Interview with Patrick Collison | Hacker News Reader

Interview with Patrick Collison

noahpinion.substack.com

186 pointsjger1553 comments

Almost 20 years ago, I met Patrick at a Lisp meetup in a pub in London where he talked to us about Croma, the Lisp he developed for Web use. I learned that he was looking for a Lisp machine, and as I had a Symbolics 3630 which I hardly ever used, I arranged for someone to pick it up and deliver it to him. I wonder if it still works and if he is still using it.

Donald! It's great to hear from you. I still have your Lisp machine and in fact booted it last time I was at my parents' home in Ireland last year. Using one was quite formative for me... Lisp machines significantly elevated my bar for what developer experiences could be and, on some level, that whole line of thinking is what led to Stripe.

Thanks for sharing! What a neat moment to observe a friendly in-person interaction, reconnected via the internet years later.

To me it seems like this is an example of how "the explosive expansion in access to opportunity facilitated by the internet" isn't limited to business opportunity.

Twitter update from pc with a photo of the machine: https://twitter.com/patrickc/status/1369320446827528192

Hi Patrick! Glad you that found it interesting and that it still works. I rescued it from a former employer (BT). Initially I was offered a Symbolics 3660 which was a lot bigger and had a 30 amp plug with round pins, but I turned that one down.

Interesting. In what sense(s) is the developer experience in Lisp machines different?

By the way, Patrick, while you're around: any plans to GPL CROMA at some (remote) point in the future? Have always been curious to see it...

He's 32 now. So he was 12/13?

Probably in 2005.

https://www.irishtimes.com/news/limerick-pupil-s-web-languag...

> Swiss nationals have won more than ten times more science Nobels per capita than Italians have. Ten times! And yet they're neighbors, and Italy certainly isn't lacking in scientific tradition -- Fermi, Galileo, the oldest university in Europe, etc. The "how" of science just really matters.

One thing I worry about in the US is homogenization in the way science is done. You won't find all that much difference between how scientists operate in California versus how they operate in Massachusetts or Idaho or anywhere else. There's pretty much the same grant application process, same peer review process, same journals, same conferences. And that would be fine if we knew that we were doing things the Right Way -- but as it is, it just feels like putting all our eggs in one basket.

Can you expand a little? It's not clear what problem could come from having a standard process for funding, reviewing, and sharing science. I know there are problems, like the replication crisis, but I'm not tapped in enough to know if/how that kind of thing links to the way science is done.

Sure! As an illustrative example, let's look a bit at the standard process for funding.

One of the main parts of being a scientist, at least as measured by time spent, is writing grant applications. Scientists spend a huge percentage of their working hours asking for money from an elaborate, high-overhead funding system that goes to great lengths to try to avoid spending money on the wrong things. But in practice the correlation between attractiveness of grant proposals and the quality of the resulting research doesn't seem to be strong, meaning that this process might just be, at least to a first approximation, an ever-more-expensive random number generator.

You can certainly imagine other ways to do it. Funding agencies could try allocating money at coarser granularity, e.g. larger grants shared among more researchers with much less specificity about how the money is to be spent; I hear this used to be more common many decades ago. Or they could do a quick-and-dirty screening for what 1-2 page research proposals aren't obviously terrible and send out money to the vaguely plausible ones at random [1], which would at least have lower overhead. Or you could have grant money allocated by individual people instead of committees, with less deliberation and guaranteed fast turnaround time to a yes-or-no decision, e.g. [2] and [3].

In general there's usually more than one way to do something, and I worry about a monoculture getting stuck doing things in a deeply suboptimal way but unable to change. I could tell similar stories about the modern peer review process (a relatively recent thing, and arguably not very good at solving the problem it's meant to solve), or the persistence of bad statistical methods in a lot of fields because they're the standard that everyone expects, or various other things.

[1] https://www.vox.com/future-perfect/2019/1/18/18183939/scienc...

[2] https://marginalrevolution.com/marginalrevolution/2018/09/em...

[3] https://fastgrants.org/

Is the problem the grant process or the lack of cash to award? There are pre-proposals for say department of energy that are only a couple of pages so you can see if you should waste your time on an application. I am not sure about other agencies. You do have grants that go across groups of researchers (MURIs/MRSECS/etc) On the tangent that you mention, I sadly agree with you about the difficulty of dealing with bad statistical methods. I am curious about your thoughts on peer review.

Very interesting discussion/thread! I have been thinking lately, that scientists whom research area is around developing vaccination, might not need to worry anymore about getting grants or funded because of covid-19. It might be considred one of best consequence of the unfortunate epidmic.

If anything America has a way more varied system of universities and research centers, for virtue of being bigger,richer and with a culture of generous donations by private individuals. A system like that has good things and bad things, compared to say Germany or France where most universities and research centers are more or less equivalent in funding, prestige, salaries and lines of research.

Interesting comparison, but I offer a different interpretation.

First of all, unfair to compare CH (Switzerland) with IT (Italy), instead of IT to FR (France), DE (Germany), ES (Spain), UK (United Kingdom), countries of rather similar population size, economy, land mass, etc.

If you look at the list of countries by Nobel laureates per capita [0], you find the following:

CH: 4th; 27 laureates; 31.6 per 10M people.

UK: 10th; 133 laureates; 19.4 per 10M people.

DE: 14th; 109 laureates; 13.2 per 10M people.

FR: 17th; 70 laureates; 10.6 per 10M people.

IT: 33rd; 20 laureates; 3.3 per 10M people.

ES: 43rd; 8 laureates; 1.72 per 10M people.

In fact, CH is in the same ballpark as Sweden, Norway and Denmark. Smaller countries, richer countries, countries with a better education system overall.

In addition to that, of the 20 laureates in science in CH, 4 (20%) were born in other countries [1].

This is not (as it might certainly appear) to be pedantic, but rather to show that interpretation of data can give us a different story, based on what we are looking for.

[0]: https://en.wikipedia.org/wiki/List_of_countries_by_Nobel_lau...

[1]: https://en.wikipedia.org/wiki/List_of_Nobel_laureates_by_cou...

Would you expand on this? I’m missing the “unfair” part about comparing IT to CH. Even compared to FR, DE and UK, IT (and ES) are significantly under-represented by the figures you present.

It is not an order of magnitude. It is a 2x / 3x difference, and actually Spain performs worse than Italy.

I had the chance to briefly meet Patrick in Estonia some years ago when he came to our offices at e-Residency. He immediately struck me as an incredibly intelligent person that had a crazy mix of intellect, curiosity, and eq. This interview definitely solidified for me that he is probably one of the best, and underrated, entrepreneurs out there right now.

Underrated by the public but Paul Graham holds the Collison brothers in incredibly high esteem, lots of other SV people do also.

Having worked with them in the past (when Patrick was a newly "on paper" millionaire and not a widely known billionaire visionary), I feel the same way.

They're smart, sure, and have great ideas and execution, sure, but they're also just really, really great people, and fun guys to hang out with.

The one thing that stands out for me - and this goes for both brothers - is their modesty. There are a lot of people thumping their chests all day long about very minor achievements and these guys just plug away at what underpins a very large fraction of all e-commerce. Stripe one day will be really too big to fail.

My partner cooked for the very-early Stripe crew, and has shared some of her fond memories from that time with them.

pc - No telling when we'll get back up to CA, but there's buffalo chicken mac 'n cheese on offer :).

Patrick is a great many things but I wouldn’t include underrated on that list. I don’t think anybody operating with or around him believes he’s anything but one of the most competent executives on the planet today. If you surveyed 100 venture capitalists and asked them the likelihood of Stripe hitting a $1T market cap, I think 99 would agree and one would be trying desperately hard to appear contrarian.

Yeah, for sure not underrated in the valley and by people who know him but I meant generally. I doubt the average person knows much about him or Stripe

Stripe’s not a publicly traded company yet — it will be at some point — and I get the sense that Patrick and John are enjoying the relative obscurity while they still can.

"relative" I think John was on some sort of an ad on all the transit signs in SF, maybe during Dreamforce a few years back.

The general public only cares about consumer goods. Why would they know anything about a payments processing CEO? The "star" CEOs are all from consumer goods - Steve Jobs, Bezos and Elon Musk are probably the only ones any American can name.

This is interesting and I'm trying to think of others. You're on to something. Zuckerberg comes to mind. Jack Dorsey, maybe. Warren Buffett - consumer business.

I think the Shark Tank cast are probably pretty well-known by now, at least as much as Bezos. Though I don't know if they're known for their companies, so maybe it doesn't make sense to include them. In fact, Daymond John is the only whose business _I_ know, and that one was a consumer business. They're all much smaller time, ofc, than the other names mentioned. Cuban makes me think a bunch of sports CEOs are probably well known. Vince McMahon.

Jack Welch, Lee Iacocca, and Ray Kroc I think were big deals, back when.

Probably there's no value in the CEO of a non-consumer company having celebrity. Nobody's deciding between PWC and E&Y based on a TV appearance by the CEO, so the CEO won't spend time on that.

I heard similar opinion recently about Square hitting $1T market cap too. Let's hope that market is big enough for both companies.

I think picking on the FAA is a little unfair. For the past ~20 years, they've been enormously successful in overseeing commercial airliner operations - I think the last commercial crash in the US that killed all onboard was in 2009

I think that's entirely the problem. I'm worried it's a sign that the FAA is no longer as effective as it once was, and the problems are going to quickly grow as time goes on.

This is a guess, of course. Time will tell.

This is what institutional decay often looks like! They can still carry out a set of procedures allowing them to do the same thing they've been doing for a long time -- but they get less efficient, less flexible, more ossified. They are no longer the FAA that figured out how to oversee airline operations; they're the FAA that follows the script left to them by their predecessors. Collison gives a nice example:

> The avionics you see in cockpits are bafflingly primitive because it's so hard, slow, and expensive to get the FAA to approve new technology. As a result, pretty much every pilot flies with an iPad running sophisticated flight planning software -- their connection to a world that the FAA doesn't encumber.

There was a time when the FAA was much better at dealing with new technology. There was a time when all the technology in their purview was new. Something changed.

iPads can run out of battery or get smashed in turbulence or fall prey to bugs/malware; old-fashioned steam gauges cannot. The FAA's regulations are written in blood - the aviation fatality rate has fallen sharply over the years [0] thanks to a combination of technological advancement and regulation. Aircraft builders are still free to do pretty much whatever they want within the 'experimental' category, provided they mount a placard informing occupants that the aircraft is experimental.

[0] https://aviation-safety.net/graphics/infographics/Fatal-Acci...

Just a few days, at the top of HN, was this post[0]: "FAA safety engineer goes public to slam agency’s oversight of Boeing’s 737 Max".

It links to a Seattle Times article [1] that outlines (from the inside) systemic failures in how the FAA attempts to implement it's own mandate.

That article would lead one to believe that it is in fact difficult to be _overly_ critical of the FAA.

It has a stranglehold on innovation in the industry, and... well, just because you grew up with it (presumably) doesn't mean it's not the laughingstock of international aviation-oriented industries.

The FAA is a trashfire. Persons who grew up in the USA don't have a benchmark on what it would look like for an agency that did it's job well.

[0]: https://news.ycombinator.com/item?id=26376549

[1]: https://www.seattletimes.com/business/boeing-aerospace/faa-s...

> I think the last commercial crash in the US that killed all onboard was in 2009

The FAA approved the MAX and let it keep flying after the first crash because it wasn't Americans who died. They only acted after more died in a second crash and they could no longer pretend they weren't at fault for those deaths.

In the early days, when I was implementing Stripe and we were about to launch, he gave me his phone number in case anything went wrong or I had an emergency. And he answered (I think it was the middle of the night). :)

> France today gets three quarters of its electricity from nuclear power.

Yup, and one of the surest way to loose an election and / or media support in France is to ever mention the fact that you might even consider this to be a not-entirely-terrible thing.

Anyway, this person seems really interesting.

> ever mention the fact that you might even consider this to be a not-entirely-terrible thing

If I'm parsing this correctly, you mean that nuclear power is unpopular in France?

It's... complicated, but in a nutshell, yes, it's unpopular . The closest thing we have to data would be polls [1], which shows that the view is now positive only for a minority (although the ratio is 47 / 53, so you might say it's "slighly" unpopular.)

It's clearly became a purely political and partisan topic though ; the right-wing, older and upper class will be more favorable than the left-wing and youngster.

The French Green party, the most "visible" NGOs, left-wing press and public media are (rather) openly against nuclear.

Right-wing press and the industry is only slighly more favorable. Quite surprisingly, even the far-right movement is going "full nuclear" those days.

A large part of the population is concerned by security and the handling of nuclear waste ; and too many believe that our nuclear plants are contributing to global warming [2].

The debate is not going to get better any time soon, as we're about to reach both the end-of-life of the 40-year-old plants, and... an electoral year.

[1] http://www.odoxa.fr/sondage/nucleaire-lopinion-sest-retourne...

[2] https://www.lepoint.fr/economie/rechauffement-les-francais-a...

> Quite surprisingly, even the far-right movement is going "full nuclear" those days.

Just to clarify, it's "surprising" because they can be quite sceptical about science (the leader of the "main" far-right party has made comments about covid vaccines that borders on conspiratism ; the "smaller" far right parties are crossed the border a long time ago. )

They're trying something that is being more "anti-wind" than "pro-nuclear", which makes sense as they want to gather votes from the southern countryside with is heavily biased against wind turbines.

We'll see how that fares next year when they have won the election.

A bit of a shame that this is not available in audio format, it's quite a long read.

I never used Stripe the financial system but I like that Stripe has a book division, the books they have published so far are right up my alley so keep up the good work.

There's a chrome extension called audiblogs [0] that can convert any article into audio format. Here's the audio version of this article: https://audiblogs.com/share/4ce7e9c6-0970-4fa6-8cdb-ac244104...

[0] https://audiblogs.com/

Yes and screen readers are also a thing for people that need audio for accessibility reasons. For anyone else I don't see any point to this... Listening to audio is slower than reading, but it can be more interesting & engaging to hear people's voices and get a glimpse of their personalities in an audio interview format. An automated text to speech translation is the worst of both.

This seems unnecessarily dismissive of audio content. Some people prefer audio content to reading. And with audio you can do passive consumption while doing something else.

> An automated text to speech translation is the worst of both.

I think you're underestimating how much better we're getting at making automated voices sound human. This site is way better than the average text to speech, and technology is only going to get better at it.

I guess I've drank the koolaide whole, because not landing a position at Stripe is one of my larger regrets in the tech world. Oh how they've grown in the last few years! Curious how the culture is now.

> Nobody in financial services thinks that real-time settlement is a bad idea

Maybe this is referring to something else, but e.g. Matt Levine thinks it's a bad idea: https://www.bloomberg.com/news/newsletters/2021-02-26/money-... discusses how real time settlement on spot prices for electricity would be worse than the alternative.

https://www.bloomberg.com/news/newsletters/2021-02-25/money-... discusses how real time settlement would require you to have all money on hand immediately, making trading more difficult and expensive. Also amusingly:

> It is not unimaginable that the stock market could move to real-time settlement; you could put stocks on the blockchain hahaha.

Loved the section where he talked about technology the world needs.

I took each idea down and started taking notes on it.

https://joshdance.medium.com/request-for-technology-from-pat...

It's curious that Patrick lists advances in medial technology as a higher need than Climate Change mitigation tech. Does he elaborate on this idea anywhere? It is because we don't actually need better tech to fight Climate Change and it's mostly a socio-political issue?

My favorite part is when he calls "GEB" a slog. I think it's one of the most overrated books of the last 50 years.

Reply on news.ycombinator.com