Breaking the internet, and the scientific paradox on The Future, This Week

Disclaimer We'd like to advise that the following program may contain real news, occasional philosophy and ideas that may offend some listeners.

Kai So, Sandra, what are we talking about?

Sandra I know you've got one story you're dying to bring up.

Kai Krispy Kreme reveals its plans for an initial public offering trading as DNUT.

Sandra That's D N U T?

Kai Yes, yes, it's in the New York Times. I wanted to bring up this story for a couple of reasons. One is Krispy Kreme, a doughnut company, isn't exactly the company that comes to mind when we think IPO these days.

Sandra Mostly because it's been around for eighty some years.

Kai Yes, and it is not a tech company. But Krispy Kreme has had a great pandemic year, which is the interesting aspect here.

Sandra Krispy Kreme doesn't really seem to be on trend, right? Most people during the pandemic have learned how to bake sourdough bread and have been exercising and doing yoga and, you know, improving on their health habits, but the numbers seem to tell a slightly different story.

Kai Yeah, so, Krispy Kreme sales grew 17% to $1.1 billion in the last fiscal year, which kind of paints a different story to what we're seeing on Instagram, or what is portrayed on endless Zoom calls and on Twitter, where people seem to make everyone believe that they're living this healthy lifestyle, and they're doing all this sport. So maybe what happens off camera is doughnuts.

Sandra And that to me is not all, because often companies that are not quite on trend are ignored and we think they're not doing well, how could they not company be doing so well at the time when all the trends point towards health and maybe something that's got almost half your daily intake of sugar is not something that would perform really well during the pandemic. But a similar thing happens in tech actually. Last week, Hangouts reached 5 billion downloads.

Kai Just a reminder, Hangouts is a Google product that has supposedly been dying for the past five or so years.

Sandra And yet still, 5 billion downloads shows that it's very much not dead, it's supposed to be phased out sometime later this year but seems the jury is still out. And it reminds me of those conversations we had with Simon Kemp about all these websites and tech companies that get written off because they're not the sexy thing, but that are very much still dominating by sheer numbers.

Kai Anyone remember, Yahoo?

Sandra Yahoo still has a larger active user base than Snapchat.

Kai It's the fifth most visited site on the internet still, and it is very much a go to place for things like financial information, for example.

Sandra And often that's what happens to the not sexy thing. And I remember back in the day, we did this episode about the innovation that was happening and how well eBay was doing and no one was talking about eBay because it wasn't the sexy company.

Kai And the New York Times Shira Ovide's "On Tech" made a point the other day about those companies in the middle, we tend to think about the spectacular failures Theranos, or WeWork, the scandals; we think about the unicorns, like the Googles, Apples, Amazons, TikToks, Ubers. But we forget that there are many companies in the middle that are actually the breadth of what is happening in tech, solid companies like eBay, or Dropbox, for example, which is a solid performer, is used by a lot of people, but they're not the kind of stars that make headlines all the time.

Sandra So we should pay attention to things that are not obvious. But in the meantime, things that are very obvious, which we have to bring up. This has been happening for the last few weeks, a lot of the big energy resources, companies are doing interesting moves to cut emissions or to address climate change or to make really interesting left field investments. Chevron shareholders had voted for a proposal to reduce emissions, and similarly, Exxon Mobil also suffered what's been called a shareholder rebellion from climate activists and other institutional investors who are demanding that they set some strategy for a low carbon future. And they weren't the only ones in the news…

Kai Such as a Dutch court ordering Shell to also cut emissions.

Sandra And in an interesting other move PTT in Thailand, the Thai oil giant, has started investing in plant-based proteins in an attempt to become less dependent on fossil fuel businesses. To be fair, it is a $9.6 million partnership and market capitalisation of $38 billion.

Kai So a oil drop in the ocean…

Sandra But still, it's interesting to see a move towards, and I think it's a protein from jackfruit that they're using to create meat, which I think I've actually tried, it's not bad.

Kai I remember that and I think we've mentioned that on the podcast previously.

Sandra And something we'll come back to, I'm sure in a full episode, but there's interesting conversations about where all these decisions will lead. And there was a very interesting article in the South China Morning Post where Joseph Stiglitz the economist, Nobel Prize winner, interviewee on one of the Sydney Business Insights podcasts was warning that carbon pricing may inadvertently trigger a global financial crisis, that carbon prices will be much higher than current prices, and then fossil fuel assets will come down and that the transition will actually lead to something similar to the subprime mortgage crisis. And his reasoning was that carbon accounts for a much bigger piece of the global economy than subprime mortgages did.

Kai And what he endorses is actually quite a radical position, which is straightforward regulation, or basically banning fossil fuels entirely over time, rather than trying to price it out of the market.

Sandra He's saying that would be a much easier solution than actually having a pricing system because the transition that that pricing system will bring about will create a financial crisis.

Kai And of course, this sounds like a radical solution, because we're so accustomed to working with economics or incentives to solving these problems. But he points out that the risks are far greater and a much cleaner solution would be an outright ban. I found a story in Wired Magazine, it's titled changing minds about why doctors change their minds. And it goes right to the whole discussion around COVID health advice and vaccines, and basically, tries to discuss why public discourse, the general public seems to misunderstand what it means when doctors or scientists more generally changed their minds.

Sandra And really goes to the heart of what science is and how we do research and how we make progress. So I think it's a really good story. And it's a very timely story given all the news, we hear about vaccines, about treatments, about health guidelines.

Kai And it talks about the scientific paradox and we really should look into that.

Sandra So we're definitely doing that one but we will need to start with someone broke the internet.

Kai Again, happens from time to time. Amazon was down, so many other websites.

Sandra This is the story about parts of the internet going down.

Kai So a range of websites that were no longer available: news websites, shopping websites, government websites, there's a list.

Sandra Thousands of government news and social media, including the New York Times the Guardian, Reddit, large parts of Amazon, CNN, PayPal, Spotify, Al Jazeera.

Kai Etsy.

Sandra Oh, no, surely not Etsy.

Kai Yes, and Pinterest.

Sandra So it's not only you couldn't buy things on Amazon, or you couldn't read the New York Times. But significant infrastructure, things like PayPal, or the UK government websites were down for almost an hour.

Kai So let's have a look of what went down when the internet went down.

Sandra Let's do this.

Kai Let's do this.

Intro From The University of Sydney Business School, this is Sydney Business Insights, an initiative that explores the future of business. And you're listening to The Future, This Week where Sandra Peter and Kai Riemer sit down every week to rethink and unlearn trends in technology and business. They discuss the news of the week, question the obvious, and explore the weird and the wonderful.

Kai So the story we've picked is from the Guardian.

Sandra But we really could have picked any story as everyone reported, or at least tried to report, because as we said, the New York Times was down and the Guardian was down and reddit was down.

Kai Are you really posting an article when no one gets to read it?

Sandra Interesting you should say that because while vast chunks of the internet were down the Verge's website was also down, so they shared a Google Doc, we'll put the link in the shownotes, where they were doing live reporting in Google Docs.

Kai Audience numbers are unknown but it is good to see that the internet comes together in a moment of crisis.

Sandra The explanation for why this specific outage happened was fairly straightforward.

Kai So should we do a deep dive into how the internet works? No, okay. So what happened is a major outage at a company called Fastly, that most people have never heard of. Fastly is a so called content delivery network provider, it basically sits in the middle of, you know, the companies that host your websites, these cloud service providers, and the ISPs that deliver your content. And what Fastly does is it basically hosts a lot of big files around the world for all these companies that we've just mentioned, so that people can get their videos in their content faster, but if this piece of it Infrastructure fails, those sites are no longer available.

Sandra No. So if we're accessing the New York Times, and let's assume the servers for the New York Times are in New York, and we're accessing the New York Times in Sydney, there would be one of these companies like Fastly, that would make the content available somewhere closer to us, so that pages do not take that long to load. And this is important, of course, for news content, and pretty much for any website, because we now assume any website will load really, really quickly. But think about video streaming, think about the most popular social media apps that we use, we want that content quickly and in order to do that you have companies like Fastly.

Kai In essence, that are layers of infrastructure that provide us with the internet, in this instance, a big piece of one of those layers failed. Which however, raises questions as to the vulnerability of the internet as an infrastructure.

Sandra And what the outage at Fastly has highlighted is one of the problems with the, with the CDNs, with the content delivery networks, which is that there's really only three big players in this space, and Fastly is one of them.

Kai And the other two are Akamai and Cloudflare. And Cloudflare, incidentally, had a major outage in 2020, when one of the engineers accidentally pressed the wrong button, and they pushed huge amounts of internet traffic to just one data centre in Atlanta. And that thing basically failed, which again, left lots of websites offline, including the video game League of Legends.

Sandra Well, at least Etsy wasn't affected.

Kai We don't know that.

Sandra So the issue here is about concentration in what has now clearly become infrastructure. And concentration in this case is not actually a bad thing per se, much of this concentration was done with the aim of making these companies extremely resilient. Size was allowing these companies to deliver the type of service that many of their clients wanted, a company like the New York Times, or Netflix or Amazon or PayPal, require service providers that are very large, and that can offer the level of service that they need.

Kai A service that needs to be fast, it needs to be efficient, that also needs to be secure. These companies invest large amounts of money into cybersecurity, of course, but it also increases the risk of failure when it is just a few big entities that can bring down large parts of the internet.

Sandra So while having companies of this size and having that concentration thus provide resilience, it at the same time takes resilience out of the system, because there is no excess capacity in that system.

Kai Which brings us to different notions of resilience. On the one hand, size of each one of those players allows them to be more resilient, that gives them the resources to build these resilient entities. But on the other hand, it compromises resilience at the network level. And I think it's important to understand that the very idea of the internet going all the way to the 1960s, what was called the ARPANET, which was basically funded by the US government, universities, and also the military, was to create infrastructure that is distributed and resilient to the extent that you could take out individual computers or even entire sub parts of the internet, and the rest would still function that was always the idea of the internet. However, the developments over the past few years, be it bringing in place these content delivery networks for speed, or otherwise, concentration means that this network resilience is now increasingly compromised.

Sandra And this is usually what happens when you have a scientific or engineering idea that comes up against business models and day-to-day economics.

Kai Because there is of course, a trade off between making infrastructure more efficient and making it more resilient, which means building in redundancies that are only needed should things be failing.

Sandra But this story also highlights one other point, which is, amongst all the things that went down, it was also all the websites of the UK government.

Kai And among them a website that is used to register for COVID vaccines.

Sandra Which highlights that they have become a utility. And it's interesting because Cloudflare one of the other content delivery networks actually remarked that it's hard to imagine another utility like electricity or water or gas, coping with a sudden and continuous increase in demand of 50%, which is what happened during the pandemic. And there is an interesting Brookings Institute analysis of that, which we'll put in the shownotes, but it highlights the fact that companies like Cloudflare have become like public utility services.

Kai And not only are they nodes that make the internet vulnerable to cyber attacks, for example, they also wield a fair bit of power, which we've highlighted on a previous episode.

Sandra And we did discuss this back in 2019.

Kai That was the story when the CEO of Cloudflare decided to ban a bunch of Nazis from the internet, right?

Sandra Yeah, one fine morning in August, the CEO of Cloudflare, Matthew Prince, sent an internal company email in which he just decided to kick people off the internet. And the quote from the email was "my rationale for making this decision was simple. The people behind the Daily Stormer are assholes And I've had enough. Let me be clear, this was an arbitrary decision," he continued, "I woke up this morning in a bad mood and decided to kick them off the internet. It was a decision I could make because I'm the CEO of a major internet infrastructure company."

Kai So this case, as well as the outage points to the fact that critical public infrastructure, the internet, really depends on a few commercial entities. And in fact, a few individuals really, who not only run this critical infrastructure, making it vulnerable, but also make decisions about, you know, who gets to be on the internet in the first place.

Sandra And to me in the end the story really came down to how invisible a large part of the internet is to us, and how we always think about concentration of power or big tech when we think about the likes of Amazon or Google, but now we have this entire invisible infrastructure to the internet that is similarly critical but we don't really know or spend any time talking about.

Kai You're absolutely right. All this discussion about you know, free speech on Facebook and content moderation, what we tend to forget in the quest for regulation is that there's this whole hidden layer that needs taken care of, in order to protect the internet as a piece of critical infrastructure. Because as with every industry, this one is also affected by concentration and economic rationale, it's it's just that we don't get to see it.

Sandra But on to our second story, changing minds about why doctors change their minds. It's a story from Wired that tries to change our minds about how we perceive uncertainty and doctors changing medical guidelines or medical advice.

Kai So the article makes the point that when doctors, when scientists, change their mind, because new evidence becomes available, and findings change, that this can erode trust in the general public, because people say, "you know, yeah they've been wrong, who knows that they're right now, they've changed their mind we don't know what to trust and who to trust," when in fact, this is actually a feature of the scientific process that doctors and scientists can change their mind in the face of new and better evidence.

Sandra And this has really come to public attention during the pandemic as medical advice has changed as we better understood the disease, for instance, advice around wearing face masks, or how it was contagious, but also around the efficacy of vaccines, when doses is needed to be taken, and how many of them are what the side effects were.

Kai Or who should take which vaccine, what age groups, as we learn more, as the results of more studies become available as more results from the actual rollout become available, we can fine tune and adjust the health advice based on the medical evidence.

Sandra But for the general public, this meant "oh doctors must not, must not have the right answer, surely they've been wrong before they can be wrong again, how do we know that they are right this time around?"

Kai So we really want to take a look at the scientific process and how it differs from how we typically tend to think about the world. So we want to set aside any of the emotional discourse that happens on social media around this, or nefarious actors posting disinformation, which also contributes to a lack of trust in science.

Sandra And we also want to set aside the fact that there's been a gradual erosion in trust in experts across the world. And they've been number of books over the last decade about this even Al Gore's "Assault on Reason" spoke to this. Tom Nichols' "The Death of Expertise," they campaign against established knowledge and why it matters, spoke to this erosion in trust in experts. And really numbers over the last 50 years have shown that it's gone down from about 73% in the US who have confidence in for instance, medical experts, to about 34%, just 10 years ago.

Kai And so while there are a number of different factors at play, we want to concentrate on the scientific process and in particular, the notion of what it means to be wrong or what it means to change your mind.

Sandra And we want to do this for two reasons. One is that we keep seeing this story come back again and again. I remember us doing an episode on Corona Business Insights where people were really struggling with pre-print publications and how these were distributed, and scientific studies that were still undergoing peer review, but they were being reported by the media. And the second reason is one that the article itself brings up is that professors quite often half jokingly tell students that what they learned in class will be outdated by the time they graduate. And that is true for the medical profession as it is for business schools. And that there is a good reason for this, and that's the reason we want to talk about it.

Kai So we have this common idea that science is here to tell us the truth, which in essence, is what science strives for, it tries to get to the truth, but it does so sometimes in a zigzag kind of way, having to incorporate the evidence from different empirical studies.

Sandra So in medical research, when recommendations shift, they shift so after many, many studies, multiple studies bring incremental but increasing evidence that suggests that the guidelines should change. And the COVID-19 pandemic has accelerated a lot of these processes, but has also made them a lot more visible and a lot more unsettling for people who are not part of the medical profession.

Kai And so when scientists change their mind, when they admit that they were wrong, that doesn't point to a problem with science, it's actually a feature of the scientific process.

Sandra So it's a feature not a bug?

Kai It is a feature, not a bug. Changing your mind is an important quality in a scientist, when the evidence changes, or more evidence becomes available, I should say, and being wrong in the sciences can be a sign of progress, because you're learning something new, you're making a leap in understanding and knowledge.

Sandra Oh, and this reminds me of the conversation I had with Danny Kahneman.

Kai Danny Kahneman, the Nobel Prize winner in economics, who you interviewed for one of the episodes of Sydney Business Insights.

Sandra And I actually think we should play a bit of our conversation, this is not something that ever made it on the podcast.

Kai It is such an important aspect of how science works.

Danny Kahneman Failures are routine, in academic work, and in thinking work, but I actually enjoy changing my mind. And you know, the occasion for changing your mind is always when you find that you've been wrong. And for me, this is a real joy, finding that I've been wrong, because that discovery means that I've learned something. So failure and success are inextricably linked, in my experience of thinking and discovering that you've been wrong, of correcting yourself. And it's those failures, those challenges, you know, that's like, the kind of work that I do exciting, I changed my mind about wellbeing in a very radical way. I thought I had a point of view, and then I decided that point of view does not work.

Sandra So I think an important insight at this point is to highlight what being wrong means in the sciences and how it's very different to what being wrong in general means. Because in general, if if one of us was wrong, we made a mistake.

Kai But in the sciences, that doesn't mean that you made a mistake, it's just that you didn't have enough evidence yet, or you didn't understand the problem well enough yet. And so changing your mind means you're making progress towards the truth. It is not a mea culpa, a walking back of a wrong decision, it is rather the scientific process unfolding in areas that are not yet well understood, like a new virus coming in, the development of new vaccines, these are uncharted territories. Doesn't mean that we made a mistake, it's just that we haven't had the right evidence to understand it in the best possible way. So this should actually engender trust it's the scientific process working properly.

Sandra And that's the case with a lot of terms that have a slightly different meaning when people use it for research purposes, or when we use it in a day-to-day conversation. So when we say "oh but that's just the theory."

Kai Yeah, you're right. In general parlance, you know, having a theory is just a statement of belief.

Sandra Or "that's my theory about."

Kai Anyone can have a hunch or a theory.

Sandra But in physics, gravity is a theory.

Kai Exactly. And so in the sciences, the word theory doesn't just mean a statement of belief, it's basically the truth unless proven otherwise. Gravity is what the world is like, it's our understanding of this phenomenon in the natural world. Unless, you know, we've developed further evidence that might debunk gravity, that is the truth as we understand it. So scientists are often by necessity, tentative in their arguments, because they allow for the fact that future evidence might actually change their minds.

Sandra But it doesn't mean theories are just beliefs, they're grounded in evidence, and the process by which we reach these theories, it's a cumulative, usually quite slow process that has because of the pandemic, we made very visible, and has become part of conversations that do not deal well with, "well, it's complicated," or "we need more research." But they either need a one liner that is a good headline, or that makes for a good tweet.

Kai So science often comes up against, on the one hand, unrealistic expectations by the public or the media, that, you know, there should be an unequivocal truth, just tell us what the world is like, as you said, it's a slow cumulative process. And on the other hand, a misunderstanding of what it means when scientists are tentative in their arguments that they're 95% sure, or that they're never 100% sure about a given phenomenon.

Sandra And maybe one last thing to add here, which is very much part of the research process and of the scientific process. As we said, this is cumulative but it's also a conversation, a debate that people have. So often in research or in scientific discourse, there is a lot of criticism, of constructive criticism, where people on purpose try to take down other people's ideas, to see how robust they are, to stress test them, to build on their ideas. So constructive criticism, and building and changing ideas is very much part of the process. But that can lead to a lot of confusion if it's taken outside of the research process and into a more day-to-day and often politicised debate.

Kai So really what we need yet again and over and over again, is a better understanding of what we mean by terms like theory or being wrong, changing your mind, having debate, disagreeing with each other, or not being 100% sure, that these are actually necessary ingredients in, you know, wrestling with nature and revealing the truth about how certain things operate.

Sandra And that is actually what we call the scientific paradox, which is the things that are meant to instil trustworthiness in the process, the constructive criticism, the debate, the changing the mind, can often lead to people trusting the scientific process less.

Kai And so the two of us are happy when we can change our minds because it's usually a good publication opportunity because we understand something that we didn't.

Sandra And we should get back to our research.

Kai That's all we have time for.

Sandra Thanks for listening.

Kai Thanks for listening.

Intro This was The Future, This Week, an initiative of The University of Sydney Business School. Sandra Peter is the Director of Sydney Business Insights and Kai Riemer is Professor of Information Technology and Organisation. Connect with us on LinkedIn, Twitter and Flipboard, and subscribe, like or leave us a rating wherever you get your podcasts. If you have any weird and wonderful topics for us to discuss, send them to sbi@sydney.edu.au.

Kai So I have another story.

Sandra Not the chair one, right?

Kai It's such a pandemic story, it's–

Sandra It is the chair one isn't it?

Kai It's the chair simulator, it's a game about sitting.

Sandra Megan, it is the chair simulator.

Kai It's a great game, apparently. So you pick a character, and then you have to buy a chair, and then you have to sit in the chair, and sometimes you have to get up to you know, when the discomfort level gets too high. But you basically earning sit points, and with the sit points, you can go to the store and buy a better chair so you can earn sit points faster. And you just sit in the chairs.

Sandra This is the future of chair, not the future of business.

Kai But it's a game, it's apparently very popular. It's so pandemic, right, you're stuck at home and rather than just sitting in a chair, you can now sit in a chair and–

Sandra I feel like I'm playing it right now.

Kai You can play a game where you simulate sitting in a chair while also sitting in a chair. It's very meta.

Sandra I think I'm winning.

Kai We're putting it in the shownotes. Am I earning sit points? We could mine sitcoins. It's a whole new product.

Sandra Cut.

Kai Cut.

Sandra Cut.

Kai Oh, come on this is good.