Page 29 is the first code. You need to work through 'War and Peace' of text before you get there. This is not good for a "young person". They need to get there quicker IMHO.
I always say that the reason I got into programming (at the age of 10) was because of the BBC Micro User Guide [1].
* Page 5: first code written (and importantly, code that has a visual component)
* Page 6: Drawing lines
* Page 7: Playing sounds
etc.
I realised very quickly that I was able to make a game. And that was it, I was hooked for life.
The Raspberry Pi took its inspiration from the BBC Micro (Model B). It's a shame that there wasn't more of a push to make it as easy as the original Model B to get going and start programming.
(Of course I realise that it's running Linux where nothing is simple, but yeah, I'm not sure it necessarily helps kickstart the next generation in the way the original BBC Micro did)
I think the title is just a play on "A young person's guide to the orchestra" (by Britten). Nobody in their right mind would want to learn programming in BCPL, the predecessor of the predecessor of C.
> Nobody in their right mind would want to learn programming in BCPL
I agree, but that's not what the author thinks:
"When a new programming language is designed it is invariably strongly influenced by languages that preceded it. One thread of related languages is: Algol -> CPL -> BCPL -> B -> C -> C++ -> Java, indicating that BCPL is just a small link in the chain from the development of Algol in the late 1950s to Java in the 1980s. BCPL is particularly easy to learn and is thus a good choice as a first programming language."
Emphasis mine.
Also,
"This document is intended to help people with no computing experience to learn to write, compile and run BCPL programs on the Raspberry Pi in as little as one or two days, even if they are as young as 10 years old."
So, unless I'm missing something, this document is intended to teach programming first concepts with BCPL. And it seems like it's aimed at young people, regardless of the play on words (which I didn't know, thanks!).
> > Nobody in their right mind would want to learn programming in BCPL
> I agree, but that's not what the author thinks:
The author of this document is Martin Richards, the creator of BCPL. Of course he thinks you would want to learn it.
We've all been a young person[0], so it's easy to criticise from that angle.
However, I'd like to think the readership of HN are hacker enough to like to think that, given a desert island (with minor deps like sand and the Friday hardware fab!) and a few spare years, they'd be capable of Robinson Crusoe'ing up a personal development environment.
From that point of view, I find[1] this effort admirable: it's several hundred pages, by Martin Richards, about the port of the BCPL toolchain (language, compiler, bytecode interpreter, libraries, debugger, etc. all by Martin Richards?) to a new system (was it a new arch as well?)... by Martin Richards.
How many among us get a machine with new graphics, decide to write a flight simulator, and then —only as a minor[2] implementation detail mind you— plumb float support through our entire language ecosystem? Who needs a shaved yak when you're pursuing the buttery-smooth shaded yak?
[0] or maybe we even still are: https://news.ycombinator.com/item?id=42682353
"when I was your age all we had were 1's and 0's, and sometimes not even 0's, so we'd stay up all night xor'ing 1's to have enough for the next day..."
[1] when I did this sort of thing to pay the bills, I had someone else doing the tech writing and someone else providing toolchain and debugger sources, and it was still not a trivial lift.
[2] compare Wirth FPGA'ing up an entire homebrewed arch just so he could have a personal workstation using his favourite mouse.
You're right. I'm somewhat stumped.
Chapter 4.1 (installation of BCPL, before you can start coding) is indeed somewhat(!) more complicated than how it was for the BBC microcomputer (or any other back in the day) - just power it on, enter
10 PRINT "HOW ARE YOU"
RUNNice introduction text with detailed description to set-up a R.Pi and interesting examples (AES, calculating Euler's number, Rubik's cube etc.). Shame it uses an utterly obsolete language of which most interesting feature (byte code interpreter making porting the compiler to new architectures comparatively easy) was important in the seventies with its plethora of architectures (many short-lives), but hardly today.
I'm going to test out the assertion that children as young as 10 can use the guide, by asking my 11-year-old to use it. I don't believe it; it assumes way too much knowledge of computer terminology. We'll see.
BCPL also looks dry as hell, I'm not convinced it's a great first language.
It one is going to play with obsolete languages, then wouldn't Algol 68 be a better pick? At least it has types beyond "cell".
Or is the idea that BCPL is quite close to the bare metal?
Algol 68 Genie: https://jmvdveer.home.xs4all.nl/en.algol-68-genie.html
See this, code starts on pg 11: https://jmvdveer.home.xs4all.nl/learning-algol-68-genie.pdf (after 18 pages of preface)
See this for how BCPL begat B which begat C: https://www.bell-labs.com/usr/dmr/www/chist.html
> At least it has types beyond "cell".
I prefer MCPL ( https://www.cl.cam.ac.uk/~mr10/mcplman.pdf ) but I believe BCPL did recently pick up a floating point type, after Richards got a Pi and before he wrote a flight sim.
[does the C in CPL stand for Cambridge, Combined, or Christopher? Yes.]
Wow, this is an amazing book, nearly 800 pages; counless examples of BCPL applications, by the author of the BCPL language.
My total knowledge of BCPL: it’s what AmigaDOS was written in before they ported it to C.
https://en.wikipedia.org/wiki/TRIPOS was written in BCPL (in 1978) at the University of Cambridge.
TRIPOS was ported to the 68000 in 1981 by Dr Tim King. In 1984, he joined the company MetaComCo, and they bought the commercial rights to his 68000 port.
The Amiga OS was well behind schedule in 1984. MetaComCo were contracted in March 1985 to integrate TRIPOS into the Amiga's existing OS code (mainly Exec and Intuition), and the Amiga launched in July 1985. That's a pretty fast turnaround time!
TRIPOS is most evident in the AmigaDOS portions of AmigaOS, specifically dos.library, the CLI, the directory structure (C, L, S are TRIPOS standard directories for commands, libraries, scripts, hence the odd mix of LIBS for Amiga libraries but L for AmigaDOS handlers and filesystems), the standard disk filesystem, and all filesystem handlers.
For AmigaOS 2.0 (released 1990-1991), it wasn't quite "[TRIPOS] ported to C"... AmigaDOS was rewritten entirely. There was no BCPL or TRIPOS left, except deliberate structures/functions (written in C!) to support backward-compatibility with BCPL software.
AmigaDOS's API was greatly extended to provide official OS calls for what were previously BCPL globvec internals, and most of the command-line programs were rewritten in C. This work had already been advocated for and done in 1987-1989 by the third-party "AmigaDOS Replacement Project" https://aminet.net/package/misc/antiq/ARP_13
BCPL--hadn't had my first coffee before I read this, I thought it was going to be SBCL.
BCPL ("Basic Combined Programming Language") [0]
Insanely frustrating that BCPL doesn't seem to be defined anywhere in the document itself.
It's defined in the BCPL book.
University of Cambridge Revision date: Tue Oct 23 16:58:19 BST 2018
What makes it specifically for young people? I looked at the paper but didn't spot anything obvious
In the context of BCPL, "young person" simply means "under 90".