Jon "maddog" Hall talks Unix and Linux history
Вставка
- Опубліковано 29 бер 2018
- Jon "maddog" Hall gives a brief history of the period of 1969 to 2019 with regard to 50 years of Unix and Internet advancement. Some of the high (and low) points of that period and its meaning to computer science of today. He calls for a celebration in the year 2019 of the women and men who made these advances possible. 2019 will be 50 years of Unix, 25 years of usable Linux, Linus Torvalds's 50 years old, 10 years since the start of the ideas to setup Linaro and more.
- Наука та технологія
“In corporations many things are not achieved by going through formal chanels, but by asking for reciprocal favors or friendship.” How thruthful!
Wow! I'm impressed with Jon's skills for story telling. I felt like I was reading a novel.
This interview is GOLD for me.
Many people don't have a minimum idea about software history.
Hope somebody could make history book about software
and the people that make possible for the future.
It's a pleasure to listen to this man.
he, Stallman and linus should be assigned a Nobel prize for the influence on the humans race development...
mope worthy of peace prize than Obama.
@@Canadian789119 that isn't hard to achieve.
Richard Stallman, Definitely Yes.
Send the suggestion of the peace prize to Norwegian parliament that get nominate and then decide who get the peace prize. No, that one isn't decided in Stockholm.
Agreed.
Omg... His memory of details and events is incredible... I don't even remember what I ate yesterday!
that's because you brain stores only important information that is interesting, the brain doesn't store memory like a computer.
@@ChristopherGray00 Yep absolutely right.
Yes, great person with great memory.....
Love this guy. He was the author of Linux for Dummies for a few editions.
Man , Maddog i remember this guy from some old linux online forums, wow this guy is really amazing, sharp mind.
I was under the impression that Torvalds began work on the Linux kernel not in the hopes that it would be useful for GNU, but rather as a personal project to teach himself about kernel design. I did not realize that he was aware of the GNU project from the very beginning.
This is a really great and informative interview.
Yes that too. He in fact wanted a *nix-like OS for his own use. There's many reasons floating around.
It was hard _not_ to be aware of GNU back then, because their tools were often better than what was shipped with commercial UNIX variants, and it was Free Software. GNU was the obvious choice for the application layer.
Excellent rundown of Unix history
I appreciate his TIME for the interview.
Total respect for you, sir. Thanks for sharing this knowledge.
Wow, this guy is crazy interesting! I could listen to him all day. Such a great interview!!!
Great job, Charbax! Thoroughly enjoyed watching and learning!
thanks for sharing this amazing interview.
Because of AT&T's legal monopoly on telephone service at the time, AT&T was also not allowed to sell computer products as such, which is one reason why UNIX was released free for educational purposes.
Actually Unix was not really free for educational purposes. It was *very* low price (a few hundred dollars) for a site wide source code (the only way it was distributed at the time).
Unfortunately if you were not a *research* university (Stanford, Berkeley, Carnegie Mellon, MIT, etc) the price was tens of thousands of dollars per CPU.
I know that, because I tried to buy it for my small two-year technical college.
Very clear explanations of these key events. Thanks for sharing this.
Very illuminating history! Thank you Mr. Jon. I am a DEC VAX11/780 , VMS fan. Like Mr. Linus and own a pi 400. It was like he was telling my thoughts and telling me things that I didn't know! 🙏🙏🙏
I really want to spend an afternoon with this guy listening him talk about unix history
Great interview with a great man, cheers!
Thank you for sharing this.
Wow, it has taken my five years to find out about this youtube video. I first met "Maddog" when I was Ultrix SIG Chair for DECUS Europe in 1989. That is also when I met Kurt Reisler my USA counterpart. The European DECUS was in places like Den Hague and Cannes. By 1994 my term as SIG chair had ended but I went to the DECUS conf in New Orleans. Great memories !
On the UNIX side: In the 80s I worked for Imperial College Computer Center and we got one of those university copies of Unix source. At the time I had just written a C compiler for the CDC6500 and could also cross compile for the 8086. So I got Unix running on an iSBC86-12 Multibus system around about the time IBM PCs started.
Feels so humbling to sit and listen to this guy
I wish engineers would learn computer history as default. So many of those questions would be already well knwon, and this interview could have so much more interesting technical questions answered! This guy has so much to teach and that is omitted in this video...
Probably the thing that most people noticed first about Berkeley Unix vs. Bell sysV was the BSD had job control (CTRL-Z to suspend a running job, fg, bg, etc.) Previously, in UNIX if you launched a program without &, you were stuck and had to wait for it to complete, sit on your thumbs until it was finished. Not many people paid the $160k for UNIX (is this even true?) but many universities took a sourcecode campuswide site-license for $20,000(~7 cars) for UNIX.
I wonder when they invented the program screen. I could probably look that up. But why would I? You facts are more interesting :)
PS. had to look it up. en.wikipedia.org/wiki/GNU_Screen#History
nice interview! a legend
This video is a gem
Loved this video, I got into Linux in 2018 with my Linux Mint, I came from windows haven't looked back since. Thanks for sharing.
The Holy Trinity... Maddog, Stallman, Torvalds.
Naah man. The Holy Trinity is Thompson, Ritchie and Kernighan.
@@FranciscoMNeto I think there is a whole bunch of holy people who got funded in one or another way and now we got where we are.
What about Ken Thompson, Dennis Ritchie and Brian Kernighan - "The three Wise men"? Don't forget about them!!
2 yr. old thread......oh well.....still
Yeah, and Stallman is only a clown! He have made him self a clown. His only trying to get glory to himself, from anything.
Stallman is a gifted programmer and a complete f*cktard
There are so many interesting facets in the history and evolution of, well of everything we see and do today in a digital setting.
That was a pleasure, thanks
I really enjoyed this, thank you!
31:46 cracks me up every time😂😆
22:55 nice explanation of the importance of open source
Why isn’t this taught in schools?!?
I have a photo with him. An honor to me
Great history. Terry Davis applied many of these principles in the development of TempleOS, and I'm convinced it is a revolutionary new tech.
"Send (the alpha computer) to Finland right away ..." ... a Finn here :)
Amazing stories!
One of the reasons early computer science departments liked Unix was that they could buy their own PDP11 or VAX computers to use without accounting software.
Before the 1980's the computer at Rice was a room-sized IBM-clone owned by the University and used by the administration to run the school. The math and science departments had to "buy" computer time from the University for their students.
If a student made a mistake (such as an infinite loop), an entire semester's budget could be spent in one day.
When I started as a freshman at Rice, I was a "mathematical sciences" major, but by 1981 they had a real "computer science" degree. It's hard to understand today just how liberating Unix was in 1980 by allowing students and researchers to play with the computer without having to worrying about an accounting bill.
Awesome video Charbax! long time no see! was hoping to see Maddog in this year's Fosdem but didn't make it?
Fascinating. Worked with the Alpha running Unidata under DG-UX, had the "Live free or die - UNIX" numberplate. There was a version of Windows (NT?) that ran on the Alpha, I wonder how many people actually used it? IIRC there were several ex Digital people involved in the creation on Windows NT.
I think the machine named perch at our university was an Alpha. It was definitely risc. But I was younger and didn't know asmuch back then.
This conversation covered Unix to Linux history in depth. But I was surprised that the C programming language was mentioned little or not at all. My understanding was that Unix's development demanded the creation of C almost from the beginning and that Ritchie and Thompson were deeply involved in both.
I think C on all over this superb "novel" was like tomatoes on pizza, so he forgot to mention it
Red Hat For Dummies was my first book on Linux, before you could just get everything off the internet.
Ran down to WHSmith to grab that, and didn't even have a PC to use the CD that came with it, just read cover to cover for a year. Dude still a legend to me.
I am glad you liked LFD.
This is a great video, thanks for posting. Around 11:00 when maddog starts talking about rms is some sweet Gnu history.
I remember when I was a kid in the early 80's My dad worked for AT&T and had an AT&T 6300 PC that ran UNIX and MS-DOS. MS-DOS stuck long-term because it had games...lol But I remember him teaching me UNIX and DOS. I think somewhere in his basement we may still have the orange and gray books on both UNIX and MS-DOS.
They were good days. I was born in the eighties, but in my country I did not have a computer. I entered these interviews to see the people who owned at that time and their impression of the regime.
maddog the legend!
Amazing, He is truly the Legend!
I really enjoyed this
What a computer science lesson!!!
This is crazy interesting.
SO insightful !!!
Very interesting.
Great video on the History of Unix / Linux... Im using Openindiana OS which is Open Solaris based on System V Unix
In 1989 I asked a dean what should I do in college he said computers are going to soon burst onto the landscape you should take as much of it as you can. I liked this history it tells alot about the people in it I first got Collderia Linux in a box tried it out it was mostly broken didn't work well at all then PCGamer gave away free Linux cd's I grabbed it that was the first real working Linux I got I run Solus Linux and Linux Mint Solus is new and keeps a more updated Kernel and run a bit faster great video
Great man, interesting video - definitely work the 'like'.....
impressive!
*bows in sheer admiration*
31:47 when you meet Linus Torvalds about getting Linux running on the Alpha system, but he tells you he's going to use the power pc instead!!
I'm lop the hhhhhhhh
I transform into a crow. Obviously.
Wow! What an incredible man
Legend !
Ncomputing was doing FPGA triple headed graphics cards back around 2009 too, for thin client applications, thought that was cool.
Linus has told me that he would not start a kernel project today. Instead he would learn how to program an FPGA.
31:46 the true maddog manifesting for a brief moment
THAT - the compiler story . . in some time.. the compiler was a barrier..
Always good to hear from these guys talking about how it was when one person could make a difference. And except for Torvalds, none of them really made tons of money for their contributions.
I do not know about Linus' finances (nor do I want to know), but I can almost guarantee that Bill Gates has more money.
Another giant is interviewed ... Great
@Charbax The Title should be: Jon "maddog" Hall talks Unix and GNU/Linux history.
I like it when he said “ I dropped my drink” .
R.I.P. Dennis Ritchie
32:41 why did the video get cut?
it turns out that the battery runs out
31:47 aaaah
Fork off .. And die was and is reality ;) unix is and was.. The centre of my universe.. Cool video charbax , can you hear he doesn't talk down to you.. or sales or any other kind of.. Product.. It's all about potential.
I love in the whole open source community, that you can share your knowledge for free. And even technical books are quite cheap. This is one rease why I love Linux. Linus did not "hold back" his knowledge.
I love the way John ignores the interviewer and just ploughs on telling his story. Like the future stuff as well...
Still can't believe I've actually met him personally
Hey, Jon, are you still around? I'm still around, still running Linux! :)
Hahah! No better way to manipulate a socially awkward intellectual than with the strategic application of alcohol. :P
at 23:36 the interviewer asks Maddog what was his the first impression upon meeting Torvalds for the first time, and Maddog just answers absolutely phased out stuff
He doesn’t have it quite right. Bob Marsh negotiated the first contract with AT&T when he was with Onyx. At that time there was a young gentleman running manufacturing at Onyx by the name of Scott McNealy.
There were many people trying to negotiate licenses with AT&T about that time. I think you might admit that Sun was a bit more successful.
@@penguinnh I am just trying to set the record straight. Bob Marsh and Kip Myers’ influence in those days is underrated.
Stallman didn't write the kernel because he couldn't. It took him decades to finally complete a GNU kernel (HURD). It's still not really a viable product. Part of the reason for this is that Stallman stubbornly failed to follow the wise advice of Linus Torvalds regarding how to build a kernel.
The kernel was (should be) one of the simplest programs in the system. At the time there were different paradigms floating around on the approach to take in designing a kernel. From what I understand there were multiple attempts (by different groups) to do this and the result was a series of separate programs working introspectively to make the system work. OTOH Linux is monolithic, one program that performs all operations under a single hood. The whole GNU system was a joint effort by many groups of people, you can't single handedely slate RMS for it.
@@Newtube_Channel rms is a gifted programmer, but a complete and utter f*cktard.
I wonder what Jon "maddog" Hall now thinks about Richard Stallman. And AI, and Machine Learning, and Quantum computing.
I have been hearing that "AI is almost here" for the past fifty years, but I have a relatively large expectation of "real AI".
Quantum computers? I am glad I am retiring.
I couldn't finish this video because of the movement.
31:46
Get to the choppa'!
I tried Linux for 6 months via Zorin OS 16. I just recently went back to Windows. When I would be typing online the system would suddenly stop, and I had to restart my browser and start all over again. When I tried to get help with this on the Zorin Forum I ran into a demigod moderator that soured me from ever going there again. I suppose Linux is great for IT guys but for us novices it is just too much maintenance when something goes wrong and good luck getting help with it. This is just for egg headed tech guys.
Should have chosen something like pop os or mint or ubuntu
The man profshnal
Probashly.
"and I never drop a drink"
Maddog is awesome, but the guy who is doing the interview, it's awful, he barely can talk, and it's struggling to formulate even basics ideas on each question.
i didn't know daniel dennett knew computer history
The dollar cost comparison does not work. Inflation, incomes vary too much. How many Big Macs? This might be better?
How the fu** did this get even a single dislike? Much less 2? Insanity...
Dude was like Ginger Baker, back when he was younger ;-)
I would abolish patent and copyright laws. The Chinese pretty much ignore them which is why everything you own was made there.
> The more functionality you put in the kernel, the more likely there's going to be some kind of mistake.
What's the Linux kernel now? Like two milion lines?
Most of it is drivers.
More like 20 million lines
Indeed once you have established the foundations of a system, the system takes a life of its own by becoming abstract. You can then build in modules as and when you see fit. These are the device drivers.
The kernel is complex for many reasons, but you can get an idea of "shoving functionality out of the kernel" by looking at the number of system calls vs the number of APIs in the libraries. What if all the library functionality was moved into the kernel.
13:17 I just checked Wikipedia, and it says the Hurd kernel has been under development since 1990, which actually predates Linux. So the idea that Stallman was holding off on the kernel in order to concentrate on userland seems wrong. The main obstacle seems to be his insistence on making it a microkernel architecture. That sacrifices performance. It is supposed to give you increased reliability in return, but the fact that he has found it so hard to get it to production quality seems to belie that claim. In the meantime, “monolithic” Linux has gone from nothing at all to production quality across about two dozen different major processor architectures.
The discussion of micro kernels vs modern monolithic kernels (with loadable device drivers and loadable kernel modules) is a long discussion over many beers.
Angry Grandpa! No wait...
He didn’t say much about free BSD.
Peter Paul Chato When Apple makes an OS that requires T2 chip will you drop your silly hackintoshing and join the Linux community?
;)
@@jjwachter82 Prophet John, is that you?
Can't we just keep these guys stay alive on earth By using some sort of potion. Ahh! It's so imotional
Somebody should tell him to clean his glasses :')
The operating system called "eunuchs"
"You need performance and you don't do that with Java. I'm sorry" lol
Students need to continue to focus on fundamentals as does everyone else.