CHM Live | Ethernet@50

foreign Wow Come On In everybody all right welcome everyone to this fun nights program I'm Daniel Lewin I'm the CEO of the museum I've been here five years thank you yay I am very excited to welcome you all here tonight on behalf of our trustees our members our volunteers our supporters the staff there's a whole handful of our trustees here Gordon Bell the founder of the museum back in Boston Len schustick who founded this Museum here 20 plus years ago and a whole series of others as well these moments in time 50 years retrospective we'll talk about that in a minute bring together amazing people and you're all part of that crowd so thanks for coming this is a really terrific program as most of you may know if you've been to any of our programs or peeked at the website lately we have evolved our mission we will always collect and preserve but we are focused on decoding technology the Computing past the digital present and the future impact on Humanity because as we all know life doesn't exist without Computing these days at least anyone that we will reach and so it's really about the implications of computing and without the likes of ethernet we wouldn't be reaching everyone so tonight's program is particularly relevant I think given the purpose and attempt intent of of the museum and our new mission to reach people around the world so very excited about that the um the program will actually mirror our mission in that respect we will be talking a lot about past present and future things and you'll see that over the course of the the evening but before we get into the program I want to thank Ariston networks uh Andy beckleshine and Jay Sheree are here and um also the ethernet Alliance Peter Jones is here from the ethernet Alliance and their support makes a program like this happen so thanks very much uh also the IEEE group the members there are many members here today and there's a brick wall out in front of the the building and um they have honored us once again and they're going to be um embedding uh in the front Courtyard a plaque associated with with ethernet so thank you for that and your ongoing support we appreciate that as well and then later in the program towards the end our partners at The Verge there's a group of people here from The Verge we'll be presenting a program at 8 15 and you're all welcome obviously to stay for that it's particularly exciting as we start to reach again audiences in and around the world so with that I'm going to pass the program off the Marguerite gong Hancock she's vice president in charge of our programming and broader Innovation agenda and it'll be her pleasure to introduce the program as she does so well so thanks Marguerite great thanks so much Dental well I'm delighted to add my warm welcome to each of you tonight we're celebrating ethernet at 50. it's transformed how we connect to each other and people around the world and it's revolutionized as a foundational technology in the way that we are all connected in today's connected World more than five billion users which enables so much of Modern Life we also celebrate the co-inventor of ethernet Bob Metcalf here at chm we're very proud that he was a founding member and early Trustee of our predecessor the computer Museum in Boston and he became a chm fellow in 2008 and next month he'll be honored with the ACM touring award so thank you Bob and we'll have a chance to hear from him later well we have three parts of our program for you tonight and our first panel is going to explore the birth of ethernet in a memo dated May 22nd 1973 note exactly 50 days 50 years ago today while at Xerox Park Bob Metcalf described what he called a broadcast communication Network for connecting within a building some of the first personal computers Parks Altos we celebrated 50 years of Alto last month here at the Museum he developed the network with the late Dave Boggs and the first ethernet ran at not three but 2.94 megabits per second Bob is very precise about that 2.94 megabits per second which was about 10 times 10 000 times faster than what it would replace but how was ethernet developed and what was an environment that enabled this breakthrough that's what our first panel will discuss then a second panel of distinguished innovators entrepreneurs and Company Builders will come up on stage to discuss ethernet standardization and its growth into the stunning Global industry that we see today and in the third part of the evening finally Bob will reflect on this momentous occasion of thinking about what it means today and looking forward and time for audience questions and then ask me anything format so let's meet our first panel continue our chm traditional introduce them using five numbers first yogan Dalal two industry standards co-authored TCP and ethernet 10 Journal Publications three startups co-founded five IPOs while a venture capitalist and 50 years that is now Bob Metcalf as Xerox he played a central role in the creation of the 1980 through net specification with deck and Intel he worked within surf at Stanford on key Network protocols he co-founded Claris metaphor glucos recently of course there's a VC at Mayfield for over 25 years join me in giving a very warm welcome to the open thank you next Bob Metcalf and here are his five numbers 50 years since invite invented co-invented the ethernet 2.7 billion dollars the acquisition price of three com by HP in 2010 V equals N squared met cost law the value of a network includes the square of the number of the connected users 32 feet the length of his yacht the enthusiasm and six the number of different careers so what are those six after co-inventing ethernet azurex Park he co-founded three com in 1979 later he was CEO and publisher of idg's infoworld magazine and then a VC at Polaris Partners he's been a professor of innovation entrepreneurship at UT Austin and now he's in a sixth career as a computational engineer at MIT the one and only Bob Metcalf [Applause] there's one other person who was important to recognize as part of this story The Late Dave Boggs co-inventor of ethernet during the panel we'll show a video of from our chm collection of Dave in his own words reflecting on ethernet so we'd like to honor Dave as well as acknowledges Widow Marsha Bush who's here tonight Marcia thank you so who would it be who would be the right moderator for tonight's program we're so lucky to have Rich Karl guard he's editor at Large Global futurist and the former publisher at Forbes he co-founded The Churchill Club a Silicon Valley leadership Forum which ran 35 years and its collection is now part of our collection he also co-founded upscale magazine and garage technology Ventures so with his extensive experience at the intersection of technology and business in society which is a perfect person to lead us in tonight's conversation please join me in welcoming rich carlgarn thank you well thank you for your attendance at This Magnificent event Bob I want to start with you I just happened to be reminded reading something completely unrelated to this that the by Legend the founders of of Rome were Romulus and Remus now the founding principles of Silicon Valley as we know it today it's really the arpanet on one side and the microprocessor and silicon on the other describe you're from the arpanet world the packet World describe how ethernet came about given your Origins and what you were attempting to solve at Xerox Park in the 70s so the arpanet was a really conceived of as a resource sharing Network for arpa so they didn't have to buy a PDP 10 for everybody you could log into promote PDP 10s and uh and I I was got involved in connecting machines to that network but then I ended up at Xerox Park which had decided to build the first PC and I got the job to design the network for them and what it turned out to be really for was not the dumb terminals that were implied by the arpanet but PCS and the thing that ethernet did was to bring packets to the desktop so that PCS could process them and what did what what did you see I mean how describe how you and Dave Boggs who were different personalities is often you know dynamic Duos who end up inventing things and the valley is full of you know the great pairs who went out and accomplished described how you and Dave Came Upon This and and the development process can I just stop with Dynamic Duos I was I bought a mile of coax and I was trying to figure out what happens to a pulse when you send it down because they didn't teach me that in computer science School and I kept cutting myself with the exacto knife and there was this grad student over there in the corner and he saw me bleeding to death in the corner so he came over and apparently he was very good at stripping cables so he stripped a few of them for me and then he and I became Inseparable for two years we were known at Park as the Bugsy twins well let's talk about how um name a breakthrough or two along the way and then we'll talk and bring yoga in about how it became successful through standards I mean every project like that is a series of experiments and and things that don't work and things that do work and doubling down on what worked what were some give us a sense of a breakthrough moment or two in the development of ethernet so one of them you just touched on to do with the standards the original strategy of ethernet was not to make a standard we had didn't even form that word but in putting deck and Intel and Xerox together to make a network so they could tie their products to one another a bunch of lawyers showed up and said you can't meet because this could be a meeting in Restraint of trade so I called they didn't like the acronym dicks there's been a lot of discussion over the years about that but it turned out that Howard Charney my fraternity brother had sued IBM and antitrust so I called him up and I said how are how can we get deck Intel and Xerox to meet and he gave me a list of four or five things that had to be true like no price fixing no marketing people in the room and the goal had to be an open standard so that's the moment that very moment was the moment we had oh okay standard is going to be a strategy of ethernet proliferation um yogan talk about your involvement in creating the standards in your background and tcpip and now this new thing ethernet without the standards it wouldn't have become the foundation that it did well I'd met uh Bob in the summer of 73 just after this memo had been written but it was you know a secret Bob and David Boggs never spoke a word about it until the summer of 1975 when the blue and white report from Xerox Park was published about ethernet and my first impression when I heard about it was wow you know this is something that's amazing you know technologically very creative and yet very simple even though as Bob and others have said maybe a little complicated to engineer in 1970s technology and I think that particular feeling I had was shared by virtually everyone who came across ethernet and just to give you a very simple story people would ask me how does this thing work what is all this csma CD and I tell them I said look it's very simple you're at a cocktail party having a discussion with a group of people and that's multi-access you know people are talking but in polite Society you don't speak of somebody else is speaking so that's carrier sense and occasionally the two of you or three of you speak at the same time collision and again in polite Society people back off and then somebody starts again so this very simple metaphor captured everyone's imagination and attention and everyone wanted to be part of the ethernet bandwagon as it unfolded and I'll talk a little bit but I was an Internet Protocol person and joined Xerox really to create the next generation of what you would call Inner networking protocols my background at Stanford on TCP where I met Bob and so when I joined Xerox here we are ready to go building these protocols and ethernet and then as Bob said the big magic word deck Intel and Xerox came into being and it really was a very profound moment because I think you have to go back into the 70s as Bob alluded and think about Innovation Innovation was in the air the seeds of today's world were planted in the 70s with of course the arpanet and deck data General Doug engelbart at SRI had come up with the mouse and was developing software for the knowledge worker and so and 71 the microprocessor and and then of course you know Intel and Fairchild were revolutionizing the world of vacuum tubes in creating the micro processor in devram so things were happening and so in that world you know I got asked to basically say hey why don't you join the team from Xerox that's going to work with deck and Intel I'm an electrical engineer by training but I'm really a software guy and so it's people like Ron Crane who's passed away and others who did the electrical engineering but the thing that has survived ethernet is its packet format so well that that led to I think there was a lot of um sort of a clash of theories around that we're going to hear Dave Boggs right now but if if you haven't read the New York Times obituary on Dave Boggs it's magnificent and it ends with this great quote of Dave vogs telling somebody well ether about ethernet well it doesn't work in theory but it works in practice so if we could if we could roll that then I want to have Bob comment on sort of the you know the the the telecoms people versus the package switching people and the different theories if we could draw the clip to early 70s was a time when uh there were office systems based on time large time Mainframe time sharing machines and uh I guess it was a fairly radical thing at the time to issue a mini computer to each office worker these things typically cost thirty thousand dollars and then typical office might have 30 of them that's a million dollars worth of Hardware there what's a what's a secretary going to do with a thirty thousand dollar computer I wouldn't sort of understood at the time we said well we the Xerox Park guys said send email write memos run spreadsheets and um and if you want to exchange that with anyone else you have to call the mini computers together somehow so so the decision to go to issue a computer to every office worker sort of required some way to hook them all together and that was our our take on that was ethernet Bob described sort of this just sort of almost Venus and Mars thing going on with ATT and and why they were upset that it didn't work in theory what was their theory that didn't allow it to work well there were two there were two monopolies standing in the way ATT and IBM and they both had to be broken to build the internet at T had a uh was a monopoly and that Monopoly thinking infused the uh the company oh and one of their uh ideas about what should work is five nines of reliability you know 99.999 I remember breaking the news to them that in the internet world it was five it was nine fives [Music] but the trick was to make a reliable system out of unreliable components so you didn't have to make the low level components in 99.999 if you had re-transmissions for example up here so it was probabilistic versus deterministic they were Newtonian you and you were modern physics I like that yeah so by the way one of these cocktail parties are going to get started so at T had a they eventually came around and especially after they got broken up in 1984. which you know used to be you couldn't hook anything to the telephone Network that was not owned by a t like answering machines for example were a big breakthrough and uh but that got fixed and suddenly we started uh getting higher and higher bandwidth at one point the uh the strategy was to use a t copper to do local distribution and the the at T people tried to discourage Us by telling telling us that those circuit that copper was so cruddy It could only carry 14.4 kilobits per second and no sooner had they written that statement then our our customers were buying 56 kilobit per second modems at the store so their their Theory had broken down how did you go from there then to three com I mean when did you see the commercialization opportunities and and what uh did you try to persuade Xerox that there was commercialization opportunities um so some of this is personal I left Xerox because it was my time to go start a company not because I saw a sudden commercial opportunity for ethernet but it took a few months to find uh to get deck until in Xerox um expressing cooperation and that looked like huge potential to me so that's when our company started focusing on serving that market ethernet compatible products you've had a great career in not only in network computers but in software uh you're a co-founder of Claris you're a 25-year venture capitalist with Mayfield what did the whole experience of of ethernet and being involved in the early days of ethernet um how how did that sort of form your philosophy you know standards is one probably you that became really important in your mind I would guess but how did it teach you about business and and ecosystems but well you know having that standard out there sort of made networking possible so the fundamental rule of networking is that you've got to be able to communicate with each other but from my perspective when I left Xerox a few years after the spec was published and was a variety of startups but eventually as a venture capitalist I learned what is it that drives Silicon Valley and you know some people say cheaper faster better some people call it the innovators dilemma but for a while everyone was waiting for the year of the land you know is it 1982 is it 1983 it's in 1984 but at some point the year of the land came and went because we had sort of crossed the Tipping Point but I think as an investor you were always looking for that innovators dilemma where the older company couldn't adapt to what the newer companies wanted to do and it always had to do with performance and configuration coax cable gave way to a 10 base team what you think of ethernet now CAT5 Jack that plugs into one of your devices and then going from 10 to 100 to a thousand so there's a pattern that repeats itself but the pattern works when there is a standard that allows entrepreneurs to be creative in conforming to the standard but Breaking All the Other rules that don't apply to that standard Bob you were sharing in The Green Room an interesting story you know we're going to hear on the next panel from Judy Esther and Andy bechtelsheim the first time you met Andy Bechtel shime can you share that story sure although he and I don't agree that it ever happened uh in fact Andy came to 3com which was a brand new company and uh and Bill Krause had arrived so we were focused and so Andy said listen I have this workstation and I don't have anyone to build it and you guys build stuff here so would you like to build this workstation for me and Bill and I we were focused and we said no we're focused and we'll we'll sell you ethernets but we won't build your workstation so Andy walked right out of the door and founded Sun microsystem we'll hear the other side of that unofficial story that doesn't exist on the next panel um the uh as we come to the end of this panel not only I mean we're here to celebrate the 50th anniversary of of something you and Dave Boggs brought into the world but you also brought in as Gordon Moore did a a rule that turned out to be a really good rule of a way of predicting them oh not a rule well a law a law I I thought a man I thought a man of your I thought I I thought well no I'm talking myself into a cul-de-sac synonyms talk about the origin of metcalf's Law and what was uh well described for people who may not know what it is and when did you first come up with that one so we began uh shipping ethernet for IBM PCS and one of our problems was that there weren't any IBM PCS so they were very small numbers so we developed a kit a three node kit that you could buy for three grand and it was three ethernet cards a piece of cable that could connect them together and a discount with software on it and the software did pfmts to make a long story short print file mail pfmts and so we started selling these things for three thousand dollars and the customers came back and said it does everything you promised it's just not useful so at well that's pretty bad so I was I was head of sales and marketing I went over to Stanford where there was an alto and I made a slide this was uh before PowerPoint I was on the board of the company did PowerPoint we sold it to Microsoft in 1987 for 14 million dollars but this is before that so I made a 35 millimeter slide that said the cost of the network goes up linearly with each card of ours that you buy but the value of the network the number of possible connections is quadratic and it goes up as N squared so there's a point but after which a critical mass point where your network will be extremely valuable so you and what is the remedy to this problem it's to buy more of our product and and they did no they did and we went public shortly thereafter and then in 19 It was 95 George Gilder saw this slide and he called it metcalf's law that the value of a network grows as the square of the number of attachments and I've been defending it ever since it's a big hand thank you yeah this is fun are we done are we leaving yeah you're coming back thank you oh thank you Rich yoga and Bob now we begin our second panel as much as ethernet is a story of technology is also a story of massive commercial success and so our next panel focus on the companies building the companies and the industry according to IDC in 2021 revenue from just one part of the market ethernet switches topped 30 billion dollars in in 2021 so how did ethernet triumph over their competitors how did they open IEEE 802.3 standard become so vibrant as it is today what contributed to the founding and growth of generations of ethernet related companies and what lessons if any might be applicable to today's innovators and entrepreneurs here are our next two speakers five numbers first we have Andy becklesheim he gave us five dates 1976 was the year he hears about ethernet at Carnegie Mellon 1978 sees ethernet at Xerox Park and starts the sun University Network project 1982 co-founds on Microsystems the first to bundle ethernet on every product 1995 co-founds gigabit Ethernet Pioneer Granite system which was acquired by Cisco in 1996 and 2004 co-founds Arista networks which is a market leader in Cloud networking he's a Serial a successful serial entrepreneur who's been founding and building networking companies for over 45 years and he's been an angel investor and a mentor for many companies including Google and today he's co-founder chairman and chief de Mountain officer of Arista networks please join me in welcoming Andy next I'll introduce Judy Estrin and here are her five numbers 21 was the age as a Stanford graduate student when she worked on a team developing TCP eight companies co-founded 2008 the year her book closing the Innovation Gap was published six corporate boards she served on including Disney and FedEx and 32 years in her most important role as mother to her son David she's a Serial entrepreneur yes including as co-founder of bridge Communications a pioneering Network company acquired by 3com she served in many boards and executive roles including scto and Cisco and today she's CEO of jlab's LLC please give a very warm welcome to Judy esper [Applause] when I think of you Judy Esther and I think of almost royalty in the world of Silicon Valley and and networking and your parents being computer scientists at UCLA and you're being working as a very young person on tcpip but you founded eight companies having to do with networking what explains the success and the endurance of ethernet why are we here celebrating it and why will Verge be talking about the next 50 years so before I do that I just want to say thank you to the museum for hosting this talk about Nostalgia but also to Bob for not only uh the spark the design and your persistence and your charisma which clearly is what uh created the spark for an industry and for my career and I also want to thank yogan because uh yogan York continuous thoughtful contributions to the industry and my life have gone from when I first worked with you or studied with you at Stanford so thank you to you guys first I also just uh need to say as I get have listened how it's so clear how history those of us who report history come from the perspective that we sat at the time reporting it because I realized that my remarks are not uh conflicting with Bob's and yogans but just a very different perspective because my career was built on that creation and in terms of taking ethernet and turning it into so many different functions and bringing those different functions to um to the marketplace and so I was thinking about uh what made it successful both in terms of technology and a lot of non-technology uh issues and some luck and I also just want to acknowledge how many people in the industry not just those of us who are Engineers but the sales and marketing people the Ops people the investors and the customers who all together we took this thing and and made it work and from a tech perspective the Simplicity of the design that you heard yogan talk about was absolutely critical the layering that allowed us to build different Technologies underneath the the packet format was critical to the success and the foresight of whether it was the 48-bit address that yogan was part of or the modularity one of our first deals at Bridge one of our first big deals was that Southwestern balance St Louis but we had to support for fiber and we didn't support fiber there was a codenol transceiver that we plugged into our systems and we were able to bid the deal what all the testing though all of our lab testing could not um guess the uncertainty of humans because when we installed it several of the offices didn't work why because they had been painted the week before and spray paint got into the unterminated fiber connections that simply didn't work so um it those types of experiences um are were important in terms of the flexibility but that experience also was another key non-technical thing that helped make the technology successful there was no consumer Market in those days we sold to Enterprises and Enterprises created the not the standards in terms of the network but the standard level of we had to make it work and interconnect their equipment it wasn't them meeting our needs it was we had to make this work in their environment and I think that created a system that had longevity the standards that Bob talked about whether it was the de facto standard dicks that he in those days said you don't need standards we have a standard but many of us went on to uh work on the IEEE 802 standards which became critical I I have a view of the culture at that time of being intensely competitive but very collaborative and we knew that it was we had to work together for our shared Vision against IBM or at T to move forward so lots of entrepreneurs and last the VCS that invested in US didn't demand moats they demanded differentiation so they allowed us to work with each other and still survive and then we've got to recognize that the success of ethernet enabled but also went along with the success of Internet TCP network computing and it was a time when we moved from vertical uh proprietary systems to broad-based horizontally integrated open systems unfortunately with mobile and social we went back to vertical integration but that time of innovation entrepreneurs could Thrive because we could all build things sell them to customers in a compatible way I'll stop there andy you are from Germany and you started your formal education as a college student Jeremy but you wound up at Stanford as a grad student and Bob tells the story of you visiting three combat but I I believe in reading your background you visited Xerox Park in 1976 when you're in the Stanford program can you describe what you saw yeah so actually it came as a graduate student enjoyed Carnegie Mellon news in Pittsburgh where semfullo's I believe you tonight was my advisor and Sam was a graduate of Stanford and he went back there once in a while to see what's going on so he came back from one of these trips and he told me here's in the future and it's called ethelet and I go like what's ethnet well it's this yellow cable and you can just stick the adapter Center and then the computers can talk to each other it's not like a really really good idea so then I I transferred to San Francisco 77 and it was lucky enough to be a what they call it a no fee consultant to Syracuse Park which means they didn't pay me anything but it would just hang out there and pretty soon that's a true story I took classes during day but I was spending all my nights at Xerox Park and the security guard was like wondering what is this guy doing is it coming in at eight o'clock and leaving after midnight every night but you had to sort of experience this environment for yourself to really understand what this meant you know for the future and I have to admit that the whole Sun Works range there was actually started from that experience of spending all this time there and it was clear to me that every you know future software developer or engineer or you know cat designer would need one of these machines I wasn't focused on the office workers for which the machine would have been too expensive but rather they the productivity working uh gains in engineering well I want to come back to both of you but we have a clip so so the internet story hang on for me when I started the the sun Direction stand at Stanford in 78 with first basket Association that was my advisor we were waiting for the 60 000 chip which was late so with the first card which is end was a three megabit card to be able to talk to the auto computers that serious graciously donated to Stanford and they also had a laser printer so the very first use case was actually connecting other equipment to the laser printer so people could print these large documents at high speed we have a clip from Gordon Bell now Gordon Bell is here but if we could play the Gordon Bell clip because I think this whole issue of standards of of deck and Intel and and Xerox was a really important one so Gordon Bell clip I've always been a kind of a standard not the Dax position was we absolutely had to have a local area network and that uh uh it was a kind of a almost a failure company on on the idea march of 1978 uh we had the first meeting with uh uh uh a deck Intel and Xerox at Xerox Park and that was the beginning of the formation of uh of the creation of the ethernet standard that ultimately was completed in September of 1980 I think the main thing the Xerox brought uh it's patent it brought a lot of expertise because the researchers at Park had had built a successful uh 2.75 megabit shared wire local area network Intel came in as really uh uh the need to be able to have a volume product so they were betting on the uh a the the smaller machines frankly ethernet is kind of everywhere on everything uh doing every task these cables that's a 10 billion dollar business and then these things there's uh boxes like this that's a 20 billion dollar uh business so so it's really really significant and sort of uh sort of this infrastructure is is so large and so important well given the focus of this panel as Internet the ethernet is a platform for um those 20 billion dollars and much more that have resulted from companies built around that platform let's can you talk a little bit about bridge and then we'll talk about sun and sort of your belief that this was the one that you could bet on well I think uh so one of the things that uh helped the market that I didn't say was that there were two different uh at least two different types of vendors approaching the market so there was Sun PCS building adapters to sell their systems into the market that ethernet was part of Bridge initially sold devices that connected different networks but also connected existing devices that the Enterprise had into the network so the the discussed with which Bob talked about dumb terminals that's what we did we connected those dumb terminals into the ethernet as well as other kinds of devices so but those were very complementary because if you wanted to convince a Enterprise to install ethernet they needed everything to be or one of the benefits was having everything on top of ethernet so Bridge was in the switching routing connect Legacy systems part of the business as opposed to Sun and I have my own Andy story of going to have pizza with Andy before he had started three common we had before he started sun and we were thinking of starting bridge and we had a conversation about whether we wanted to build a company together bill and I wanted to build a network internet company Andy clearly wanted to build a workstation so we did work together but not in the same company well if um Bob day Boggs were the dynamic duo then you and McNeely and vinod and Bill Joy were The Fab Four what well they came later yeah so so the the use cases of um that the original Sun pressure at Stanford were in fact you know uh zero line integration bridge I mean Cisco and Advising the design so I think that bridge as well right the early British products we used your CPU award in Britain so I had this site business it's time for to sell licenses to this you've always had a side business um and there was a company trying to do Lisa printer controllers Etc um but the weird thing to me was that the biggest opportunity was clearly the workstation application because you know you would need one on every desk and it would just bundle leave an adapted to learn to connect and other people looked at this and they were criticizing me there would be no software you know for this product it was well people will rate this software it was completely obvious right that it was useful but most people didn't see it because they didn't go to Xerox Park well I want to ask you about because it relates to this idea of both workstations and workstations that were networked when sun um I think you debuted your plans in 1982 was it the company started and started shipping and I believe that the first tagline was the network is the computer and people couldn't figure out what that meant yeah I don't know who came up with that but it was quite descriptive actually because it was the benefit of being Network connected and some of course ran Berkeley Unix which had the full TCP stick so email in all these functions we take for granted they were bundled as part of the product which was not the case for the Macintosh or the PC at the time what was uh when did you know that you had gotten it right that your your vision of both workstations but Network workstations companies generally aren't born of Immaculate Conception they experiment they've known all starts we had like hundreds of pending orders while it was a student at Stanford it's just a way to build the machines so we started with the backlog actually it's true story that's why you're The Fab Four well let's um I'm going to go to a another guest video with um John Chambers so if we could uh connect up to John Chambers I went to one of the major Automotive companies and I met with them uh and uh they said John we're talking to you about routers but there's something in switching going on that you ought to be aware of and you might want to think about should Cisco be playing there and I said okay and uh they explained to me this concept called Fast ethernet I got it I understood it had a great meeting and can they got the orders for the routers and I went back to headquarters uh and then two weeks later I was up with Boeing and the first company was Ford and the second company was Boeing I was in the meeting of Boeing and I was working on a major router opportunity and they said well John there are a lot of changes going on uh in the industry and are you aware of uh ethernet and fast ethernet's potential role and I said yes I am I think it's very likely to be the future here are all the positive things that I've learned two weeks before that and uh and then I said can I have my battle order for 10 million and they laughed and they said John if you buy this company called Crescendo then we could really talk and as simple as it sounds it's about listening to your customers understanding what was important to them everybody at that time thought 80 am to the desktop was going to win all the smart money was there including our own investment on it yeah yet customers told you that an open standard that moves rapidly was going to be the future and it worked out great for us that's part of leadership but Lessons Learned Innovation if you can do it yourself quickly and you have the expertise do it yourself if you can't do it quickly and the Market's already taken off or you don't have the expertise and it's strategic to you you better do M A or a strategic partnership but then while everybody tends to think it's about the decision and the right uh selection of the technology the real issue is how well M A Works has a lot to do with how you integrate it into your company and do you build a machine that really does that extremely well and are you able to keep the people because in high tech you are normally acquiring Next Generation products which is a nice way of saying your Aquarian people if you don't keep the people candidly you don't get the Next Generation products out so I think ethernet will be one of those major technology changes for the whole internet that's going to have huge business implications and major benefits from that I think it is one one that is a model for others to learn from and I think it's got many legs left each time we thought ethernet has kind of achieved its maximum uh in what Bob and Judy and so many other people have done it keeps having a life after that a life after it so I think Ethan has been a huge amount of the success of the past but I would argue it will be a huge amount of success to the Future Bob it's rare that somebody has a chance to make a difference in a huge way not just on technology but technology that changes every aspect of Our Lives thank you for your leadership thank you for your openness thank you for your courage and thank you being a part of this future [Applause] you've uh you've been a founder or co-founder of eight companies you've been an extremely successful corporate board member and here I'm going to inject a one-minute tale about how valuable Judy is in the world outside of Silicon Valley I wrote a book in 2014 called the soft Edge and it was about enduring corporate cultures and one of the people I talked to was Fred Smith the founder and CEO and chairman of FedEx and he FedEx of course it's trucks and airplanes but it really is almost like NASA level you know computer rooms that manage the logistics and all of that you can imagine probably outside the U.S military there isn't anything more logistically complex than FedEx or UPS or the like and they were built on a Mainframe background in the 80s and around the uh late 80s he knew he was hearing more and more and the CIO was a guy named Jim Barksdale we're hearing more and more about network computing and so we sent Barksdale out to Silicon Valley and and he said we got to understand this networking network computing Revolution because it's going to change our business and if we don't adapt to it somebody's going to run by us and he said to Barksdale and I quote find me the smartest guy in Silicon Valley I want him on our board so Barksdale came back with Judy Estrin [Applause] and she served 21 years on FedEx's board in Fred X Fred X Fred Smith of red FedEx said she was the best board director we ever had because they helped FedEx make that transition from Mainframe Computing network computing so I want to ask you about what is it about sort of ethernet and network computing that gives you an understanding of how businesses and societies work and and organizations work at their best it's a philosophical question but it seems to me it's a question that in addition to having the right technology and the right standards this idea that it it is the right philosophical answer to a lot of challenges is very very important and talk about that so uh my perspective of how networking influenced my leadership philosophy and how I operated within those systems I will say that not every successful company has that or every successful leader there's no there's no one way is number one but it's clear that during the time that I was on the FedEx board I also was on the Disney board starting a little later going through that transition that companies needed to figure out how to embrace an open systems philosophy and it's a very different philosophy I mean I realized that my leadership style on boards and as companies is I lean towards favoring Distributing power I want to interconnect diversity for power as opposed to top-down autocratic Styles now there are very successful autocratic leaders but in the system one of whom rejected your offerings from bridge yes yeah I would not have necessarily put him in the top-down autocratic he had his autocratic moments but he also had his uh hiring the right team moments do you want me to tell that story is that why you accepted that um you can take it whichever Fork let me just finish I think that there is a philosophy around open systems which is interconnecting diverse systems that are all that are good at what they do FedEx grew out of acquiring different companies that had Legacy systems and they needed to interconnect those systems uh without dropping they needed the 99.999 uh availability to deliver packets but I think there's another piece of it which um had to do with the layering of Internet the of ethernet the internet does the same thing which is assuming that uncertainty exists acknowledge uncertainty and acknowledge that things happen and plan for how to fix it when it happens and so the layering of ethernet allowed reliability at a different level and large companies have to figure that and it is the opposite of a deterministic view IBM's philosophy was very much about deterministic there are people who believe they can make the future happen a certain way I believe that humans are wonderful we're uncertain and that we need to plan and acknowledge for that and I think FedEx and the other boards I was on and I will the last thing is as a board member I'll go back to yogin's description of csmscd you need to listen before you talk and you need to listen as you're talking because you need to be sensitive to what is going on on a boardroom there is no one person in charge there's a lot of Egos and a lot of conversations and so I brought csmacd into the boardroom with me as I was 20 years younger than anybody else and the only person from California in in Memphis board meeting so it was a fascinating experience and then I'll quickly say that when we were at Bridge we decided we wanted to try to expand past our system business and see if we could sell our ethernet to some system vendors and I went and had a meeting with the Apple team and we wanted to convince them to put ethernet as an option in addition to Apple talk and Steve was at that meeting and he basically looked at me and said why would I want to put anything this bulky and this expensive in my computer here his Aesthetics were showing even then apples only need to talk to Apples why would I do this and so uh I knew I wasn't going to win that order and eventually obviously Apple opened up a little so Andy Not only was Sun's a spectacular success but your posts on Career both as an entrepreneur and investor have been remarkable Granite Systems Arista and then one of the first two outside investors in Google so I'll ask you kind of the same question I asked Judy it it clearly you're doing something right clearly you're able to recognize things in the in signals in the noise that that have proven to be really valuable signals so is the sort of the ethernet networked way of looking at the world a very useful way of interpreting the world and organizing the world well I I find it really hard to compare Network protocols to either corporate decision making or investment decision making so I'm not sure that's a that's a useful way to go but going back to Easton for a second here to me you know what really enabled the overarching success of ethernet over time is quite frankly its original Simplicity and flexibility and you know those of you in the room who've worked on various attributions may know this but there's almost 100 different ethernet specifications they can all talk to each other but there's a you know 10 megabit 100 megabit gigabit two and a half five gigabit 10 gigabit 25 40 50 100 200 400 800 and pretty soon 600 and then they go over you know twisted your copper fiber all kinds of Optics it's just amazing how many use cases could be covered you know with this same packet format and the same you know underlying Network principles and this is in the I mean if you compare the token ring which was like one spec and that was supposed to solve a problems it was clear the token was just you know landlocked into its own view with an Ethernet that the real breakthrough was when you know people realized they could go 100 megabit instead of just 10. and there was this uh some people may know the story but there was an informal agreement between the I Tripoli and ANSI ANSI owned ftdi the fiber distributed Network and the agreement was that ANSI would do every speed at 100 megabit and greater and I truly would everything below 100 megabit which didn't make any sense in retrospect but that's why they actually didn't think of 100 megabit originally at ethernet it was actually Crescendo the company that did the original cddi fabric Channel over copper development and then they realized it would be much more useful just to send the ethernet packets over the same 100 big copy interface right and that was the birthplace of 100 megabit twisted pair and that was followed just within the Furious that there was gigabit Ethernet over twisted pair and pretty soon 10 gigabit and the whole rapid Evolution started actually in the early 90s so the first I don't know 12 13 years of ethernet was fixed 10 megabit and everybody thought that was the final answer but the this proliferation and all these different use cases and the beginning of the switching industry that allowed these things to be connected was actually a much more recent phenomenon basically the last 25 years all from something that didn't work in theory but only in practice let's give our panelists a big hand thank you thank you so much to Andy and Judy and Gordon and John Chambers so we're just going to change the set here as we get ready to bring up Bob on stage we're really thrilled that for our final segment we're going to have a chance to hear a keynote from Bob and then in his inimited style he said let's turn this into an AMA and ask me anything so you have these questions on cards on your your chairs please go ahead and write your questions that you have from and a team member will come by and pick those up so here's a chance to welcome Bob Metcalf back on stage the single most important new fact about The Human Condition is that we are now suddenly connected since October 29th of 1969 5 billion people have gone on the internet two-thirds of the human race and it's affecting everything we do so I'm out advocating that connectivity be a tool that everybody has connectivity is extremely powerful tool and now one of the problems with connectivity is we have been so successful generating it we have too much we don't know how to handle it so there have been a series of pathologies that have come along from not being able to figure out how to handle connectivity the first one was hacking I wrote the first memo in Christmas of 1973 reporting two high school students had hacked into the internet of the arpanet and uh the hacking has not stopped it continues to this day is that right then came pornography they almost shut the internet down because it was carrying pornography and so the solution was to pass the communications decency act and as you know there's no more porno on the internet and then came advertising so advertising cropped up I think it was a couple of lawyers in Arizona but or was it a t I'm not sure started advertising on the internet but those of us who had built the internet were above that I mean advertising was a crass commercial activity and we'll be damned if advertising will be on our Network a few months later all investment and all Opex and capex all the money going into the internet was coming from advertising so we learned that advertising was the at least for this epic the machine then when we've fixed when we realized that we still had spam and spam was a was a a pathology and it's pretty much under control we don't talk about it much anymore so now we have fake news which is there are several of them but that's my favorite latest pathology so because we haven't figured out how to use all that connectivity we've generated we're getting fake news and there are all sorts of solutions to that you know for example have the government filter all news for us it's a great idea or better yet have uh meta have them sort all news for us anyway we're going to solve uh this this pathology just like we've solved all the preceding ones and here and here's how we're going to do it and you heard it here first branded composable filters so the government isn't going to filter it and meta isn't going to filter it we're going to filter it and we're going to have a tool to help us filter it and that tool will allow us to compose filters the New York Times a little bit of New York Times a little bit of this a little bit of that so branded composable filters that's eventually going to be the solution to that pathology would anyone like to ask a question please okay uh 1971-73 it was a massive collection of innovation to change the world how do you see and check GPC in large language models put that in perspective the the title of my undergraduate thesis written from Marvin Minsky in 1968 a Neuron model and some of its information processing capabilities sadly all copies of that dissertation have been lost so I've been watching AI come and go come and go come and go over the years of the Decades longer than ethernet and the principal problem has always been the models run out of data and without data they lose their predictive power and AI just goes into remission for a while there's a chance that this time that's not going to be the problem that is because we have the internet and we have all this connectivity that I've just mentioned so the the data underpinnings of the AI modeling has a chance this time because of the availability of all that data on the Internet it's not a sure thing but that's so my model is optimistic that the data that on the internet will feed AI this time adequately but I don't know um much about chat GPT other than the fact that the Magna Carta was signed in 1066. uh anyone another question hey thanks hey Bob congratulations on the touring award it's just terrific I'm not allowed to ask this question but how many smoots really is across the Harvard Bridge sorry the real question is this um I moved to silica Valley in 81 and joined venture capital in 88. and one of the concepts we talked about was when new real estate is created that's when you go out and build New Deals new building real estate is the most valuable creation Mankind's ever come up with and Ethernet or networking or just connecting everybody is like real estate we can connect more things we can build things on top of it you enumerated a bunch here porn advertising and stuff like that but what's next what's the future of our connected world what applications would you look in would you be looking for what what can we do with this now that we're all connected all seven billion of us so as you as you knew when you asked that question is a really hard one uh but one one of the big unresolved things is we we had in 2019 it is as if the internet was built to handle covid in 2019 which was the 50th anniversary of of the internet happened to be 2019. so by the way I'm proposing that the acronym covid mean collaborative video it's and in all seriousness that's playing itself out now we don't it's it's the one of the big futuristic questions is is everybody going to go back to work now and sit at their desk as they used to no is everybody going to spend all their time at home you can't work at home there's just too many interruptions I'm sorry I have two dogs I can't wait to get out of the house and leave them alone but uh I'm staying at a hotel here in Mountain View that's next to next to an office building the office building is completely empty it's a see-through and downtown Austin has a few see-throughs in it including a brand new Google building which as far as I can tell is empty uh so that's plenty and I don't know what the answer is but an answer your question we're going to find out how that's going to play out how real estate is going to adjust to the connectivity so we have all this connectivity uh and um okay so there's one potential answer to the question about the Future Real Estate another question please save me some we have some questions that people have written in here so um we'll take a break from the future somebody wants to know about going to the past 50 years ago the influence of Aloha net on ethernet so aloha net is a wide area network radio network that a famous paper was written about in 1970 by Norm Abramson and at that moment we were designing ethernet and the uh one trip to Hawaii convinced us that our land our ethernet land would not use radio because we saw the modem and the modem was about this big and it ran at I don't know 9.6 kilobits per second so we Lo we uh gave up on the goal of having zero wires and chose to have one wire the zero wire thing took 25 years it's called Wi-Fi but it took 25 years to get the chips necessary to go back to Aloha Network that is a radio network but there was another feature of a low Network that we really loved which was randomized re-transmissions we we were looking for an access method that allow us to put packets on the network without a lot of apparatus that you could and you and yogan gave us the cocktail party rendition of that a few moments ago so we had uh three key Technologies we had the coaxial cable with Gerald vampire Taps which I'm sure you all love and then we had uh uh Aloha Network using randomized re-transmissions and then we had Manchester encoding that is you could turn that cable on and off and so if you wanted to send a bid under Manchester encoding you send the bit and then you send the complement of the bit and that guarantees that there's a a transition in the middle of the bit cell and you use that to recover clock which then toggles those bits into a shift register and puts them in memory so three Technologies Gerald Taps Manchester encoding and Aloha randomizory Transmissions but by choosing we chose all three of those but by choosing coaxial cable instead of radio we ended up running a 2.94 megabits per second and not only that because we were using cable we could do carrier sense which Aloha Network wasn't doing and we could do Collision detection which a lower Network wasn't doing and not only that we were running thousands of times faster than Aloha Network so we thanked and we to this day thank you and I always get a question on the Aloha Network I don't know what it is somebody out there thinks that I'm I don't give enough credit to the alola network but we're very grateful and I spent a month studying this with Norm Abramson the University of Hawaii and Manoa campus and so we incorporated a lower Network as an access method but I want to uh so I've listed the three key Technologies which were ethernet at Park in 1973. none of those persist they're all gone we don't Pierce cables anymore we don't use Manchester encoding anymore and we don't we don't use randomizory Transmissions much there hasn't been a collision in decades so what did ethernet actually deliver if it weren't those if it wasn't those three Technologies what was it and here they are uh first packets to the desk until Ethernet there were dumb terminals I'm sorry Judy dumb Terminals and we were transitioning to PCS and those PCS needed the packets to come all the way to the desk not stop at the host which is where the arpanet started the second thing is abundance of bandwidth 2.94 megabits per second is 10 000 times faster than the network that it replaced on my desk uh not 10 you thought I said 10 didn't you no no 10 000 times faster so following ethernet band uh the network had abundant bandwidth so it made sense to upload cat pictures and and finally ethernet joined the internet standards bandwagon so you had TCP you had FTP but you also had ethernet made a standard by the IEEE the IEEE committee is one of the most effective organizations on Earth and they've been at it since February 1980 at the Jack tar Hotel in San Francisco which is gone now but 802.3 is still humming along and I'm I've just visited there the 802 3 meeting just to say that just to say you guys are great keep it up um another question um when we finally are able to beam up Scotty will it be over ethernet next question okay an easy one what's your opinion on of Elon Musk my hero used to be Steve Jobs and as you may know he passed away so he can't be my hero anymore and the new Steve Jobs is Elon Musk and I I am rooting for him I have a Tesla uh our Island camp in Maine is served by starlink this guy is phenomenal and I'm a big fan of his and I I defend him every chance I get just like yes [Applause] yes well a more silver variant of the beam me up question here is what is the maximum speed you envisioned when inventing ethernet and how far can it go the maximum speed spanned with capacity of okay so the current committee is working on a standard for 800 gigabits per second 800 gigabits per second now they cheat they have this is on optical fiber and they have Lanes I don't know 200 200 gigabits per second Lanes I don't know the exact number but they it's multiple wavelengths on the fiber and I keep demanding gigabit I'm sorry terabit ethernet on one Lambda on one lane that's been my goal but uh they're still catching up to my goal but they're they're aiming for 1.6 terabits per second in uh with multiple Lanes that's really a lot faster than 2.94 megabits per second by the way would you like to know why it's 2.94 megabits per second when we were get when Dave Boggs and I were given the challenge of building this network and made David rest in peace we were given 60 dip dips chips to use from uh the Texas Instruments catalog and a card that could hold 60 of dual in-line 14 pin packages and so that was our limit so we started putting function you know my memory and fifos and clocks and everything on there not clocks and it came time to clock the packets out onto the network you have to tick tick tick tick push them out we didn't have any more room on the card so David says wait a minute Bob there's a clock on the back plane of this Alto personal computer let's just run a wire over there and use that as the clock for our packets and uh it we did and it worked and it had the advantage that unlike many other communication systems including some that are 99.99 reliable there's no clock synchronization required among the stations each packet has its own clock encoded right in it and that clock came off the back plane well the system clock ticked every 179 seconds and if you Manchester encoded that means each bit is 340 nanoseconds which turns out to be 2.94 megabits per second but then journalists say it's three megabits but why don't you just say three simple the Intercontinental trunks of the internet in 1973 ran at 50 kilobits per second the rounding error in going from 2.94 to 3 is bigger than the Transcontinental Links of the internet so that's why I don't round It Off [Applause] I wonder if you could share this question question or ask some stories about the role of Bob Khan in all of this you know efflorescence in the 1970s tcpip driving very large system integration of workstations SGI Etc so Khan is a is a perennial intellectual of networking thinking way far out in deep thoughts so he was around when uh in 1973 Vince Irv held a had a conference at Stanford to plan the next generation of protocols after the NCP protocol Network control protocol needed upgrading by the way there's a funny story about that uh vincerf and I both attended that seminar wanting to have a network number the NCP packets did not have a network number they just had a host number so to expand it we want we both wanted a network number but we discovered at the meeting he wanted to connect a network per country and since there are less than 255 countries he thought one byte would be enough for a network number but I was attending the same meeting building ethernets in which there would be multiple ethernets for building we I needed a huge field for the network number so we compromised on 32 bits and it's worked pretty well so far but the the hilarity of having the two of us we both knew we wanted a network number we just wanted completely different things that's all uh you didn't find that hilarious did you no but I found it deeply philosophical and and um what was the word you used to describe bobcon oh so so saw that yeah and then he he partnered up they both lived in Washington DC and he immediately grabbed hole of vent and they started working on the successor protocol to NCP which was tcpip famously so and they wrote a paper together in 74 proposing this next Generation so Khan has been over the years decades has been a consistent intellectual about how you should do networking here's a technology question that stretches out into the future can ethernet interconnect silicon to quantum computers I don't know oh but here's something I know going back to the question about chat GPT I've looked at the structure of the of the neural networks uh there's great potential for expanding the connectivity of those neural networks so that's the hope of improving I I find Hope in approving the the AI machines not So Much by improving their loss functions but by improving their connectivity because the the brain the real brain has much more connectivity than the neural Nets that we're currently playing with here's a question you've probably been asked a million times why did Xerox not commercially take advantage of all the great technologies that were developed at Park that is really a boring question uh my but my answer which I've given 9 000 times is Xerox was a monopoly and we work there I love the way my colleagues use the third person when they refer we were working there guys it was our fault whatever happened and my belief is it was the Monopoly mentality Xerox was could had a monopoly on copiers but didn't have a monopoly on workstations or so when it went into that market we got our we got our faces eaten by a highly competitive environment so it thought it could as a monopoly it thought it could control the pace of development and roll out yeah and adequacy of product was wrong Steve Jobs loved Apple talk because it was adequate it only ran at 250 kilobits per second it was a lot of bad features but it was adequate and and ethernet was uh for the stations at that time was over engineered it was more than was needed um this is a very end of a question asked before um but but a lot of these cards kind of go into buckets that are sort of similar and that is looking to the Future does Ethernet have any are there any natural successors to ethernet or does Ethernet have the evolutionary you know long Horizon to to be able to uh just keep going so let me tell you a secret we've been cheating yes every time a new technology emerges to supplant ethernet they hold a marketing meeting to decide what to call it and every time they call it ethernet because they're not stupid and you know Ethan is a very powerful brand so if you have a product you want to I remember HP once came out with a fast ethernet and I I was writing a column for interworld at the time and I studied what this new technology was and it it was a hundred megabits per second but it was an Ethernet so I uh called him up and I was called them out in my column I said this may be a lovely technology but it ain't ethernet and uh it's they changed its name to 100 VG any land you see the ethernet would have been a better name if I wasn't there to give him a hard time well we are coming to the end of this session and by tradition the Computer History Museum likes the honored guest to be able to share summarize their vision and share it in one word can you do that for us as oh you've caught me completely by surprise are you ready yeah are you ready you let the cat out of the bag that that's the real explanation Behind metcalf's law selling you're selling yeah yeah I have high regard for sales people and selling I wish I had learned selling before I started three come rather than after I started three times and uh so I tell my whenever I have students I tell them the importance of learning how to sell and I'm willing to share with you the secret right now the secret of selling I could I can tell you right now would you be interested in knowing the secrets listening and listening I learned this bill Krause is here so Bill joined uh 3com and he came to our first Monday morning Ops committee and as our new president it was his job to run that meeting I had run it before but now it was his turn so I sat through the meeting and I watched bill who's here are you here bill yeah there was Bill sitting in this he won't remember this he'll deny this story too so I'm watching Bill writing on this yellow PAD as the meeting progresses and I'm intensely curious I'd like what does Bill know that I don't know I think I'll go look so being chairman I had the it was My Prerogative to stand up and walk around behind Bill where he was taking notes and this yellow pad had written all over a DNT DNT DNT do not talk so that was the first of a thousand things that bill Krause taught me about how to run a company and as you can tell I still have trouble listening thank you all right well thank you both um let's just take a moment again and thank for everybody here it's [Applause] um what a privilege to be here tonight so I want to thank Bob and Rich for that and earlier Andy Judy yoga and John and Gordon Bell as well thank you all for what you've done um I'm struck by a number of things and I don't know about you but most importantly um I think it has a lot to do with the people and that I think Bob's closing remark about selling you know if you're not buying or selling you know and for those people who believe that Christopher Columbus discovered America someone sold them a boat right I mean I'm not sure he discovered America but somebody sold him a boat so I'm I think sales is a good thing too there's no question about that there's a QR code that'll pop up on the screen in a minute and as an institution we're working really hard to understand the implications of our work and the take away that you have as it relates to understanding past present and future and the broader implications of computing on The Human Condition so please take the time I'll speak a little bit longer for a moment so you'll have a chance to capture that and give us give us the data give us a point of view uh what you picked up and what you've learned and how you think about it um and if you haven't become a member please consider if you go to the website there's a join a give bucket and you can go and take a peek and take a look at that and when you're there what you'll notice is we have this brand new collection it's actually not so brand new but it's new to the web and it's all the Xerox Park archive it's up on our site it's an incredible collection this past year we've been working on source code releases as well we've had the Apple Lisa source code PostScript source code among others so it's it's an exciting sort of evolution for us to do a lot more digitally there's also a collection The pelkey Collection that's on the web that's the definitive source for internet related history activities so please take a look at that if you like um what's going to happen also I want to comment our partner The Verge they're celebrating the ethernet 50th Anniversary if you go to their website you can take a peek at that as well um at 8 15 The Verge will be broadcasting here there's a special vergecast podcast inspired by everything that's 50th and if it's on if you're online you'll be able to pick it up later this week and it's taking place here in this Auditorium so if you'd like to talk with the speakers when we're done closing the program they'll be out in the lobby on that side so oftentimes people will gather in the front but instead we're going to be in the lobby with the speakers but I encourage most of you uh if if you haven't had a chance to take a look at what's going on with our partner at the verge so in closing tonight I wanna because this is this is part of History what we're doing tonight we're capturing this and it's part of our archive so I'm going to invite all of the speakers who were available tonight to come back up on stage for a for a photograph if I may so great thanks [Applause] so um the next thing I can comment on is people are joining us on stage is that on June uh 20th we've got a program on the future of news if you start to think about advertising the things Bob was speaking about we've been working quite hard for the better part of a year Gathering domain expertise and defining the problem of the future of news so that'll be our next program on stage join me in thanking our panelists thank you

Channel: Computer History Museum

Views: 1,199

Keywords:

Id: T9On2L0-ObU

Length: 90min 19sec (5419 seconds)

Published: Thu May 25 2023