb, floating point etc.

I would imagine that a combination of Andy Davis's more recent X86 port of Imp77 plus the old ARM compiler sources would mean that we probably could get Imp77 running on the Pi, if anyone is interested in taking that on?

Look for the files pass?-arm.imp in http://history.dcs.ed.ac.uk/archive/languages/imp77/ - if anyone is interested I'll track down the complete build system which is somewhere in our archives. Unfortunately it was an ARM-based system so not directly applicable to unix, though it might be easier to resurrect on RiscOS first.

[Heater]

Ha, I was just scanning this thread wondering why Algol had not been mentioned. The mother of all modern programming languages.

As Tony Hoare said: "Here is a language so far ahead of its time that it was not only an improvement on its predecessors but also on nearly all its successors."

https://en.wikipedia.org/wiki/ALGOL#History

Not exactly British perhaps but certainly a lot of input from the UK.

Don't forget Charles Babbage and/or Ada Lovelace wrote the world's first computer programs:
http://factmyth.com/factoids/ada-lovela ... r-program/

[PeterO]

As Tony Hoare said: "Here is a language so far ahead of its time that it was not only an improvement on its predecessors but also on nearly all its successors."

I have an autographed copy of the manual for the Algol-60 compiler for the Elliott 803 signed by Tony and Jill Hoare and Jeff Hillmore

PeterO

[Heater]

PeterO,

Wow!

It's only after picking up Algol in my spare time using Kent university's ICL 4130 computer that I realized that these computer things were more than just toys. As I gathered from being taught BASIC in technical school a few years before.

The Tony Hoare comment I quoted above is significant for me because for many years in my early career it seemed every project required learning a new language. It all seemed like a pointless waste of time as they were all basically Algol but different, and not as good: Coral, Pascal, ADA, C, etc.

Of course the Algol legacy lives on in Java, C#, Python, Javascript etc, etc, etc....

Or does it?

A big point of the Algol idea was to have a formal specification of the language. I'm not sure that languages like Python or Rust etc do.

[PeterO]

A big point of the Algol idea was to have a formal specification of the language. I'm not sure that languages like Python or Rust etc do.

BNF (https://en.wikipedia.org/wiki/Backus%E2%80%93Naur_form) is another of those seemingly forgotten tools/techniques (like flow charts) that I still use from time to time to help with designing code.

I first used it to design a "hex monitor" programme for the first microprocessor I built while still at school (~1977)

And I still get to play with Algol-60 https://www.youtube.com/watch?v=AIxZ1i8pvZI

PeterO

[jahboater]

A big point of the Algol idea was to have a formal specification of the language. I'm not sure that languages like Python or Rust etc do.

Please don't forget Algol 68 which had an even more rigorous specification of the language (which was produced before the first compiler was implemented).

[PeterO]

A big point of the Algol idea was to have a formal specification of the language. I'm not sure that languages like Python or Rust etc do.

Please don't forget Algol 68 which had an even more rigorous specification of the language (which was produced before the first compiler was implemented).

Only fair to include a quote from Tony Hoare :
"The best we could do was to send with it a minority report, stating our considered view that, "... as a tool for the reliable creation of sophisticated programs, the language was a failure." [...] - C. A. R. Hoare

PeterO

[jahboater]

I wonder if it would have succeeded now, with todays compiler technology?

https://en.wikipedia.org/wiki/Compariso ... nd_C%2B%2B

[Paul Webster]

I was taught Algol (68R I think) and lambda notation by, amoung others, Peter Landin ... who, some years earlier, had taught the same thing to Tony Hoare.
First big programming lesson that I remember (1977 by Prof Khabaza I think) was using recursive procedures to solve the missionaries and cannibals problem.

When I started work it was programming in Coral-66 which was similar enough to quickly pick it up.

[Heater]

PeterO,

Only fair to include a quote from Tony Hoare :
"The best we could do was to send with it a minority report, stating our considered view that, "... as a tool for the reliable creation of sophisticated programs, the language was a failure." [...] - C. A. R. Hoare

To be fair to Algol that quote is referring to Algol 68. The supposedly new and improved version of Algol. Seems Tony and others considered it to be an overly complex language design and full of errors. He should know he was on the design committee. Shockingly it turns out that the committee suppressed Tony's negative report on the Algol 68 design.

On the other hand Tony has good things to say about the old Algol 60.

This creeping complexity, unusability and crappiness of a programming language design as it evolves over time seems to be a common phenomena. Look at what happened as C evolved into C++ and the increasing nightmare that has become over the years. Similarly I'm worried about what has been happening to Javascript in recent years.

I'd say Algol was pioneering in that respect

Meanwhile the means of formally specifying the language syntax, Backus–Naur form, introduced with Algol, lives on in all kind of schema for things like XML and JSON and parser generators like YACC.

Tony's quote above comes from his paper "The Emperor's Old Clothes": https://gist.github.com/jfacorro/50d0b7 ... 9541302ba0 an interesting read.

[Heater]

Paul Webster,

Wow, it's a long time since I heard anything from a former Coral-66 programmer.

I did a couple of years of Coral 66 back in the early 1980's. For Marconi Radar.

Yes, Coral was similar enough to Algol to be picked up quickly. I had never heard of it before starting at Marconi Radar. They took me on as a contractor anyway. Didn't get any training in Coral either. Just had to wing it based on my experience of Algol back in uni.

[PeterO]

PeterO,

Only fair to include a quote from Tony Hoare :
"The best we could do was to send with it a minority report, stating our considered view that, "... as a tool for the reliable creation of sophisticated programs, the language was a failure." [...] - C. A. R. Hoare

To be fair to Algol that quote is referring to Algol 68.

Yes, that's why I put it after the previous post introduced Algol-68 to the thread.

The supposedly new and improved version of Algol. Seems Tony and others considered it to be an overly complex language design and full of errors. He should know he was on the design committee. Shockingly it turns out that the committee suppressed Tony's negative report on the Algol 68 design.

Yes, that's what the quote was about.....

On the other hand Tony has good things to say about the old Algol 60.

Which isn't surprising considering it allowed him to properly describe "quicksort" for the first time

This creeping complexity, unusability and crappiness of a programming language design as it evolves over time seems to be a common phenomena. Look at what happened as C evolved into C++ and the increasing nightmare that has become over the years. Similarly I'm worried about what has been happening to Javascript in recent years.

I'd say Algol was pioneering in that respect

Meanwhile the means of formally specifying the language syntax, Backus–Naur form, introduced with Algol, lives on in all kind of schema for things like XML and JSON and parser generators like YACC.

Tony's quote above comes from his paper "The Emperor's Old Clothes": https://gist.github.com/jfacorro/50d0b7 ... 9541302ba0 an interesting read.

Yes, it provides an interesting insight into several aspects of computing in the 1960s. Of particular interest to me is his description of the failure to develop an operating system for the Elliott 503. Since I have experience of writing code for 503's earlier sibling (the 803.... don't ask) I can understand why it proved so difficult. The 503 did have a few extra features that would have made the task easier, but it still lacked some fundamental features that would be considered essential today (like a hardware stack and position independent code).

PeterO