Lions’ Commentary on Unix

ProphetOfDoom

ProphetOfDoom

20407352-F9EE-4326-8AA0-F26BCA264B9E.jpeg

I just saw this on Amazon and immediately added it to my wish list so I’ll probably be purchasing it in the next week or so.
If you’ve already got a copy I'd love to hear your thoughts.
I have no idea whether I’ll actually end up reading it. I didn’t get very far with the famous Operating Systems: Design and Implementation by Tanenbaum as it was just too dense for me. From the reviews it seems this one is a bit easier on the reader?
 
It teaches V6 Unix, originally for UNSW undergraduate course work, but was also adopted inside Bell Labs for that purpose. It was very highly regarded, and commonly held to be the most illegally copied book, ever, in computer science (because you needed to be a V6 licencee to get a legal copy).

Its great virtue is that it's approachable and exposes many of the principles of Unix with few of the complexities added as time went on. So, it's the best way I know of to grasp the fundamentals of the way that (quality) operating systems work.

I managed to borrow and photocopy a set of originals back in late 1970s (commentary and source code). Plus I bought the "legal" copy (above) when it came out in the mid-1990's (mostly out of respect for John Lions, whose health was failing).
 
Thanks gpw928 , bakul
I wonder if it would be possible to build the Unix sources for something like a BeagleBone and start a whole new lineage of Unix? This is idle speculation of course, I imagine one could lose years or decades on a project like that.
There was a comp sci professor from UNSW who used to post his undergraduate lectures on YouTube about 15 years ago and I watched them avidly, I wish more schools would do that. He was really passionate about the subject.
patmaddox /anyone else interested I also recently read Unix: A History and a Memoir (published quite recently) and it was a really interesting read.
7B50E87E-A2E5-45CC-BCF1-00D72ECA47B5.jpeg
 
ProphetOfDoom said:
I wonder if it would be possible to build the Unix sources for something like a BeagleBone and start a whole new lineage of Unix?
Click to expand...
Not inconceivable. A port of Version 6 Unix to the Interdata 7/32 was completed by Richard Miller and Ross Nealon at Wollongong University, Australia, during 1976-1977. This was the first time Unix was ported. I believe that Bell Labs got Unix going on the Interdata 7/32, but after Miller and Nealon. The principal virtue of the Interdata 7/32 was that it was cheap.

Of course, V6 is a long way from where Unix is today. However, getting it going on a PDP-11 simulator would be quite plausible.

[By the way, the printed book contains the commentary and the sources, but the URL posted by bakul above has that and more.]
 
Nealon completed porting ed to Interdata32 but I believe Richard Miller did almost all of the work. Also see Miller's own paper from the era:

Richard Miller has long since moved on to plan9. He was the one who ported plan9 to RaspberryPi (part time work over 4 months). He made it work on pretty much every model long before any BSDs, including support for Wifi (& may be sound as well).

Porting v6 to new hardware is probably not worth it.
 
The Lions book will have to wait a bit as I ordered Absolute FreeBSD and am having a slight cash flow problem...
Incidentally, reading the terminal commands for installing a new unexplored OS gives me exactly the same feeling in my chest as asking someone out for a date. I honestly wonder if I’m on the autistic spectrum or something.
 
You must log in or register to reply here.
Back
Top