When Unix Landed - Computerphile

we're talking late 70s over to early 80s a crucial period actually for computers in many regards yes there was definitely a british's best britain first purchasing policy either for government institutions like radar establishments like the police all this kind of stuff and particularly for educational uh establishments the feeling was computers are expensive so if we can spend a lot of money it might as well be fed back into the british economy by buying british certainly applied to universities one or two were allowed to escape i think university of newcastle on the computer science front was allowed to have an ibm so of course was the university of cambridge i can't remember what happened at oxford it was a way for me to find out by the back door about the advantages in a way of sharing a hardware platform with not just other universities but other government institutions and in one of the previous videos i've talked about how we only felt able to use algol 68 as a teaching language because that well-known government institution blessed them the royal radar establishment had actually put a huge amount of effort into setting up a very good algol 68 compiler and of course it had to be on ico machinery they were a government establishment so were we so we benefited from that a lot how did you get into unix you went to the states didn't you can you tell me about it yes i it's true i went over to the states in 1978 for the whole calendar year and visited in short order and it was all highly educational a visit to oklahoma state university where they'd written an alcohol 68 compiler so given my interest in those days that was a no-brainer but i then thought no no if i'm going to be away for a year i want the california experience and so i finagled my way into doing some teaching at ucla of course in los angeles but also in the cal state system cal state northridge is a well-known california state thing and i did i lectured all sorts of computer things there taught fortran in oklahoma and so on it was weird actually but in some ways at nottingham we were a little bit ahead of the curve there because before i left we've taken our first tentative steps into getting this thing called unix which enabled you on a underpowered pdp-11 to sort of run your own programming experience multi-user but only on dumb terminals yeah multi-user multitasking operating system hopelessly under bad machinery but it sort of worked and this was must have been about 76 when we bought um first of all the pdp 1105 just to get used to having our own computer of some sort could do a similar programming but i think just more or less as i got in the plane for america colleagues told me that no we're going to get that upgraded to a pdp 1135 as quickly as possible and that when we do we will be ready to receive this operating system called unix done by bell labs and people said to me today if you'll find just like why did a radar establishment have to support your compiler you'll now be saying why does a telephone company have to be the people that provides me with a little operating system that i feel happy with and uh it was it was as yogi bera once said like deja vu all over again oklahoma state they were saying wow you know you're ahead of us we want to go to eunuchs we can't get approval all this kind of stuff so when i came back uh 1979 very cold winter all that it really was a liberation to know that while we might have to rely on the competing center for a lot of bulk teaching and running programs nevertheless we could start to do our own thing on pdp-11s with an eye to the future but even so we couldn't on a peter 11 35 give a bulk service to our compsci undergrads we continued running algol 68 on icl machinery with a huge extended overlap with us getting eunuchs because we couldn't become self-hosting on units or have enough big powerful machines until well after this era we were made very aware that we were getting this for a few hundred dollars because it came from bell labs bell labs was a so-called regulated monopoly and that meant they were caught in the crossfire from everywhere they must not be seen to be using their massive leverage and obscene telephone bills to squash little startup telcos they must not get in the way of ibm because there was a suspicion that you know when people said to bell i was well what are you doing for the public good they thought it was a nice cool idea to say look we've got basically a computer science research department we're not a computer company but these researchers it really was a bit like marvin's radar defense establishment we are not a computing company absolutely not but we have developed some software that universities might find interesting so we're proposing for the public good to virtually give away this software if anybody's interested and it it's a start and ibm won't see it as a threat uh so we're all right you can get hold of this so long as you are a bona fide degree giving institution you can get hold of this unix you must have a course uh p2p11 got to keep america's balance of payments in the black and it must be at least a pdp 1135 not the most minimal you could get if you do that this unix will get you started in the great world of multi-user on simple terminals you can do your own thing you'll run out of steam after about four or five users but nevertheless it's all yours so we cheerfully signed up for this and this as i say it was back in 1979 it's all about to happen version seven of unix i think we did somehow get hold of the previous one version six but serious usage started with version seven to find out more about this version seven that we were getting how did you find out let me be boring and oldie and say yet again there was no internet there were no web pages to tell you about this there was not even email you had to send out posters to departments you had to use letters you could send faxes right and you had to say we are the uk unix user group principally populated by representatives from computer science departments in uk universities we all almost all of us absolutely long to have eunuchs we have persuaded the american master unix user group and att the ultimate owners of bell labs who ran all the patents and licensing they're going to come over and tell us how to get licensed for an academic use only license for unix and what's it all about and it was a superb chat shop and ideas swap shop as it were enjoyed it hugely the star of the show was a visitor from at patents and licensing frank j riffle jr i remember his name and uh oh boy did he have the job from hell trying to keep everybody happy and he said i want to license you all you're wonderful you know uh we love universities and i'm in charge of patients and licensing but i gotta warn you you must not do any commercial usage with this it is purely for degree giving teaching only right and he kept on nuttering to himself there were a lot of gray areas and it's my job to stamp them out and what a heck of a job it was he said just remember you must be a bona fide degree giving university institution then you will be licensed no others a hand goes up from the audience a guy i got to know quite well later on but he really was the odd man now han goes up and said well how about me you've licensed me for version six but i'm not at the university where are you from sir and he said marcus gray this guy's name was he said i'm just a college i'm pre-university you know from basically we concentrate from 11 to 18. i'm marlborough college i'm a public school now i think even the americans know that there's nothing more badly named than an english public school it is not available to the poor and the needy as it probably should be or was originally no some public schools are so expensive you need serious money to attend them principally eaton is the most famous but marlborough college isn't far behind i mean the tuition fees at marlborough college will make your eyes water so this guy said no no i'm not i'm pre-university i'm marlboro college but sir over here you know most colleges are like universities and marcus said no it's not the case with me and riffle was getting more and more ruffled that's a joke and said sir let me try and summarize are you trying to tell me that you are more or less a very expensive version of what i would call a high school and marcus said sort of said yeah that's roughly right he just looked at mark and said sir you should not have been licensed and he didn't say it but in parenthesis i could hear it going on in the background oh boy somebody's going to be in deep trouble when i get back to murray hill new jersey he said sir you should not have been licensed but frank j riffle is a man of his word i will not take your license away sir truly it did change everything and i do remember very well we had a lot of conversations at the uk units group saying those guys at bell labs are just so good the two of them obviously dennis and ken we've talked about them many a time they've invented the c language for system implementation they've developed unix and it does things sensibly and this is absolutely great aren't we lucky i thought as a community to have them being on our side and i did think i wonder if they'll ever get any recognition from the acm which shall we say about what they've done and i was very pleased it took till 1983 i think but eventually so many academics in the u.s and all over the world were using unix that dennis and ken were given the acm touring award it was actually awarded to them in 84 it was won and achieved in 83. so in 1984 they had to give a talk about what their work meant and normally when you write a paper to go with your presentation you thank everybody which he did but then you go on to say here's how units came about here's the nuts and bolts of what it's like to really run it we didn't get that at all we got thank yous from ken in cryptic terms but very heartfelt to a lot of people so he said i now just want to point out something to you effectively you've got something here where we use c as a system programming language so the operating system is written in a high-level language have you any idea the problems that is going to open up to the potential flaws security holes let me just demonstrate one of these to you you'll be amazed and amused and i mean you know what standards we have on computer file we don't swear so i'll allow myself to say that one of the rather celebrated people that we're going to point you at in the next video just described ken's contribution as a total effing bombshell and he wasn't wrong a lot of people couldn't understand why it was a bombshell but it really really is and i'd like to do my best to explain to you just how what a fundamental earthquake this caused once people understood what he was trying to say a few seconds or half a second of time then gave him a half and then give her a little bit and just go round and round the people who wanted to do some kind of computing and because computers were even the signal is moving in so the speed of those pulses tell us the rate we know which direction it is because we know which roller it's coming from and we can look at those

Channel: Computerphile

Views: 110,105

Keywords: computers, computerphile, computer, science, Unix, University of Nottingham, Professor Brailsford, Coding, C Programming Language, Ken Thompson, Dennis Ritchie, Bell Labs, AT&T

Id: fCDsn7OTNMg

Length: 13min 12sec (792 seconds)

Published: Mon Mar 08 2021