Algorithms + Data Structures = Programs - Episode 59: Otto & Eric Niebler! (Part 2)

Episode Date: January 7, 2022

In this episode, Bryce and Conor continue with part two of their interview with father and son Otto and Eric Niebler!Date Recorded: 2021-12-02Date Released: 2022-01-07The Remaking of Wall Street, 1967 to 1971NYSE SpecialistNYSE (New York Stock Exchange)General ElectricWorld’s Fair1964 New York World’s FairADSP Episode 48: Special Guest Dave Abrahams!Boost C++ LibrariesSwift Programming LanguageIntentional SoftwareCharles SimonyiP2300 - std::executionUNIVACWorld’s First Computer BugIBM Personal Computer XTPongCppCast PodcastIntro Song InfoMiss You by Sarah Jansen https://soundcloud.com/sarahjansenmusicCreative Commons — Attribution 3.0 Unported — CC BY 3.0Free Download / Stream: http://bit.ly/l-miss-youMusic promoted by Audio Library https://youtu.be/iYYxnasvfx8

Can you that's a good question? It really is. And I've thought about it many times. I'd like to say I was pressing it. Welcome to ADSP, the podcast episode 59 recorded on December 2nd, 2021. My name is Connor. And today with my co host, Bryce, we continue part two of our conversation with father and son Otto and Eric Niebler. If you haven't listened to episode one, I highly recommend listening to that one first. I mentioned that the Wall Street area was going through some fairly heavy trauma during the late 60s.

I don't know if you're familiar with that part of history of Wall Street. Wall Street has a tradition of being notoriously old school and reluctant to change. But what was happening is that the stock market was growing exponentially. And between 1965 and 1968, the volume of transactions on the exchange itself went from $5 million to $12 million within three years, which always destroyed Wall Street. Basically, whatever transaction occurred, a stock was traded. They would manually transfer the stock into the owner's name. It was a very manual process. It was on paper. You get a certificate, in effect. That destroyed three or four different brokerage firms. They couldn't keep up with it, and the SEC shut them down. I think the most notorious was Hayden Stone Brokers. They literally shut down that brokerage firm. And then they introduced something called Central Certificate Service,

where they basically had a banking service introduced, but instead of money, they would have certificates in accounts. And to do the trading, they would actually do accounting postings where they transfer one into the other without any paper transfer. And that saved Wall Street's bacon for the most part. But that all required computers. So they all had to get on board very quickly and get all their back office operation translated into an automated system, but not without a lot of collateral damage. They lost about six to 10 brokerage firms that couldn't handle it.

And they could not find anybody or couldn't find the staffing, the technology at that time. And if you were anything close to a computer, they would gobble you up and give you anything you wanted in terms of salary. If you could spell the word computer, you're hired. That's basically what it was. They had headhunters all over the place trying to get technology staff from one brokerage firm to the other brokerage firm. There was piracy on Wall Street just to get the right people.

That's basically how traumatic it was piracy on Wall Street just to get the right people. But that's basically how traumatic it was during the late 60s. Wow. Yeah, it's super interesting to hear because at one point in my career, I wanted to be like a quant and wanted to work on Wall Street. So I've read a ton of books about the 90s until the early 2000s. But it's a it was it's you know it's completely different world like to hear the early days of having to actually physically transfer oh i own this stock and it's a certificate and that was a trading house i couldn't keep up with that and then they would go under like versus now i mean it's basically algorithms and you know the human in the loop is is very minimal it's like you know a trader staring at 12 screens just monitoring what

a bunch of algorithm you know proprietary algorithms they're executing it's hard to even imagine like that time where you know computers aren't even uh in that system like that wall street it's amazing once they learned what computers could do wall street really grabbed the bit that ran with it uh they basically uh uh were uh totally automated within three or four years and but they began to realize the other potential other than being more efficient they also said that they could do a lot more in terms of scientific investing with the computers, doing the analysis rather than having people. So the automation process went very quickly once they realized the upside potential of computers. But it's an industry.

It's an interesting evolution on Wall Street because Wall Street prior to had a very heavy reluctance to automate anything. Another interesting story about Wall Street is automating the specialist. The specialist on the floor of the exchange is a very unique individual. He basically buys when everybody's selling and sells when everybody's buying. His primary purpose is to stabilize the market so it doesn't go crazy. That's a unique feature of the New York Stock Exchange. But working with them personally, they're the biggest crooks on Wall Street. They basically are responsible for maybe a half a dozen stocks, and they basically have to stabilize the stocks.

But they trade on inside information very easily because they know what is going up and what's going down before anybody knows about it. And they all have proxy accounts that they could use very easily. And the SEC is always on top of that. One of the projects I was assigned is automating the specialist book. The book is where you post your trade. If you're a broker, you would approach a specialist and say, when this stock hits this price, sell or buy this stock for me. And he keeps that in a book. And it's a manual book. But we wanted to help him. And I was a very naive manager at the time, and I went down to the specialist and said,

we're going to do a great thing for you. We're going to automate your book. And the words I got back were not very pleasant. To the effect is, like hell you will. And they didn't want any part of it because they feel that having the footprints of what they do available to the SEC is comparable to suicide. They didn't want any part of automation whatsoever.

They did not want Big Brother looking over their shoulder at all. But that type of automation was strictly unwon by the floor of the exchange. So did you end up being successful or what happened? At some point it said, thou shalt automate that book. And that was one of my projects as well. We automated the specialist book. And it turned out to be a blessing because it freed up the specialist, made him a little bit more honest, but also freed him up and made him a little more efficient.

So he was able to do more work and get more money out. So it worked out well for the specialist. It was not without a fight. it's interesting that to hear that the thing that really drove the automation was that that increase in volume that is that the volume reached a point where there was no other there was no other alternative no that was that was a condition of the economy at the time the automation didn't have much to do with it automation basically alleviated the the stress of the added volume. And it was a lifesaver for Wall Street. It really was. But it did not generate that traffic. It was other things that caused that traffic.

So not to switch topics, but a couple of times earlier, you said referring to tech and how far things have come. If I recall correctly, you said in 56, when you saw the computer, you sort of knew it was going to be a big deal in spite of your parents or your dad being like, come on, you should reconsider staring down people's throats and be a dentist. And I'm not sure what the equivalent is today. If it's, you know, seeing something that how, how do you know it's going to have the impact it does? So do you know what led you to think that computers were going to have, like, cause my guess is there was a bunch of folks that saw computers and we're like, okay, like, I'm sure this will be neat, but like now they're they're everywhere what made you think that it was going to be it was going to be this game-changing thing or do you know can you that's a good question it really is and i've thought about it many times i'd like to say i i was prescient about what's happening to computers but i wasn't it was a type of thing that I thoroughly enjoyed doing. I looked at it as a game, actually.

And I said, would people be willing to pay me money to do something that I enjoy so much? And I said, well, I don't know if they will, but I'd like to find out. And I think that's what drove me, not the vision of tomorrow and what role it plays. GE had a lot to do with bringing the computer age into being. I don't know if you remember the 66 World's Fair. They had a beautiful exhibit about technology tomorrow. And that basically brought home the message of what really could happen with technology. I had the vision a lot before that.

Not the vision. I had the urge to be part of it well before that. I think that type of mentality is what really brought people really into interest and looking at it as a career. There is a tremendous potential. And I think one of the big companies like GE and others started talking about what could possibly happen if you really applied your imagination to it. What could possibly happen with a computer? It really boggled your mind.

What were some of the things that they showed at the 66 World Fair? It wasn't just computers. It was just the different types of transportation, different types of automation of the home. It was different types of communication. I can't be more specific than that, but they gave us examples of what it would look like in Tomorrowland. I think they had an actual exhibit called Tomorrowland

and it showed you a whole bunch of things, mostly around the home, around the transportation, around industry. At that time, Dick Tracy's talking watch was considered leading-edge technology. I'm trying to look up when the last actual, because I don't think World fairs were a thing when I yeah so when was the last world

world fair while you're looking that up we should ask Eric about his time as a freelancer because I think that's one of the more, one of the more interesting periods of Eric's career. And this ties in with our other guest, Dave Abrahams, does it not? Yeah, yes, it does. Oh yeah. Yeah. Yeah, actually. Yeah. So right, right, right. So, so I was,

I was working at Microsoft and I started getting involved in, in boost a little bit. And I think I went to some sort of a, maybe I hadn't been involved in Boost just yet, but I went to a conference and at a conference I met Dave Abrahams. And I remember sitting at a bar with Dave just chatting over a beer. And we just seemed to be two peas in a pod. And we've looked at software and the current challenges in the software industry the same way. And I'm like, man, this is a great guy. He's doing interesting things.

And I went back to Microsoft. And I was thinking about it and you know at this time dave was dave was already uh independent and he had uh his um boost consulting a business uh it was called boost consulting then um it wasn't boost pro yet um and uh and i thought you know that'd be a great life you know and i i kind of have this dream of travel and, you know, it would be nice to be free in order to travel and work remotely. This was a long time ago and I was like, hey, you know, if you need another consultant, let me know. I'm not particularly attached to my job here at Microsoft, and I thought nothing of it, and, you know, several months later, I got an email back from

Dave. He's like, are you still interested in being a consultant? And I was like, hell yeah. Quit Microsoft like post haste and started working with Dave. And it was fantastic, you know, because I was a nobody, right? I couldn't have started a consulting business on my own. I didn't have clients. I didn't have any sort of reputation in the industry or anything at all. So I got to ride Dave's coattails a bit. And, you know, he sent me work. I worked away. People seemed happy with the work. I enjoyed it. A lot of it was either proprietary development for companies or open source development that was sponsored by various companies that wanted to

let various libraries built. A couple of the Boost libraries are ones that I built while I was consulting with Dave. And I got very active in the committee. I got very active in boost.org and on, on the, on the mailing lists and, you know, learned all sorts of stuff about library development from all sorts of smart people. It was, it was a great time. You know, I, I, I got a chance to learn a ton of stuff and, and, and rub elbows with a lot of really great people. And so that was a fine time for me. And then eventually I did make myself location independent, not right away. And so I worked, I did a lot of that work while I was kind of knocking about.

You know, I was either in Australia or in Thailand or in Vietnam or New Zealand or anywhere in the U.S. or Canada. Or, you know, I just kind of knocked around and would set up shop for three, four months and then, you know, move wherever struck my fancy and do it over again. And I did that for several years while consulting. That was great. Eventually, you know, Dave got sick of doing, you know, running a business and got a good offer, like an offer he couldn't refuse, so to speak, from Apple to work on Swift, on Swift and to be the person steering the development of the standard library for the Swift programming language. And so Dave decided to snatch that opportunity.

I would have too, I suppose. And that kind of left me high and dry. I was like, well, I was subcontracting to this guy. And now he's gone and for me. I enjoyed that a lot. You know, a lot of the clients that Dave had had, I inherited. So that was great. That made it real easy for me. And at that point, I had a bunch of libraries in Boost. I'd written some articles.

Some people knew, you know, oh, okay, this guy's got a bit of a reputation. He's a C++ programmer. You know, hire this guy. And that was all wonderful. It was terrific. I loved it. I had a really brief stint when I came back from my travels at a startup called Intentional Software, which is a a programming uh they it was started by charles simoni with um you know if if folks don't know charles simoni he is um one of this microsoft ultra ultra rich people he he wrote excel i think he wrote word um And so he started this company about, you know, he wanted to popularize what he was calling intentional programming, which is basically the idea of, like, you know, domain-specific languages. You know, building a little workbench on which you can develop domain-specific languages to target various specific applications.

Which is kind of a cool idea. But, you know, when you have as much money as Charles Simonier does, then there is no real pressure to ever ship a product. And so they just kind of like spun their wheels endlessly rewriting the, the engine. And, you know, after, after I think five, I lasted five months there after that, I was like, okay, this, this is crazy. I actually, I actually wanted to deliver value. So, so I quit that. And, and, and then, then I went back to consulting and eventually got myself a family and decided I needed a steady job.

So I went and worked at Facebook. And that's how I ended up there. That's my story. Yeah, I would love to consult again. That would be that was a great, great time. But, you know, working at a company like NVIDIA has its has its perks. You know, I get to do bigger things because there's there's there's more will. There's more long term vision. There's more funding and more resources. There's more smart people that i get to interact with um and so uh you know like the work that i'm doing right now for instance like trying to um trying to give c++ um an async programming model um that's work that i you know i would have to either not do that or do it pro bono in my spare time if I were still consulting. Nobody would be

paying me to do that. So I consider myself actually really lucky right now to be not a consultant and working at a company like NVIDIA. It's interesting. I entered the C++ world in about 2010 and 2011. And at that time, it seemed to me like there were a lot more people who were freelancers who were doing consulting. And over the course of the last 10 years, it seems like pretty much all of the people who were freelancers have been gobbled up by some company. And it seems like it's sort of like it's it's uh it's this lost era where we had a lot of you know that that 2000 to 2010 era of boost development and we had a lot of folks that were um you know working as freelancers as consultants and uh or at you know small

companies and and i'm sure that i'm sure that there are still plenty of freelancers out there, but it seems like it was in some ways a different time. Yeah, I'm not sure if that's my perception also, but I'm not sure if it's an accurate perception. I have no way of actually figuring out whether there are fewer consultants now than there were then. It could also be, you know, you know, this is like a station in life. You know, when, when you're in your twenties and you don't have any responsibilities, you know, the idea of becoming a freelance consultant is really appealing. It doesn't matter if your income stream dries up for a certain amount of time. But, you know, and then, and you can, it's all right if you have to pay for your

own insurance, health insurance, if you're in the U.S. and you have to do such things, then that's fine. But, you know, as people get older, they end up getting more responsibilities, families, house sort of thing. And then the idea of having an uneven or uncertain income stream is far less appealing. You know, it's sort of, that makes me think of another question, which is, at what point did, you know, software engineering today, being a software engineer is, you know, considered like a really good job. It's a really high paying job. Was that, that couldn't have always been the case, you know, at what point did it go from being, you know, a pretty good white collar job to being one of the, one of the jobs that a lot of people aspire to because it's

so highly valued and, you know, well compensated? I mean, that's a good question. I mean, it certainly wasn't the case when I graduated college. I remember my starting salary at Microsoft was about 40K, which, you know, is not that much, you know, and not that much, you know, and it wasn't even all that much at the time. I mean, it was okay at the time, but it wasn't like, wow. So I think what happened was the internet, I think.

I think the internet and like the first bubble and all of these people that suddenly made themselves super rich in the Bay Area from the first bubble. All of a sudden, tech stock became a hot commodity. People can make a lot of money off of it and and that seems to have driven a lot of the software economy and and yeah so now now programmers you know get are highly in demand and get compensated really generously I mean I definitely know it was like when I, I grew up in the nineties and the, you know, 2000 to 2010. And I certainly had no notion at the time of, of software engineer being, you know, a high paying job or, you know, like a good job. And even when I like got into the industry at first, I didn't realize that like, it was a really good, a good place to be. It was just like, it was something that I was really passionate about. Yeah. That's interesting.

I, you know, we, we just, I guess we just take it for granted these days, but that's, it's not always been that way. Yeah. That's like, it's, it's, it's, it's interesting to have heard Otto earlier say that, you know, back in the day that your dad didn't want you to be a computer programmer. Because, of course, like it was that wasn't even it was not just that it was not a job that people thought was high paid. It was not even something that people thought was a job. You couldn't even take it to university as a subject. You had to take math. Regarding the amount of pay that the system programmers get, it can be tracked to the reliance that industry has on computers. In my days, the use of computers was not considered a critical element within a corporation operation. It became essential, but its growth was gradual.

And as the growth became more and more, and the dependence on computers became more and more, I think the salaries of the supporting personnel, technology, technical personnel grew as well. And it really grew as the dependence on computers grew. In today's world, people can't envision what would happen if there was no computers. In my time,

it was considered an overhead expense. It was not considered that essential. That's in the early days. But it grew as the dependence on it grew, and so did the salaries. I remember you telling me in the early days, companies would buy these enormous computers and have them in display rooms so that people could see them just to impress investors and clients and such.

And they wouldn't even like be using them to do anything they just have like blinking lights and the the whirling tapes and everything like that and you know to that to that point uh one of my first accounts when i worked for univac with a japanese import export company called mitsui located on located on Fifth Avenue in Midtown Manhattan. And I was the account rep on it, the technician to help install the product. And they had a ground floor operation where their computer site, like so many others, were glass enclosed and raised platform, drop ceiling. It occupied half the floor, bottom floor. The other half was on the other side of the aisle, both was glass enclosed. It was all accountants with compactuses. And the first tour I got from this gentleman, who was my counterpart,

we walked down this aisle. On the left was this beautiful computer installation we just installed. And on the right was this glass-enclosed accounting room with at least 50 to 75 accountants using their practices. And I asked him, I says, what exactly are these guys doing? He says, every piece of output that comes out of this side, he points to the computer, is checked by this side. I wish I had a camera to take a picture of that because it represents the entire spectrum of accounting today.

That was the story. It was interesting. That's hilarious. We're going to have to find a link to some sort of article detailing. Oh, you're going to have a lot of research due for show notes. There's like at least 10 things that I counted that you're going to have to link to. You're going to have to go and find that news story about the XYZ stock from the 60s. Yeah, it definitely exists.

I would love to find that. Yeah. It's got to be on the Wayback Machine at some point. Not the Wayback Machine. The internet's not existing. I was just actually thinking that. I was like. You might have to go look at like Microfees or something.

Yeah. Just to warn you at microfees or something. Yeah. Just to warn you guys, there are other articles. There were many mistakes made on the ticker. Not by my computer, not by my program, but this is false transactions that came out that they took exception to. That caused a lot of excitement. So there was many of this type of situation, not as bad as mine, but there were many mistakes made on the ticker that were corrected almost

immediately on the ticker. So that I'm sure got some, some press as well. Would we be looking at like Washington, the, the journal, is that it? The Wall Street Journal, is that what it is? Wall Street Journal, if you go back and track it, you should find it there. They would be the ones who would blow it up the best.

And when was that? That was, what year was that, you think? Okay, let's see. That was somewhere between 67 and 70. It's late 60s. I would think it was after we had the machine installed and running. So I would think it's 1969 would be my guess. But there was hell to pay on that one, I'll tell you.

If I find the exact some some sort of do you think it was your name in the paper when the story came out or no no okay they basically said it was a computer error the word actually actually was actually used i don't know if they said bug it was could be a mistake of the code i mean whatever they say at that time i don't know actually entered the lexicon i believe that bug was coined uh world war ii or post world war ii yeah grace hopper the mark one right the um yeah the moth that got stuck in the electro mechanical relay is that a true story i've heard i've heard it's a true story and then some people say that that's a story that is made up um is it true is it not you can go to a museum and see the actual moth the first bug yeah you can yeah i think it's at

the computer history museum and uh it's certainly certainly at the very least it's from that era that terms from that era i have a good i have a good last question to finish on. So you mentioned too, Otto, that it was not just Eric that ended up, or maybe Eric, you were mentioning that you had a sister. Someone mentioned another offspring that also ended up. My daughter and Eric's sister. Yeah. And do we want to, do we, I'm not sure if it's a dangerous question to ask who turned who turned out better.

You know, you bought the computer. You had two kids. One one father is a relative term. Better is a relative term. Expert politician answer from a clearly experienced parent. We do very different things in in computers um christine is is mostly um um she does a lot of administration and and troubleshooting that sort of thing okay so not on the not on the code writing side no she's never gotten into writing code i've encouraged her to but um but she she doesn't seem to be all that interested in it did she ever ever play around with the computer that you were doing? What was it called?

That was the IBM XT. The IBM XT. Did she mess around with it as a kid? No, she messed around with playing games like Eric did. Pong was a big one for her. What was the other one? A couple of very basic

games at that time. Interesting. I don't actually have recollection of my sister in front of the PC playing games. Not like you. Not like you. Interesting.

Well, maybe in some five-year anniversary or something, we'll do... We're going to have to put together some sort of bingo board. We've just checked off the Father-Son podcast. We'll have to get the Brother-Sister podcast at some point. Someone, actually, here we go. Listener out there, if you're interested in making us a bingo board, we will be taking submissions. And at some point, we will choose a bingo board

and we'll make our way through them. But yeah, this has been an absolute blast for me. Oh yeah, definitely. I think a ton of our listeners are going to have or the listener will have a fantastic time. These kinds of stories are just

it makes my mind sort of bend. Every once in a while you listen to CBPcast, you hear Jason talk about his, you know, PDP eight or whatever assembly that, you know, but like, this is even going back. This is going back literally to before. I think the first year that people say like, what's the earliest language?

Is it Fortran? And like, I think that language is credited with like, is it 1955 or 1957? So to go back all the way to the beginning of that is it's super, super interesting to hear sort of what it was like back then. Cause it's, it's so unfathomable. You know, my first computer was, you know, in elementary school and we were already doing graphs and Excel and like we were word processing and it was, it was already pretty,

pretty good situation. There was definitely no punch cards. It'll be interesting to see. Yeah. I'm sure there's some folks out there that listen to that have similar sort of whether they're wall street stories or other stories back in the day, a million, a million dollars, X, Y, Z. Am I glad that I only write like backend code and a bug for me is just like some manifestation and some Python code or something like that. I definitely don't have newspaper articles being written. I was the person that like filed the bug on Libstd C++ for them to make the ABI breaking change to std list in C++ 11. The one where they made it for one release and then like very rapidly put out another point release to un-ABI break it.

That's like the closest I've come to doing real hard. In for me, yes. Well, this has been an absolute blast. This is really fun. So much fun. Thanks for listening. We hope you enjoyed and have a great day.