" /> net.wars: November 2022 Archives

« October 2022 | Main | December 2022 »

November 25, 2022

Assume a spherical cow

SphericalCow-IngridKallick-370.jpgThe early months of 2020 were a time of radical uncertainty - that is, decisions had to be made that affected the lives of whole populations where little guidance was available. As Leonard Smith and David Tuckett explained at their 2018 conference on the subject (and a recent Royal Society scientific meeting) decisions under radical uncertainty are often one-offs whose lessons can't inform the future. Tuckett's and Smith's goal was to understand the decision-making process itself in the hope that this part of the equation at least could be reused and improved.

Inevitably, the discussion landed on mathematical models, which attempt to provide tools to answer the question, "What if?" This question is the bedrock of science fiction, but science fiction writers' helpfulness has limits: they don't have to face bereaved people if they get it wrong; they can change reality to serve their sense of fictional truth; and they optimize for the best stories, rather than the best outcomes. Beware.

In the case of covid, humanity had experience in combating pandemics, but not covid, which turned out to be unlike the first known virus family people grabbed for: flu. Imperial College epidemiologist Neil Ferguson became a national figure when it became known that his 2006 influenza model suggesting that inaction could lead to 500,000 deaths had influenced the UK government's delayed decision to impose a national lockdown. Ferguson remains controversial; Scotland's The Ferrett offers a fact check that suggests that many critics failed to understand the difference between projection and prediction and the importance of the caveat "if nothing is done". Models offer possible futures, but not immutable ones.

As Erica Thompson writes in her new book, Escape From Model Land: How Mathematical Models Can Lead Us Astray and What We Can Do About It, models also have limits that we ignore at our peril. Chief among them is the fact that the model is always an abstracted version of reality. If it weren't, our computers couldn't calculate them any more than they can calculate all the real world's variables. Thompson therefore asks: how can we use models effectively in decision making without becoming trapped inside the models' internal worlds, where their simplified assumptions are always true? More important, how can we use models to improve our decision making with respect to the many problems we face that are filled with uncertainties?

The science of covid - or of climate change - is only a small part of the factors a government must weigh in deciding how to respond; what science tells us must be balanced against the economic and social impacts of different approaches. In June 2020, Ferguson estimated that locking down a week earlier would have saved 20,000 lives. At the time, many people had already begun withdrawing from public life. And yet one reason the government delayed was the belief that the population would quickly give in to lockdown fatigue and resist restrictions, rendering an important tool unusable later, when it might be needed even more. This assumption turned out to be largely wrong, as was the assumption in Ferguson's 2006 model that 50% of the population would refuse to comply with voluntary quarantine. Thompson calls this misunderstanding of public reaction a "gigantic failure of the model".

What else is missing? she asks. Ferguson had to resign when he himself was caught breaking the lockdown rules. Would his misplaced belief that the population wouldn't comply have been corrected by a more diverse team?

Thompson began her career with a PhD in physics that led her to examine many models of North Atlantic storms. The work taught her more about the inferences we make from models than about storms, and it opened for her the question of how to use the information models provide without falling into the trap of failing to recognize the difference between the real world and Model Land - that is, the assumption-enclosed internal world of the models.

From that beginning, Thompson works through different aspects of how models work and where their flaws can be found. Like Cathy O'Neil's Weapons of Math Destruction, which illuminated the abuse of automated scoring systems, this is a clearly-written and well thought-out book that makes a complex mathematical subject and accessible to a general audience. Thompson's final chapter, which offers approaches to evaluating models and lists of questions to ask modelers, should be read by everyone in government.

Thompson's focus on the dangers of failing to appreciate the important factors models omit leads her to skepticism about today's "AI", which of course is trained on such models: "It seems to me that rather than AI developing towards the level of human intelligence, we are instead in danger of human intelligence descending to the level of AI by concreting inflexible decision criteria into institutional structures, leaving no room for the human strengths of empathy, compassion, a sense of fairness and so on." Later, she adds, "AI is fragile: it can work wonderfully in Model Land but, by definition, it does not have a relationship with the real world other than one mediated by the models that we endow it with."

In other words, AI works great if you can assume a spherical cow.

Illustrations: The spherical cow that mocks unrealistic scientific models drawn jumping over the moon by Ingrid Kallick for the 1996 meeting of the American Astronomical Association (via Wikimedia).

Wendy M. Grossman is the 2013 winner of the Enigma Award. Her Web site has an extensive archive of her books, articles, and music, and an archive of earlier columns in this series. Stories about the border wars between cyberspace and real life are posted occasionally during the week at the net.wars Pinboard - or follow on Twitter.

November 18, 2022

Being the product

Twtter-bird-upside-down-370.jpgThe past week of Twitter has been marked by a general sense of waiting for the crash, heightened because no one knows when the bad thing will happen or what form it will take. On Twitter itself, I see everyone mourning the incoming loss and setting up camp elsewhere; on professional media journalists are frantically trying to report on what's going on at HQ, where there is now no communications team and precious few engineers.

As noted here last week, it is definitely not so simple as Twitter's loss is Mastodon's/Discord/s/SomeOtherSite's gain.

The general sense of anxiety feels like a localized version of the years of the Trump presidency - that is, people logging in constantly to check, "What's he done now?" Only the "he" is of course new owner Elon Musk, and the "what" is stuff like a team has been fired, someone crucial has quit, there's been a new order to employees ("check this box by 5pm or you're fired!"), making yet another change to the system of blue ticks that may or may not verify a person's identity, or appearing to disable two-factor authentication via SMS shortly after announcing the shutdown of "20% of microservices". This kind of thing makes everyone jumpy. Every tiny glitch could be the first sign that Twitter is crumbling around the edges before cascading into failure, will the process look like HAL losing its marbles in the movie 2001: A Space Odyseey? Ot will it just go black like the end of The Sopranos?

I have never felt so conscious of my data: 15 years of tweets and direct messages all held hostage inside a system with a renegade owner no one trusts. Deleting it feels like killing my past; leaving it in place teems with risks.

The risk level has been abruptly raised by the departure of various security and privacy personnel from Twitter's staff, which led Michael Veale to warn that the platform should be regarded as dangerously vulnerable and insecure. Veale went on to provide instructions for using the law (that is, the General Data Protection Regulation) rather than just Twitter's tools, to delete your data.

Some of my more cautious friends have been regularly deleting their data all along - at the end of every couple of weeks, or every six months, mostly to ensure they can't suddenly become a pariah for something they posted casually five years ago. (It turns out this is a function that Mastodon will automate through user settings.) But, as Veale asks, how do you know Twitter is really deleting the data? Hence his suggestion of applying the law: it gives your request teeth. But is there anyone left at Twitter to respond to legal requests?

The general sense of uncertainty is heightened by things like the reports I saw of strange behavior in response to requests to download account archives: instead of just asking for two-factor authentication before proceeding, the site sent these users to the help center and a form demanding government ID. There seem to be a number of these little weirdnesses, and they're raising users' overall distrust of the system and the sense that we're all just waiting for the thing to break and our data to become an asset in a fire sale - or for a major hack in which all our data gets auctioned on the dark web.

"If you're not paying for the product, you're the product," goes the saying (attribution uncertain). Right now, it feels like we're waiting to find out our product status.

Meanwhile, Apple has spent years now promoting its products by claiming they provide better privacy than the alternatives. It is currently helping destroy the revenue base of Meta (owner of Instagram, Facebook, and WhatsApp) by allowing users to opt to block third-party trackers on its devices. At The Drum, Chris Sutclifee cites estimates that 62% of Apple users have done so; at Forbes Daniel Newman reported in February that Meta projected that the move would cost the company $10 billion in lost ad sales this year. The financial results it's announced since have been accordingly grim.

Part of the point of this is that Apple's promise appeared to be that the money its customers pay for hardware and services also buys them privacy. This week, Tom Germain reported at Gizmodo that Apple's own apps continue to harvest data about users' every move even when those users have - they thought - turned data collection off.

"Even if you're paying for the product, you're the product," Cory Doctorow wrote on discovering this. Double-dipping is familiar in other contexts. But here Apple has broken the pay-with-data bargain that made the web. It may live to regret this; collecting data to which it has exclusive access while shutting down competitors has attracted the attention of German antitrust regulators.

If that's where the commercial world is going, the appeal of something like Mastodon, where we are *not* the product, and where accounts can be moved to other interoperable servers at any time, is obvious. But, as I've written before about professional media, the money to pay for services and servers has to come from *somewhere*. If we're not going to pay with data, then...how?

Illustrations: Twitter flies upside down.

Wendy M. Grossman is the 2013 winner of the Enigma Award. Her Web site has an extensive archive of her books, articles, and music, and an archive of earlier columns in this series. Stories about the border wars between cyberspace and real life are posted occasionally during the week at the net.wars Pinboard - or follow on Twitter or | | Comments (0) | TrackBacks (0)

November 11, 2022

Moving day

twitter-lettuce-FgnQBzFVIAEox-K-370.jpeg""On the Internet, your home always leaves you," someone observed on Twitter some months back.

Probably everyone who's been online for any length of time has had this experience. That site you visit every day, that's full of memories and familiar people suddenly is no more. Usually, the problem is a new owner, who buys it and closes it down (Television Without Pity, Geocities), or alters it beyond recognition (CompuServe). Or its paradigm falls out of fashion and users leach away until the juice is gone, the fate of many of the early text-based systems.

As the world and all have been reporting - because so many journalists make their online homes there - Twitter is in trouble. A new owner with poor impulse control and a new idea every day - Twitter will be a financial service! (like WeChat?) Twitter will be the world's leading source of accurate information! (like Wikipedia?) Twitter can do multimedia! (like TikTok?), who is driving out what staff he hasn't fired.

The result, Chris Stokel-Walker predicts, will be escalating degradation of the infrastructure - and possibly, Mike Masnick writes, violations of the company's 2011 20-year consent decree with the US Federal Trade Commission, which could ultimately cost the company billions, in addition to the $13 billion in debt Musk added to the company's existing debt load in order to purchase it.

All of that - and the unfolding sequelae Maria Farrell details - will no doubt be a widely used case study at business schools someday.

For me, Twitter has been a fantastic resource. In the 15 years since I created my account, Twitter is where I've followed breaking news, connected with friends, found expert communities. Tight clusters are, Peter Coy finds at the New York Times, why Twitter has been unexpectedly resilient despite its lack of profitability.

But my use of Twitter has nothing in common with its use by those with millions of followers. At that level, it's a broadcast medium. My own experience of chatting with friends or responding randomly to strangers' queries is largely closed to them. Like traveling on the subway, they *can* do it, but not the way the rest of us can. For someone in that position, Twitter is a large audience that fortuitously includes journalists, politicians, and entertainers. The writer Stephen King had the right reaction to the suggestion that verified accounts should pay $20 a month (since reduced to $8) for the privilege: screw *that*. Though even average Twitter users will resist paying to be sold to the advertisers who ultimately fund it the service.

Unusually, a number of alternative platforms are ready and waiting for disaffected Twitter users to experiment with. Chief among them is Mastodon, which looks enough like Twitter to suggest an easy learning curve. There are, however, profound differences, most of them good. Mastodon is a protocol, not a site; like the web, email, or Usenet, anyone can set up a server ("instance") using open source software and connect to other instances. You can form a community on a local instance - or you can use your account as merely a convenient address from which to access postings by users at dozens of other instances. One consequence of this is that hashtags are very much more important in helping people find each other and the postings they're interested in.

Over the last week, I've seen a lot of people trying to be considerate of the natives and their culture, most particularly that they are much more sensitive about content warnings. The reality remains, though, that Mastodon's user base has doubled in a week, and that level of influx will inevitably bring change - if they stay and post, and particularly if many of them adopt a bit of software that allows automated cross-posting between the two services.

All of this has happened without a commercial interest: no one owns Mastodon, it has no ads, and no one is recruiting Twitter users. But that right there may be the biggest problem: the huge influx of new users doesn't bring revenue or staff to help manage it. This will be a big, unplanned test of the system's resilience.

Many are now predicting Twitter's total demise, not least because new owner Elon Musk himself has told employees that the company may become bankrupt due to its burn rate (some of which is his own fault, as previously noted). Barring the system going offline, though, habit is a strong motivator, and it's more likely that many people will treat the new accounts they've set up as "in case of need".

But some will move, because unlike other such situations, whole communities can move together to Mastodon, aided by its ability to ingest lists. I'm seeing people compile lists of accounts in various academic fields, of journalists, of scientists. There are even tools that scans the bios of your Twitter contacts for Mastodon addresses and compiles them into a personal list, which, again, can be easily imported.

If Mastodon works for Twitter's hundreds of millions, there is a big upside: communities don't have to depend for their existence on the grace and favor of a commercial owner. Ultimately, the reason Musk now owns Twitter is he offered shareholders a lucrative exit. They didn't have to care about *us*. And they didn't.

Illustrations: Twitter versus lettuce (via Sheon Han on Twitter).

Wendy M. Grossman is the 2013 winner of the Enigma Award. Her Web site has an extensive archive of her books, articles, and music, and an archive of earlier columns in this series. Stories about the border wars between cyberspace and real life are posted occasionally during the week at the net.wars Pinboard - or follow on Twitter or Mastodon

November 4, 2022

Meaningful access

Screenshot from 2022-11-04 12-56-46-370.jpg"We talk as if being online is a choice," Sonia Livingstone commented on , "but we live in a context and all the decisions around us matter."

As we've observed before, it's only for the most privileged that *not* being online or *not* carrying a smartphone comes without cost.

Livingstone was speaking on a panel on digital inequalities at this week's UK IGF, an annual forum that mulls UK concerns over Internet governance in order to feed them into the larger global conversation on such matters (IGF). The panel highlighted two groups most vulnerable to digital exclusion: old people and children.

According to Ofcom's 2022 Online Nations report, in 2021 6% of British over-18s did not have Internet access at home. That average is, however, heavily skewed by over-65s, 20% of whom don't have Internet access at home and another 7% of whom have Internet access at home but don't use it. In the other age groups, the percentage without home access starts at 1% for 18-24 and rises to 3% for 44-54. The gap across ages is startlingly larger than the gap across economic groups, although obviously there's overlap: Age UK estimated in 2021 that 2 million pensioners were living in poverty.

I know one of the people in that 20%. She is adamant that there is nothing the Internet has to offer that she could possibly want. (I feel this way about cryptocurrencies.) Because, fortunately, the social groups she's involved in are kind, tolerant, and small, the impact of this refusal probably falls more on them than on her: they have to make the phone calls and send the printed-out newsletters to ensure she's kept in the loop. And they do.

Another friend, whose acquaintance with the workings of his computer is so nodding that he gets his son round to delete some files when his hard drive fills up, would happily do without it - except that his failing mobility means that he finds entertainment by playing online poker. To him, the computer is a necessary, but despised, evil. In Ofcom's figures, he'd look all right - Internet access at home, uses it near-daily. But the reality is that despite his undeniable intelligence he's barely capable of doing much beyond reading his email and loading the poker site. Worse, he has no interest in learning anything more; he just hates all of it. Is that what we mean by "Internet access"?

These two are what people generally think of when they talk about the "digital divide".

As Sally West, policy manager for Age UK, noted, if you're not online it's becoming increasingly difficult to do mundane things like book a GP appointment or do any kind of banking. Worse, isolation during the pandemic led some to stop using the Internet because they didn't have their customary family support. In its report on older people and the Internet, Age UK found that about half a million over-65s have stopped using the Internet. And, West said, unlike riding a bike, Internet skills don't necessarily stay with you when you stop using them. Even if they do, they lose relevance as the technology changes.

For children, lack of access translates into educational disadvantage and severely constricted life opportunities. Despite the government's distribution of laptops. Nominet's Digital Youth Index finds that a quarter of young people lack access to one, and 16% rely primarily on mobile data. And, said Jess Barrett, children lack understanding of privacy and security yet are often expected to be their family's digital expert.

More significantly, the Ofcom report finds that 20% of people - and a *third* of people aged 25-34 - used only a smartphone to go online 2021. That's *double* the number in 2020. Ofcom suggests that staying home much of 2020 and newer smartphones' larger screens may be relevant factors. I'd guess that economic uncertainty played an important role and that 2022's cost-of-living crisis will cause these numbers to rise again. There's also a generational aspect; today's 30-year-olds got their teenaged independence via smart phones.

To Old Net Curmudgeons, phone-only access isn't really *Internet* access; it's walled-garden apps. Where the open Internet promised that all of us could build and distribute things, apps limit us to consuming what the apps' developers allow. This is not petty snobbery; creating the next generation of technology pioneers requires learning as active users instead of lurkers.

This disenfranchisement led Lizzie Coles-Kemp to an approach that's rarely discussed: "We need to think how to design services for limited access, and we need to think what access means. It's not binary." This approach is essential as the of the mobile phone world's values risk overwhelming those of the open Internet.

In response, Livingstone mooted the idea of "meaningful access": the right device for the context and sufficient skills and knowledge that you can do what you need to.

The growing cost-of-living crisis, exacerbated this week by an interest rate rise, makes it easy to predict a marked further rise in households that jettison fixed-line broadband. This year may be the first since the Internet began in which online access in the UK shrinks.

"We are just highlighting two groups," Livingstone concluded. "But the big problem is poverty and exclusion. Solve those, and it fixes it."

Illustrations: UK IGF's panel on digital inequalities: Cliff Manning, Sally West, Sonia Livingstone, Lizzie Coles-Kemp, Jess Barrett,

Wendy M. Grossman is the 2013 winner of the Enigma Award. Her Web site has an extensive archive of her books, articles, and music, and an archive of earlier columns in this series. Stories about the border wars between cyberspace and real life are posted occasionally during the week on Twitter or @wendyg@mastodon.xyz.