I’m headed to Boston to attend the Tufte seminar tomorrow. It’s been nearly fifteen years since I bought — and devoured — The Visual Display of Quantitative Information. Even if there’s nothing new in the seminar beyond what I’ve already learned from the four books, I’m excited to see one of my heroes live. Over the past few weeks, I devoured the three extant Wayfarers novels by Becky Chambers: The Long Way to a Small Angry Planet, A Closed and Common Orbit, and Records of a Spaceborne Few.
Having taken away the traditional means of measuring ad effectiveness with Intelligent Tracking Protection, the Safari team is now offering an alternative in “Privacy Preserving Ad Click Attribution For the Web” (emphasis mine). Online ads and measurement of their effectiveness do not require Site A, where you clicked an ad, to learn that you purchased something on Site B. The only data needed for measurement is that someone who clicked an ad on Site A made a purchase on Site B.
Chris Bowler reflects on the relative importance of his habits versus projects: But truthfully, it doesn’t really matter if I never complete this project. On the other hand, doing an exercise 5 days per week to strengthen my core makes a big difference to my life. So too with running four times per week (and the first habit makes the second more doable). Helping my son with his reading makes a huge difference in his life.
“Using Artificial Intelligence to Augment Human Intelligence” in Distill: By being forced to find a compact description of the training examples, the neural net learns an abstract, higher-level model of what a font is. That higher-level model makes it possible to generalize beyond the training examples already seen, to produce realistic-looking fonts. Shan Carter and Michael Nielsen describe a system where a neural network learns compact representation of the vectors along which attributes of font vary and has the capability to review new primitives to describe font design or to discover and make explicit new design principles.
Distill is a new machine learning journal, designed to support interactive visualizations that clarify the ideas in their articles – much like how Bret Victor approaches explanations. Editors Shan Carter and Chris Olah lay out their vision in “Research Debt”: In research, we often have a group of researchers all trying to understand each other. Just like before, the cost of explaining stays constant as the group grows, but the cost of understanding increases with each new member.
The Slack blog describes in “Slack 101: Onboarding” how they use their own chat platform to manage their work. My first reaction to Slack was “OK this is cute but not substantially better than HipChat or Campfire”. Now that I’m a member of 11 (!!!) different Slack communities1, I’m convinced that Slack is an incrementally better team chat system. Reading about how their own internal channel-naming conventions has me convinced that they get the information architecture of communication in a deep way.
My hypertext idealism is still alive - and stirred today by Mike Caulfield’s“The Garden and the Stream: A Technopastoral”, describing his experience of maintaining a federated personal wiki. And weirdly, these links were compiled over the space of a year, just by noting things I learned or heard and linking them to things I’d heard before or that others had written. I created a wiki on issues of found art without even knowing it.
Physicist Sabine Hossenfelder offers consulting for amateur theorists, to help them connect with the mainstream research community. My clients know so little about current research in physics, they aren’t even aware they’re in a foreign country. They have no clue how far they are from making themselves understood. Their ideas aren’t bad; they are raw versions of ideas that underlie established research programmes. But those who seek my advice lack the mathematical background to build anything interesting on their intuitions.
J.D. Evans and Sam Lebovic compare “the market” to another collective system we all participate in through our daily lives: the transit system. In the United States, we are used to the term “the market” being thrown around in public discourse, as though it refers to a single thing. In fact, there are many markets, all nested and interconnected. Some parts are private, some public. Optimizing for the growth of the whole loses sight of the goals of the participating agents, which the analogy to roadways makes clear:
Camille Fournier on the value of complaining: If you do this well, you actually teach people how to understand which problems are important, and which problems are not. Letting people complain might seem like it will do nothing but encourage negativity and drama, but if you guide people to learn from their complaints it can instead help your team grow. It’s great when people can bring problems AND solutions to you simultaneously, but it’s more likely that they will need help to see the best solution.
From a Neil Postman’s “Bullshit and the Art of Crap-Detection” (emphasis mine): Each person’s crap-detector is embedded in their value system; if you want to teach the art of crap-detecting, you must help students become aware of their values. After all, Vice President, Spiro Agnew, or his writers, know as much about semantics as anyone in this room. What he is lacking has very little to do with technique, and almost everything to do with values.
“Learning How to Read”, by Niklas Luhmann: This leads to another question: what are we to do with what we have written down? Certainly, at first we will produce mostly garbage. But we have been educated to expect something useful from our activities and soon lose confidence if nothing useful seems to result. We should therefore reflect on whether and how we arrange our notes so that they are available for later access.
David Robinson wrote about his experience after a year as a Data Scientist at Stack Overflow. I haven’t read it yet, but it the opening paragraph turned me on to his series about analyzing baseball statistics using the beta distribution. Understanding the beta distribution Understanding empirical Bayes estimation Understanding credible intervals Understanding the Bayesian approach to false discovery rates Understanding Bayesian A/B testing Understanding beta binomial regression This is a great series of posts – I’ve been learning how to use the beta distribution to estimate uncertainties in clickthrough rates, so finding David Robinson’s blog was like stumbling into the lecture hall for a class I didn’t know I needed.
While home sick this week, I started learning to play Go. Partly inspired the article on AlphaGo in Wired, but also because it’s a pretty game. Hiroki Mori’s site The Interactive Way to Go is an awesome resource1. I haven’t gotten to the point of actually attempting to play a full game, but working through Go problems is super fun. Some of the patterns that emerge in Go (like “Crane in the nest” and the “Ladder”) unfold with a inevitability that makes me think of stable structures in Conway’s Game of Life.
Sarah Jane Coffey, on the experience of working in start-ups as a sober alcoholic: We began placing orders for the family style meal, when my boss suddenly announced to the room that I wouldn’t be eating meat. People began to chuckle, saying things like, “How can you be in Argentina and not try the meat!?” “You’re not going to try the beef? You don’t know what you’re missing!” Then, my boss added, “Yeah, she doesn’t drink either.
Visiting Indiana this week, with a brief side-trip to the Champaign-Urbana area in Illinois. Vacation reading: Blindsight by Peter Watts. File under “terrifying mega-conceptual first contact thriller”. Among many other fun ideas, Watts introduces vampires as a predatory homo sapiens offshoot who, due to a neurological defect, experience seizures at the sight of right angles. That is, unless they are taking their anti-Euclideans. Site Reliability Engineering, by various Googlers. Essays on how Google operates its software.
Duretti Hirpa, on using pull requests to structure time and progress: I connect present and future by scribbling notes to myself, and constructing narratives for others – which intersects beautifully as changesets, and, in particular, as a fleshed-out pull request description. I believe coherent pull requests are an act of empathy, for the person reviewing my changes, for the me in the just now – the me trying to get a quality review – and for the me in the what’s next?
Aurynn Shaw, on language/editor/tech wars in “Contempt Culture”: I was taught to be contemptuous of the non-blessed narratives, and I was taught to pay for my continued access to the technical communities through perpetuating that contempt. I was taught to have an elevated sense of self-worth, driven by the elitism baked into the hacker ethos as I learned to program. By adopting the same patterns that other, more knowledgable people expressed I could feel more credible, more like a real part of the community, more like I belonged.
Tina Rosenberg reports (in The Guardian) on a homeless shelter in Ottawa that stabilizes the lives of alcoholics by giving them alcohol under supervision: The pour is calculated for each resident to be just enough to stave off the shakes and sweats of detox, which for alcohol is particularly unpleasant – seizures from alcohol deprivation can be fatal. The pour is strictly regulated: Young cuts off anyone who comes in intoxicated.
Sarah Perry summarizes and recommends John Stuart Mill’s defense of free speech: Perhaps more importantly, in a society in which dissenting views are rarely expressed, the prevailing views are often held as mere slogans or prejudices, with people rarely understanding the justifications or arguments for why the prevailing view is correct. In my introductory example, I ask why free speech is important. But if free speech is never challenged as a value, how can we be intellectually and emotionally aware of the foundations and reasons why free speech is good?
Cal Newport, on scheduling margin around your meetings: There are few experiences more stressful than a day in which your schedule is so fractured with appointments to talk about work that you have no time to act on the results of all this discussion – leading, instead, to the awful sense of a growing stack of obligations, all being juggled in your head, that you have no idea how to define or handle.
John D. Cook, on the relationship between credentials and knowledge: When I was in college, a friend of mine gave me a math book that I found hard to get through. When I complained about it, he told me “You’re going to finish a PhD someday. When you do, do you think there’s going to be fairy dust on the diploma that’s going to enable you to do anything you can’t do now?
A discussion on the Beeminder forum got me amped up about speed reading1. I ordered Kump’s Breakthrough Rapid Reading and started following the drills, but then found Scott Young’s debunking of speed reading. Kump’s method can at least improve skimming speed – which itself might improve comprehension (skim for structure, then go back and read closely) – but hitting one thousand words per minute with full comprehension may not be anatomically possible.
“There is No Secret Notebook”: The feeling is that the answer to your question belongs to people who have some essential quality that you forever lack. You look at yourself, with your slow and fragile reasoning power, and you feel like you’re counting on your fingers, and imagine that someone out there has a supercomputer. (Or maybe that everybody on Earth but you has a supercomputer.) This is an illusion.
Matthew Noah Smith argues in Slate that law enforcement having access to personal smartphone data is a reasonable analogue to mind reading: This is especially the case when it comes to the role that our phones play in both communication and information storage. There is simply no principled distinction between the processes occurring in the meaty glob in your cranium and the processes occurring in the little silicon, metal, and glass block that is your iPhone.
When you’re driving, talking to a passenger is safer than talking to someone on the phone. Why? The passenger shares your situational awareness and knows when the conversation has to slow down for the sake of safety. Malcolm Ocean considers this as a metaphor for working with an instant messaging client open. Whereas if someone were talking to me in person, they might pick up on the physical cues of my situation, and see that I’m in a rush.
Julie Zhuo writes on Medium about trading quality for scope or delivery speed in a project. […] to create high-quality work, there has to be a minimum acceptable bar. And high-quality creators cannot trade off below that bar. They simply can’t. It would be inauthentic to who they are. It doesn’t matter if their peers, their boss, the whole wide world told them that this bar didn’t matter and that the right decision is to give up a bit of quality for speed or time or money or whatever.
Temple Grandin – in a Q&A with Grist – calls out the danger of inexperience with skills constrained by physicality. We’ve got to figure out sensible things to do. The thing that worries me on a lot of these issues is that we’ve got more and more people getting involved who have never done anything practical, because schools have taken out all the cooking, sewing, woodworking, and art. And in the real world of practical things, nothing can be perfect.