00:00:05.460
Hello, hello, yes I've gone full Madonna, which is nice. Right click to start it says on my notes.
00:00:24.180
It's so weird that two of us are using this retro software to give our talks. I brought the system up in less than two days, so I'm a solid medium in terms of my personal skill level. I'm kidding, obviously. Let me just quit out of this nonsense and start my proper presentation.
00:00:52.140
So when I gave this talk in the US, I apologized in advance for the swearing. I figured my adorable British accent would let me get away with it, but I am aware that a British accent might be received slightly differently in Europe given my nation's recent history. Maybe I can win the love of the European room by demonstrating a history of British cultural exports. I know one person here shares my love of the Spice Girls, so that's one out of 700. Maybe the Beatles and the inventors of the microphone are another British gift to the world. Firstly, sorry for all that brexity behavior and for the succession of idiots that we keep electing. If I did want the conference to happen in Brighton, I could also reassure you that your money will go a long way in the UK as well because our currency is worth basically nothing now. Secondly, the swears are staying in, even if the accent doesn't save me.
00:01:52.500
This talk is about history—not musical history, not ancient operating systems, but the story of our craft. And because who doesn't love it, a little audience participation! These are not object shapes; these are just circles. However, what you need to know is that they are different sizes. Now, I think there won't be any distortion; I believe it's all the same on all screens, but there is a fundamental difference.
00:02:05.219
So, let's see by show of hands who is a good judge of size. Who thinks the blue circle is bigger? Okay. And who thinks the red circle is bigger? Hmmm, interesting. What's the answer? Everyone who put their hand up is wrong. What was your instinct when the two circles appeared? They're the same, right? So, the real question is: how many white guys with a microphone did it take to tell you these two identical circles were different before you shrugged and said sure and tried to choose between them? Hold that thought.
00:02:29.760
As we bounce around the last hundred years or so, welcome to Cambridge, an ancient seat of learning. This is how you might imagine it: It's 2018. I’m sat in the audience within the brutally looking 60s-built Churchill College in a stifling lecture hall—a room full of people, but not anywhere near as nice as this. It's the business of software conference, lectures from the good and the great on how to build successful software businesses. What the hell am I doing there? By 2018, I'm the CTO of a small software company, having started my computing journey 30 years ago copying code from magazines into my Acorn Electron. It was a computer science degree that made me a relatively early adopter of Rails—I'm no longer new, as my receding hairline will attest. I had a little bit of a funny moment when the previous speaker was saying 2001 was a long time ago. I'm like, '2001? It’s the future! I've got decades of experience hand-crafting artisanal bugs for the internet!'
00:04:11.459
Anyway, back in the room at the business of software, a speaker comes on and starts to tell their story. I had never heard of them, and the story they told changed what I thought I knew about the history of our industry. Aged five, Steve Shirley arrived in Britain as a child refugee. As part of the Kinder transport, she attended a convent school followed by high school near the Welsh border. Steve excelled in mathematics, but given her situation, faced a battle to be allowed to take it.
00:04:23.880
You see, in the early 50s, there was no standard curriculum in the UK, and each school ran according to the whims of individual headmasters or mistresses. After leaving school, and after becoming a British citizen, Steve joined the Post Office research station in North London, part of the early Vanguard of the UK's post-war computing boom. Evenings were filled with a honors degree in mathematics, and another job followed after a decade at the Post Office, firing code into more early computers, including this little beauty: the ICT-1301. So far, so typical of a programming career at the time.
00:05:55.260
But when the story reached 1962, I really sat up in my chair. In 1962, Steve Shirley incorporated freelance programmers with a capital of six pounds. Now, as I’ve already said, that’s not worth anything now, but it was approximately the cost of a Venti gingerbread latte these days and not much more back then. Later, the company became F International. In their 2012 memoir, they said it sounded mad.
00:06:06.660
Drawbacks included the following: I had no capital to speak of. I had no experience running a company. I had no employees, no office, no customers, and no reason to believe that anyone wanted to buy my product. Nobody sold software in the 60s; as far as it existed, it was given away free with the machines it ran on. So what would happen to the freelance programmers? Would our hero be successful? Well, yes!
00:06:44.220
This was the Business Software Conference. Given it's a conference about how to be good at writing software and being in business—yes, it prospered for decades. Projects included planning routes for Titan Oil sugar lorries, calculating depot locations for oil companies, and even the black box for Concorde. They went international—the U.S., several European countries, and even grew through acquisition in India. In 1996, they floated on the stock exchange, only two years before this idiot started his degree.
00:07:03.060
This is a huge commercial UK software success story, and they pioneered a few remarkable things. The workforce was primarily remote in the 60s. They worked first with overnight post and then later telephones as those became more available. The workforce was flexible in hours worked and the strength of the independence of the programmers. Folks only had to work 20 hours a week, and the management and staffing of projects was really dynamic.
00:07:28.800
The profits were shared. In 1981, Steve Shirley established a shareholder's trust and progressively gave away a controlling share of the whole company to the workforce. So, when the IPO happened, over 70 employees became millionaires. One more thing: of the first 300 staff, 297 of them were women. Steve Shirley made a success by hiring an unorthodox workforce even as the rest of the industry struggled to find programmers as the business world rapidly computerized.
00:08:00.479
And there was a reason for that because this is who was on stage. She's an OBE, she's a Dame, and one of the highest honors you can receive in the UK. There are only ever 65 at one time. Current Companions of Honor include Ian McKellen, Paul McCartney, Elton John, Judy Dench, and multiple former Prime Ministers. This is the group that she is a part of, and I hadn't heard of her at all or her trailblazing software company. She exuded warmth, insight, and elegance from the stage.
00:08:19.860
In a TED talk in 2015, she mentioned that 'in those days, I couldn’t work on a stock exchange, drive a bus, or fly an airplane. I couldn’t open a bank account without my husband’s permission. For years, I was the first woman this or the only woman that.' My generation of women fought the battles for the right to work and for equal pay. Nobody really expected much from people at work or in society because all the expectations then were about home and family responsibilities.
00:08:51.080
And I couldn’t really face that, and so started to challenge the conventions of the time, even to the extent of changing my name from Stephanie to Steve in my business development letters so I could get through the door before anyone realized that she was a she.
00:09:43.560
Computing wasn’t just happening in little old Blighty in the 50s and 60s; it was also happening in the mighty United States of America. Let’s cross the Atlantic and find out what was happening. After the Second World War, the development of early computers occurred under conditions of wartime secrecy. Actually, wait; we need to go back a bit further and get some more context.
00:10:20.220
The term computer has been in use since the 17th century; it was a job for human beings. But it wasn’t until World War One that it became a profession. In both world wars, large numbers of human computers were required on both sides. These folks used the power of math to make map grids and artillery tables. Most of these new essential human computers were women.
00:10:50.520
Two years before this building was constructed, Grace Brewster Murray was born in New York City. A curious child, she once dismantled seven alarm clocks to see how they worked before her mother managed to stop her. I'm sure all the parents in the room understand that. Grace excelled at mathematics, achieving degrees from both Vassar and NYU, and married in 1930, taking her husband Vincent's surname.
00:11:02.760
Grace Hopper got a PhD in 1934 from Yale and returned to Vassar where she was on the tenure track. She took the classes that no one else wanted to teach, updating old coursework and trying entertaining new techniques, such as building imaginary worlds in a mechanical drawing class. This earned her growing student numbers in previously untrendy subjects along with the unbridled resentment of her male peers.
00:11:39.060
Then, when Grace was 36, this happened—well, the actual Pearl Harbor rather than the mid-2000s Michael Bay movie that’s not as good as The Rock. Grace attempted to enlist but was rejected because she was too old, too physically light, and too valuable as a mathematics professor. So, she joined the reserves anyway. She graduated at the top of her class, and to her surprise, she was sent to Harvard to become the world’s third computer programmer on a ten thousand-pound Computing machine named the Mark One.
00:12:11.700
Grace was invaluable in understanding both the machine’s capabilities and managing the volatile designer of the machine, Howard Aiken. Aiken gave Grace a code book, some strange-looking commands, and one week to write a program that could compute interpolation coefficients to 23 decimal places. Now, I don’t know what that is either. The problem itself wasn’t difficult for Grace with her PhD, but the machine itself was a mystery. It was the first computer; there was no precedent and, of course, no manual.
00:12:48.600
Grace worked incredibly hard, applying her powerful intellect and extraordinary work ethic. She wasn’t bothered by the less-than-friendly welcome of the male programmers who didn’t want to sit next to a woman. Aiken’s boat was a strict Navy operation, meaning rank and competence were uppermost. This meant Hopper rose through the ranks to become likely the most important figure in the organization despite her gender. This page from her logbook contains a sellotaped moth that attracted by the lights flew into the Mark 1 and was beaten to death by the relays. The caption reads, 'first actual case of a bug being found.'
00:13:33.600
Meanwhile, in Pennsylvania, the electronic numerical integrator and computer was created. This buzzing, electrically dangerous room of cables and vacuum tubes was faster than the Mark 1. However, there was a flaw with this new beast of a machine—whereas the Mark 1 took instructions on paper, the ENIAC needed to be reconfigured manually; cables and switches had to be plugged and flicked for every new job.
00:14:02.280
Whilst it could blaze through problems compared to the Mark 1, a second of processing could take a day to set up, and their cable management was awful. What Grace found when she arrived at the University of Pennsylvania was that she was by no means the only female programmer there. Many other women had been working in similar roles, and these women were key to this pioneering work. They latterly became known as the ENIAC six.
00:14:49.620
ENIAC itself was eight feet tall, 80 feet long, and weighed 30 tons. It had 40 panels arranged in a U-shape with 18,000 vacuum tubes each, and again, no manual. As Betty Jean described, 'ENIAC was a son of a.' To program it, they literally moved around inside the giant machine. The Bettys were the programming aces of the project.
00:15:12.600
When the ENIAC was to be unveiled to the scientific community, they were given a short 12-day deadline to do an unprecedented set of ballistics calculations. It seems tech keynotes have always been prepared at the last minute. Despite working around the clock and drinking copious amounts of apricot brandy, apparently by the night before, there were still serious errors in the program. The Bettys checked and rechecked against the test program that was hand-calculated by their colleagues. Dejected, they slumped home; the big demo was going to be an embarrassing disaster.
00:15:53.400
But like all good programmers, the answer came to Betty Holberton in her sleep. She made for the lab early, knowing precisely which of the three thousand switches to flick to solve the problem. The lights were dimmed, and the calculation for the trajectory rippled across the neon lights of the ENIAC in merely 20 seconds—which is faster than the projectile could have soared through the air. The demo had gone off spectacularly; the Bettys hustled to print souvenir sheets of results and handed them to the audience.
00:16:38.640
The event made headlines. The women remembered flashbulbs, but the photos in the papers showed only men posing with a thinking machine sold as a giant brain doing thousands of calculations per second. But as we all know, ENIAC was not thinking; it was just doing what it was told really, really fast. In reality, the only giant brains in the room were the women walking amongst the panels on the ENIAC programming it.
00:17:08.640
There are a couple of problems with the New York Times breathless reporting here. The work wouldn’t have been done by a man before the ENIAC and willfully disregards the weeks of labor by a group of bloody women to produce the demo. This complex and creative programming work was seen as sub-professional clerical work. As Jennings put it, 'if they’d known how crucial programming would be and how complex, they would have been more hesitant to give the job to women. On the 50th anniversary of the ENIAC, they rebuilt the rim-sized computer on an integrated chip that could fit in your hand.
00:17:54.300
At the same time, Catherine Kleiman contacted the organizers of the Women in Technology Conference to see how they were planning to mark the 50-year anniversary of the ENIAC six, who had inspired her during her Harvard undergrad. Yet, Whitty had no knowledge of the women she was referring to. The ENIAC six had been so thoroughly swept under the rug that even a group dedicated to furthering the status of women in tech had no knowledge of the work they had done five decades earlier.
00:18:24.300
These women invented the idea that you could write programs. The next time you write a line of code, run a complicated program, or find a bug, know that in part it's due to these six women who programmed the ENIAC 75 years ago. Despite the erasure of their work both at the time and for decades afterward, these women and women like them went on to have major impacts on the development of our profession.
00:19:16.680
Over the course of their long careers, Grace Hopper would go on to write the world’s first compiler, the A-0 in 1951. Betty Holberton and Grace would work together after the war when the military projects finished. When Betty was working on UNIVAC, a successor to the ENIAC, she ensured a numeric keypad was next to the keyboard and persuaded mechanical engineers to encase the black machine in a different color. So beige became the color of computing for 30 years.
00:19:50.700
She also wrote a program called the sort merge generator, which according to Grace Hopper was the first time a program had ever been used to write another program. The spectacular dexterity in such a concept that we now take for granted was, of course, initially resisted by the powers that be. Later iterations of Grace’s compilers became mathematic and flomatic until in 1959, she buttonholed a well-known computer scientist and explained why she thought it was about time for a common programming language for business. Meetings followed, and general agreement formed within the fledgling industry.
00:20:24.960
Using Grace’s Navy connections, the Department of Defense stepped in to sponsor the work. Now, multiple committees were formed inside CODASYL at different time horizons. Now, as we know, committees once formed become unbelievable, expanding messes where deadlines disappeared decades into the future. CODASYL was no different. Grace immediately realized the only group with a chance at succeeding was the short-range team.
00:21:15.660
Luckily, this was Betty Holberton, Mary Hawes, and Jean E. Samet. They referred to themselves as the PDQ committee because they needed it done 'pretty damn quick.' Three months later, COBOL was the result. Now, it’s pretty much no one’s favorite programming language by today’s standards, but it got done, adopted, refined, and most importantly, used because of the efforts of these three women. To give you an idea of the impact of COBOL, it’s estimated that in the year 2000, 80 percent of all production code in the world was written in it.
00:22:04.920
That was a long time ago. In fact, I even did some dangerous research. Anyone got any idea how many Ruby jobs there are currently on LinkedIn? There are 12,000 Ruby jobs on LinkedIn, but there are 5,000 COBOL jobs. So we win! Yay, Ruby community! At the jobs board on the world's weirdest social network, good, but that is still a pretty good showing for a 60-year-old language that nobody really likes.
00:22:52.920
We are by no means special in our software writing world. If you weren’t moderately well-off, heterosexual, or a white guy, it has sucked for most of history. Here are some other brief examples: Crick and Watson were the fathers of DNA, but their Nobel Prize-winning work on the double helix was based in part on the largely uncredited work of Rosalind Franklin. In 1967, Jocelyn Bell, as a research student, discovered the first pulsar.
00:23:24.720
She found it by direct observation, unrolling miles of recordings and crawling on her hands and knees to look for unusual deviations in a single wiggling line until she recognized the pattern. It was jokingly called LGM-1, her little green men, and she had to overcome her skeptical supervisor, Anthony Hewish, to prove her discovery of a neutron star. Guess who got the Nobel Prize for physics? Not Jocelyn.
00:24:02.159
Margaret Hamilton, the lead developer for the Apollo flight software, credited with having coined the term 'software engineering.' She said, 'I fought to bring the software legitimacy so those building it would be given their due respect. When I first started using this phrase, it was considered quite amusing; it was an ongoing joke for a long time.'