"Move fast and break things" was the internal motto of Facebook up until 2014 they realized, a little belatedly, that perhaps a corporation with worldwide reach and hundreds of millions of active users shouldn't make recklessness its motivating principle.
Imagine a child playing in a sandbox, dreaming up elaborate stories about the castles they've built with their sandbox toys. Now, imagine a group of eighteen adults, two teams of office co-workers, throwing around a baseball according to very precise rules. Those adults are playing too.
People tend to think of laws as set in stone, but they're very changeable. They're changeable in at least two ways: the laws themselves can be altered, through formal legal mechanisms like passing a bill through Congress, or laws can change through how they're enforced on the ground.
Information naturally flows from those who have more power to those who have less.
As we've discussed previously, all systems are built on abstractions, and all abstractions are at least a little bit wrong. The more complex and important a system is, the more likely it is that a leaky abstraction will gum up the works. So systems need feedback mechanisms to detect when an abstraction is leaking, so they can identify and fix the problem.
Technologies are built with a model of the world in mind. A hammer presumes a nail, a screwdriver a screw, and so on. But there is, necessarily, a gap between the model and the reality, a gap between how a technology is meant to be used and how it actually functions in the world.
In my last post, I talked about how all systems have flaws, and how these flaws require interpretive labor to be patched or worked around so that the system can keep running. These systems become dysfunctional when it is hard or impossible to learn from that interpretive labor what those flaws are. Without good feedback mechanisms, none of the flaws will be fixed, and the system will not adapt to changes.
As I've said a few times now, all models are wrong. That includes laws and norms, employee handbooks and product roadmaps, system diagrams and city plans. If something is abstracting away from reality, it's creating a gap. What happens when that gap causes problems?
I've posted before about why it's important to delegate decisions to the people impacted by them. For example, Toyota put the power to halt production in the hands of factory floor workers who were better able to see problems as they arose, and the German Army pushed local officers to be decisive in the midst of battle. This is such an effective strategy because factory floor workers and army officers tend to have much better knowledge of the local contexts where problems first arise.
All models are wrong. Given this, it should be no surprise that humans are constantly messing things up. But don't worry! That's often a good thing, because messing things up is how we learn.
There's a saying: the map is not the territory. Farnham Street Blog has a pretty good explainer:
My sister is interested in genealogy, and several years ago unearthed a gruesome family story: on June 11th, 1880, my great-great-great-grandfather Anthony was caught between two railroad cars and crushed. The newspaper called it a "horrible death", and added that he left behind a wife and five kids "in destitute circumstances".
Few people are happy with the US government these days. Recent Gallup polls show that 71% of people are somewhat or very dissatisfied by "our system of government and how well it works". Common complaints are that government's too polarized and full of conflict, or that it's bureaucratic and inefficient and can't get things done.
Sometimes, you just want someone to tell you what to do.
"Nothing about us without us!" is a rallying cry, a catch-phrase with a clear moral imperative: that those impacted by a decision should be involved in making the decision.
The following is from my response to a friend’s Facebook post about the value of postmodernism to engineers, and the role that education plays in introducing engineers to postmodernism.
Last week I gave two short talks as part of the IASC Knowledge Commons conference, one on our ability to contest rules in digital spaces, and another on how, more generally, we can engineer more responsive governance. Those links go to YouTube, where you can watch the talks. (Update: you can now also see the panel discussions for these talks: code as contestable law, engineering responsive governance.)
One of my communities is choosing a new governance structure, and in preparation for our first conversation on the topic, I wrote up three governance case studies. I covered DisCOs (Distributed Cooperative Organizations), Enspiral, and Python.
Facebook's 'Oversight Board' recently affirmed Facebook's choice to suspend then-President Donald Trump from its platform the day after the Capitol riots. Although many critics - including myself - support the decision to keep Trump off Facebook, we're not too pleased with the process itself. "[It's] a red herring, substituting a simulacrum of due process in certain high-profile cases for substantive reform," Will Oremus writes in a New York Times op ed. Oremus and others point out that while Facebook has promised to obey the content moderation decisions of the board, it can choose to ignore any policy recommendations. "[Facebook] did not empower the board to watch over its products or systems — only its rules and how it applies them."
There’s not enough disobedience online. Yes, that’s right, you heard me: there’s not enough space for disobedience online.
Note: I recently started writing a newsletter about the intersection of technology and governance,Hello, Governor! I’m still experimenting with format, content, etc, so feel free to give feedback about what you like or what you’d want to see.
It can be a struggle to hold groups accountable for misbehavior, even when the group is structured through legal forms, with public membership and clear leadership. On social media platforms populated by anonymous accounts, with membership and leadership often informal and invisible, accountability feels like a pipe dream.
... so I put some tech governance controversy on the platform you were using for your tech governance newsletter.
Below, I summarize the FTC complaint against Facebook, citing the relevant paragraphs so you can get more detail as needed. This was a useful exercise in helping me understand the case; I hope it also helps you.
Hello, and welcome to the very first edition of this newsletter!
It’s the last day of my mini-batch, although it’s not my last day at Recurse, since there’s an active alum community I’m eager to be a part of.
Today I finally tackled making a code madlibs program.
When we left off, I’d written some rather ugly code that gave us the basic features of a madlibs program without using any macros. But the whole point of this project is to better understand lisp macros. So today I focused on refactoring my program to use macros instead.
Building off of yesterday’s work, today my hope was to finish my humble text-focused Madlibs program, so that I could move onto code-focused Madlibs tomorrow.
Day 1 of my Recurse mini batch is complete! I spent a good portion of it meeting people and learning about the community, but I did also make some small steps on my project:
Next week I’ll be doing a mini-batch at Recurse. I’m planning to blog about it each day, and I figured I’d start by laying out my plans for the week.
I just fixed a Django bug that took me ages to solve. Part of why it took so long is that the errors I was getting weren’t surfacing the right search results. So, in case anyone ever has a similar problem, I will record what happened here and hope the search algorithms send this their way.
I have read a great many non-fiction books in my life, and I remember the details of heartbreakingly few of them. Over the last year or so, I have worked to change that pattern.
This twitter thread is a great example of "minding the gap", that is: doing the necessary improvisational work to complete a task that has only been partially, abstractly specified. I recently learned another term for this: bricolage.
When chants of ‘equal pay!’ ring through soccer stadiums, men jump on Twitter to explain why, despite performing better internationally than the men’s team, women soccer players don’t deserve equal pay because they don’t earn as much revenue.
Debunking the Capitalist Cowboy by Nan Estad at Boston Review:
Via Martha Nussbaum’s Poetic Justice, an excerpt from Woodson v. North Carolina (1976):
This TED talk by Shonda Rhimes underscores the importance of play:
It’s a Crisis of Civilization in Mexico by José de Córdoba at the Wall Street Journal:
The sum of all known reverence I add up in you whoever you are, The President is there in the White House for you, it is not you who are here for him, The Secretaries act in their bureaus for you, not you here for them, The Congress convenes every Twelfth-month for you, Laws, courts, the forming of States, the charters of cities, the going and coming of commerce and mails, are all for you.
“Many scientists prefer not to think about politics, but the work that scientists do has always had political consequences, from the quiet tragedy of an understudied disease to the undeniable horror of a city destroyed in an instant.
One prominent idea of rights, common in the US political and legal tradition, understands rights to be barriers against interfering state action: if the state just keeps its hands off, rights are taken to have been secured. The Capabilities Approach, by contrast, insists that all entitlements involve an affirmative task for government: it must actively support people’s capabilities, not just fail to set up obstacles. In the absence of action, rights are mere words on paper. Vasanti was not beaten by the government of Gujarat; she was beaten by her husband. But a government that does not make and then actively enforce laws against domestic violence, or give women the education and skills they need to get a living wage if they leave an abusive marriage, is accountable for the indignity such a woman endures. Fundamental rights are only words unless and until they are made real by government action. The very idea of “negative liberty”, often heard in this connection, is an incoherent idea: all liberties are positive, meaning liberties to do or to be something; and all require the inhibition of interference by others.”
Imagine a child playing in a sandbox, dreaming up elaborate stories about the castles they’ve built with their sandbox toys. Now, imagine a group of eighteen adults, two teams of office co-workers, throwing around a baseball according to very precise rules. Those adults are playing too.
With the exception of rare contrivances like king’s missives and broadcast television, we learn most of what we learn and meet most of the people we meet through networking. This has always been true, but it’s clearer than ever with social media. To be honest I rather like it. When a friend I trust or a writer I admire shares a link I tend to follow along enthusiastically.
I posted a few months ago about negative capability – that is, the ability to tolerate uncertainty, “when a man is capable of being in uncertainties, mysteries, doubts, without any irritable reaching after fact and reason.” Coined by the Romantic poet Keats, the term is easily associated with art but applies in all disciplines and all areas of life. Uncertainty pervades everything. So it behooves us to come to terms with it.
This link goes to a fascinating little explainer by Popehat on Twitter. Excerpted here and lightly edited for readability:
Last week I was on a panel about ‘Democracy and the Digital Commons’ at Suffolk University. At the start of the panel, each of us gave a 5-10 minute talk to help frame the discussion. While there’s no transcript of the panel itself, here are my notes for the intro. (Quick context: each of us tied our talk to the Boston Marathon bombing and in particular Reddit’s response to it.)
Back in high school and college, I had friends who offered to do tarot readings for me. I always turned them down. “What a dumb idea,” I’d think. “How could a deck of cards predict your future!”
Almost five years ago now, frustrated by the lack of trans-inclusive children’s books that my best friends’ kids had as options, I sat down and wrote a story about a tiny trans/gender non-conforming superhero named Super Princess. It’s so exciting to finally be able to put this book out into the world. This is a story about the magic of empathy and the importance of approaching the world with love instead of fear.
We often talk about giving people the benefit of the doubt, but seldom talk about its opposite, to the point that no agreed upon phrase for it exists. The best I could come up with is the “burden of doubt”, which largely applies to courtroom settings. Even with the help of judicial documents, the phrase is not very popular.
How to Survive America’s Kill List by Matt Taibbi at Rolling Stone - an American citizen’s fight to escape assassination by the US executive branch:
“[Socrates] adds that as far as he himself is concerned he believes that “it would be better for me that my lyre or a chorus I direct were out of tune and loud with discord, and that most men should not agree with me and contradict me, rather than that I, being one, should be out of tune with myself and contradict myself”. The key notion in this sentence is “I who am one” which is unfortunately left out in many English translations. The meaning is clear: even though I am one, I am not simply one, I have a self and I am related to this self as my own self. This self is by no means an illusion; it makes itself heard by talking to me - I talk to myself, I am not only aware of myself - and in this sense, though I am one, I am two-in-one and there can be harmony or disharmony with the self. If I disagree with other people, I can walk away; but I cannot walk away from myself, and therefore I better first try to be in agreement with myself before I take all others into consideration. This same sentence also spells out the actual reason it is better to suffer wrong than to do wrong: if I do wrong I am condemned to live together with a wrongdoer in unbearable intimacy; I can never get rid of him.”
The problem is all inside your head she said to me The answer is easy if you take it logically I’d like to help you in your struggle to be free There must be fifty ways to win the midterms
This old tweet has recently been making the rounds and sparked up some discussion among my Facebook friends:
There is a lot to be said about what happened to Serena Williams and Naomi Osaka yesterday, but this tweet by Jennifer Richeson captures the analysis of the incident that’s most relevant to this blog:
I recently read David Ciepley’s “Beyond Public and Private: Toward a Political Theory of the Corporation”. I recommend reading it yourself, but here’s my best attempt to summarize it in my own words.
Last week I went to a rally to protest a series of raids by ICE in my city. The rally turned into an unplanned march through the streets, and I had to make two quick decisions: first, whether to join the march, and second, whether to remain in the street when the police started to give warnings.
Principles are really important, and by and large you should try hard to stick to them. I have tremendous respect for those who have died or gone to jail for their principles.
It is in the very nature of a question like “What do I owe my parents?” that there is not and can never be a final, numerically answer. It is a question that we re-visit and re-negotiate every minute we are with them; obligation and love form an endless Möbius strip, through which our complex interdependence on each other makes the idea of paying off that debt – and of thereby severing the relationship – a sort of bitter joke. Precisely because it is a non-monetary “debt,” its function is to be an unpayable and unbreakable bond, one whose dividends never end and one that could and will never default.
This is a quote from Norman Thomas’s Is Conscience a Crime?, a book about conscientious objection during WWI from the wizened vantage point of 1923. Thomas was a six time socialist candidate for President; Secretary Baker, referred to here, was the Secretary of War under Woodrow Wilson.
How to Not Die in America by Molly Osberg at Splinter News:
Socrates, however, who is commonly said to have believed in the teachability of virtue, seems indeed to have held that talking and thinking about piety, justice, courage, and the rest were liable to make men more pious, more just, more courageous, even though they were not given either definitions or “values” to direct further conduct. What Socrates actually believed in in such matters can best be illustrated by the similes he applied to himself. He called himself a gadfly and a midwife, and, according to Plato, was called by somebody else an “electric ray”, a fish that paralyzes and numbs by contact, a likeness whose appropriateness he recognized under the condition that it be understood that “the electric ray paralyzes others only through being paralyzed itself. It isn’t that, knowing the answers myself I perplex other people. The truth is rather that I infect them also with the perplexity I feel myself.”
PyCon is one of my favorite events - the only big conference I’ve been to that feels like a small one. This year it was in Cleveland, Ohio, not too far from where I live, so three other pythonistas and I piled into a car together and turned it into a road trip. We had a blast and it made PyCon seem even more like a summer vacation.
This originated as a post to a mailing list on the subject of blockchains and how they might help the cause of open science. The quote below is the claim I was directly responding to.
The point is partly an intergenerational one. Recall from the previous chapter the idea that inequalities of outcomes in one generation becomes inequalities of opportunity in the next and its analog: free associations in one generation become inherited ways of life in the next. This is a necessary truth; children are born into particular times and places and social worlds that have been shaped by the choices their parents have made. This does not simply mean that the parents were free and the children were not; it was also true of the parents that they were born into particular times and places and social worlds. If the parents had some freedom to reshape their worlds in partially original ways, to join or form groups into which they were not born, then the children also have some such freedom. But there could be a narrowing over time; parents can join groups or adopt ways of life that leave their children with fewer choices than they themselves had.
“Refusing the parlements their right of remonstrance was an act of despotism. Liberty had taken shelter in them; it had chosen an irregular method of expressing itself; but since no one had given it another method, it was necessary to allow it that one.”
Here’s a summary of what we know as of the afternoon of Wednesday, March 21st:
I detest and fear dogma. I know that all revolutions must have ideologies to spur them on. That in the heat of conflict these ideologies tend to be smelted into rigid dogmas claiming exclusive possession of the truth, and the keys to paradise, is tragic. Dogma is the enemy of human freedom. Dogma must be watched for and apprehended at every turn and twist of the revolutionary movement. The human spirit glows from that small inner light of doubt whether we are right, while those who believe with complete certainty that they possess the right are dark inside and darken the world outside with cruelty, pain, and injustice. Those who enshrine the poor of Have-Nots are as guilty as other dogmatists and just as dangerous. To diminish the danger that ideology will deteriorate into dogma, and to protect the free, open, questing, and creative mind of man, as well as to allow for change, no ideology should be more specific than that of America’s founding fathers: ‘For the general welfare’. (p. 4)
In an article about Milton Friedman in the Pacific Standard, Rick Paulas quotes from an article of Friedman’s published in 1970:
I wrote an ‘intro to operations’ guide. The ideal audience is me a week ago.
Warning: Black Panther spoilers.
To Be, or Not To Be by Masha Gessen in the New York Review of Books:
Bureaucracy is stupid, David Graeber writes in The Utopia of Rules: On Technology, Stupidity, and the Secret Joys of Bureaucracy. As the premise for a book, it’s more fertile than it sounds.
_The law locks up the man or woman
In an essay at Brain Pickings, Maria Popova sketches out a conception of identity as a collection of fragments which is, paradoxically, being repressed and sanded down by identity politics:
From the studiously moderate Brookings Institution:
Back in college, I based my Division III (senior thesis) research on a set of empathy studies by Nancy Eisenberg. Eisenberg’s line of research hinges on a distinction between two different kinds of empathy: empathic concern, the ability to recognize and care about the hurt others feel, and personal distress, the experiencing of the other person’s hurt yourself. Counterintuitively, personal distress is not positively associated with helping behavior. In fact, it may even decrease the likelihood of helping, if there’s an easy escape route away from the empathy-provoking situation.
In a previous post, I wrote that asking questions is harder than answering them, although I qualified that in a big way with “answering [questions] involves going back over and over again and updating our hypotheses, which makes answering questions feel hard”. I want to revisit this claim.
every so often I remember that the clinton analytics team’s motto was “a decision tree grows in brooklyn” and i smile
Back when I was a research scientist, I straddled the boundary between “hard” and “soft” sciences. I did social psychology, which is a pretty soft science as sciences go, but I paired it with biology and physiology in general and endocrinology in particular, which meant getting a taste for some of the harder stuff.
Public Policy After Utopia by Will Wilkinson at the Niskanen Center:
What shade of gray would you say this is?
This is a short and subtle piece by Diana Senechal about sexual harassment claims and our response to them. I by and large agree with it, but I like it mostly for the poem (or, well, stanza of a poem) it introduced me to, Robert Frost's The Wood Pile:
It's easy to complain about "purity politics". It's easy to complain about "neoliberals" and "sellouts". But we live in hard times, and the easy route's not going to get us anywhere.
Things have been quiet on this here tumblr, in part because I have taken to blogging about a subset of topics here. If you’re interested in my thoughts about “humans and how we interact with each other, with a focus on how we can better deal with complex and large-scale social situations”, I recommend following my new blog as well.
Chalk Corridor by fdecomite 6/25/2013 CC BY 2.0
Mike Montiero recently posted an essay to Medium "One person’s history of Twitter, from beginning to end". It's a good piece, but what really struck me was this referenced Tweet from Twitter co-founder Biz Stone:
I spent a good portion of yesterday staring at my Django test cases and whimpering. No matter what I tried, no matter how thoroughly I flushed the database between each test case, state seemed to be persisting from test to test. “How is this happening?” I howled to my computer.
Andrew Gelman has a brief post up on his blog comparing the way bug reports in open source software are received to the way many researchers respond to criticisms of their work. The comments there are good, and cover my first reaction, which was, "Developers respond well to bug reports?" But that's a bit tongue in cheek. I do think that, overall, developers are a bit more responsive to bug reports than scientists are to published criticisms of their work. Here are my theories as to why that is:
Me: Wow, Django’s Case() sure is a useful addition to the annotate function! You can use it to chain together When() functions to conditionally annotate your queryset! Django is so cool.
The Supermanagerial Reich by Ajay Singh Chaudhary and Raphaële Chappe at the LA Review of Books:
You did a bad thing…
Julia Evans, in her recent summary of her time at the Recurse Center, links to this cool little project which visualizes how you use Git. Here’s my result:
It’s late Friday night, technically Saturday morning, and like a party animal I’m up googling, “Could someone mess up my server if I let them run arbitrary Python code with no imports.”
This might be the best introduction to an academic paper I’ve come across. From Roger Myerson’s Fundamentals of Social Choice Theory:
A quote by Frank Serpico, by way of writer Lawrence Grobel, by way of Metafilter:
Catching up on the Crooked Timber archives today, I saw they’ve done several seminars since last I looked, on three books/series I’d love to read: Jo Walton’s The Just City and The Philosopher Kings,Ada Palmer’sTerra Ignotaseries,and Cory Doctorow’sWalkaway.For those of you unfamiliar with Crooked Timber, it’s a community blog and their ‘seminars’ are when people all post reviews and reflections on the same work within a short period of time.
So I read Kushner’s statement today, which includes the line “I am not a person who has sought the spotlight” and, well, I couldn’t help myself.
So I’m watching my way through this pop math YouTube channel 3Blue1Brown and out of nowhere comes this extraordinarily fascinating fact: the ancient Babylonians circa 1800 BC probably (possibly?) knew the Pythagorean theorem and were generating Pythagorean triplets:
My friend Sumana Harihareswara recently posted a great collection of links on ‘call-out culture’ to MetaFilter.
If someone asks me, “What is open source software?”, I have an obvious answer: “Software that is licensed as open source and makes its source code available for users to review, modify and share.” Ditto “What is open science?”, to which I’d respond: “Science that is made available to review by anybody and available to reproduce by anyone with the necessary skills and equipment.”
I have seen this argument happen over and over again, especially in progressive spaces.
For Valentine’s Day, I want to talk about the poem I posted the other day, “Law, Like Love” by WH Auden, and why it means so much to me.
I’ve been thinking recently about the human tendency to try to put things in context, to search for patterns across different situations and conversations. It’s something I do a lot, and that my friends and the communities I’m a part of do constantly as well. Here are some examples from just the last week or so:
Law, say the gardeners, is the sun, Law is the one All gardeners obey To-morrow, yesterday, to-day.
I visited House Oversight Committee Chair Jason Chaffetz’s office today. If there’s nothing else good to say about him (and there may not be) at the very least his staff is warm and polite, even when you are implying things about their boss.
Let’s say you’ve got an engineer or other employee who costs you $150,000 a year in salary & benefits. And let’s say they lose just 2 days a year to dealing with problems caused by underfunding of the FOSS projects they work with – bugs, missing documentation, lack of features that would make life so much easier.
While I am apparently on a recommendation kick, let me share with you this essay: How Technology Hijacks People’s Minds. Have a handful of thought-provoking pull-quotes.
I’m not usually one for podcasts, but I discovered @fansplaining a few weeks ago and have really been getting a kick out of it. It is a “podcast by, about, and for fandom, hosted by Flourish Klink & Elizabeth Minkel” and it is fantastic.
Earlier today a friend/former colleague of mine linked to a dataset on the OpenScienceFramework with the commentary “OKCupid released some of their data!“ It turns out they didn’t release it, it was scraped.
The intro to testing session I ran on Saturday went well. Unfortunately the resolution on Hangouts on Air recordings is not good enough to read my code, so it’s pretty much unwatchable (though folks who saw it live were able to see what I was doing).
This was originally posted to my now-offline personal blog on February 15th, 2014. I’m reposting it now in honor of International Workers Day.
This Saturday at 11am EST I’ll be running a remote workshop with PyLadies on getting started with testing. If you’re a lady or genderqueer/non-binary person who wants to know more about the nitty gritty of unit tests and integration tests or maybe just the high level conceptual stuff like “why do people do this to themselves” you should consider signing up. I’m not sure if there’s an official cap but I know there are a lot of people enrolled (90ish! o.O) so it’s probably a good idea to RSVP if you’d like to join.
Last fall I had a bit of free time and thought I’d try out making apps. It was definitely a learning experience! I got to work with Java for the first time in a long time, and with the Android ecosystem. Said ecosystem has its good parts and its frustrating parts, but let me just say: I love Android Studio. I’d marry it, if people were allowed to marry integrated development environments.
I read a comment recently about the rules for bidding on government contracts. This isn’t the comment, but it gets at the same issue, which stuck with me:
I have never much liked Hamentaschen but it turns out “cookie” and “hat-shaped” are not terribly limiting constraints and much deliciousness can be fit within them.
So, LibrePlanet happened this weekend. It’s one of my favorite conferences, and not just because it’s local. This year Deb Nicholson and I ran a pre-conference workshop for free software projects to improve their usability and new contributor onboarding, hosted by the lovely folks at Bocoup. According to our participants it was a smashing success. Deb and I are writing up a guide so that other FOSS conferences can run similar workshops.
Today I had the pleasure of speaking to a Mozilla study group about The Little R’er. The Little R’er is a project of mine from about a year ago - it’s basically The Little Schemer for R. Or, rather, the first few chapters of the Little Schemer for R. R is, in my opinion, a much less elegant language than Scheme and I found that the socratic method eventually broke down as a pedagogical tool. The discovery of silent recycling did not help my enthusiasm levels either.
I spent a couple of hours today wrestling with encodings. Writing out the details here to help me remember, and perhaps save someone else from grief.
A friend shared this image on Facebook. I thought it was pretty funny, although perhaps not for the reasons its creator thought it was funny:
Modern dating is full of choices, whether between strangers at a bar or profiles on OkCupid. With so many options, it’s important to be rigorous in your search for a statistically significant other. Here are some tips to help you correlate with others.
There’s a body of research literature showing that people overestimate how well they can explain phenomena. Ask someone whether they understand how a can opener works, for example, and they’re likely to say they do. Ask someoneto explain how a can opener works, though, and you’re likely to get confusion, frustration, and a confession that they don’t understand as well as they thought they did.
When David Bowie died two weeks ago, I felt sad, but it was nothing compared to the outpouring of grief and remembrance from friends and from the culture at large. I have always liked Bowie’s music, his style, his interviews, the characters he’s played in movies. But I never quite got Bowie.
That feeling when someone links you to Leslie Odom Jr singing a mashup of the two best musical theater ballads* Stars and Wait For It and you spend the next two hours systematically watching all the other Ham4Hams that have ever been recorded.
Mystery Hunt was this past weekend and as usual I enjoyed the hell out of it. Highlights included Molly’s epic history rap battle, every puzzle name in the Lovecraft-themed round, and helping to solve a number of metas. The Dreamtime meta-meta was particularly fun, so that’s the puzzle I’ll explain, rot13’d to prevent spoilers. You can skip the gibberish if you just want links to my favorite puzzles.
Inspired by this letter from Thomas Jefferson to James Madison.
I’ve been reading a lot lately, but slowly. My new note-taking system seems to double or even triple how long it takes me to get through a book. Part of me wants to say “screw it all!” and go back to flying through non-fiction like novels but I can already tell how much more I’m retaining now that I’m actually writing stuff down.
Whoops I am falling behind on Iron Blogger. I’ve been busy with a lot of stuff, and have yet to master the art of the casual update, apparently.
I’ve seen a lot of trickster villains recently. Of course, tricksters are nothing new under the sun - they’ve been an archetype for a very long time - but I’ve seen a lot of them lately and I’ve noticed a bit of a pattern: trickster villains tend to be paired with vigilante heroes.
For my birthday this weekend I tried to make a Zoetrope cake. I was inspired by Alexandre Dubosc, whose cakes you should definitely look at after this one, because mine does not benefit from the comparison. I’m putting the gif below a cut, as it may be a bit of a migraine or seizure trigger.
For the record, this recipe is way easier than it looks, and super tasty.
Last year I participated in a novel and exciting meta-analysis project called Many Analysts, One Dataset. A single dataset was given to researchers in over a dozen independent groups to analyze, with the hope of seeing just how divergent the analyses would be.
The players
Last month I quietly launched a new company, Galaxy Rise Consulting. Yesterday, my first client Sumana Harihareswara less quietly launched hernew company, Changeset Consulting. If you need advice, auditing, internship support, release management, sprint prep, or anything like it for your FOSS project, I highly recommend Sumana. And if you need a website, app, or data science project developed, I highly recommend me. ;)
Chiu-Ki and Cate from Technically Speaking have challenged people to make and publish a list of public speaking goals for 2016. My goals for this year are fairly modest, as I’m no longer working with an organization that will pay my way to conferences, but that does lead to an obvious first goal:
Earlier this month I participated in a book sprint to raise awareness of the imprisonment of Bassel Khartabil by the Syrian government. Now, it is rumored he has been sentenced to death.
I did an interview with Django Girls. You can read it here.
Yesterday I backed up my entire computer, deleted and resized some partitions, and then attempted to upgrade my operating system from 32-bit to 64-bit. I was following this guide, which unfortunately doesn’t tell you to follow the instructions as root from the beginning. The result? Halfway through the process, I deleted my 32-bit version of sudo in order to install the 64-bit one.
I missed this paper when it went around last week: The prevalence of statistical reporting errors in psychology (1985–2013). Daniel Lakens has a very good review of it.
Making icons for Android is pretty tedious, until you realize you can manage everything with a shell script:
I recently made myself a new personal website, and I noticed there were a couple of problems with the mobile version: the font-awesome icons weren’t displaying, and the collapsed, mobile-friendly navbar wasn’t expanding.
Note: tumblr ate the initial, more comprehensive version of this post. :( Feel free to ping me for more details.
It’s hard to trick my niece - she wasn’t born yesterday, you know!
A few days ago I was playing Set with friends when one of them asked how many cards could be put down without making any sets. Someone else responded that whatever the answer was, there was probably a very sound mathematical/logical reason for it. After a bit of thinking I came up with this:
The other day, as I was sipping hot chocolate and talking with my college friend Cora about the play Copenhagen*, I had a thought:
It’s been about a year since I’ve blogged regularly. My perfectionist tendencies started getting the best of me - every post needed to be essay-like, completely thought out.
An editorial in Nature by Arthur Ammann points out a number of problems with our current clinical trial system:
This is a little old - I’m going through my bookmarks - but:
From a paper published in 1945 by Vannevar Bush, head of the U.S. Office of Scientific Research and Development (OSRD) during World War II:
The recent announcement that the University of California - the world’s largest public university system - will be embracing open access has been meet with celebration. However, the new policy only encourages faculty to share their work - there are no requirements. Meanwhile, the UC Regents have voted to create a new corporate entity to manage research funding:
From Retraction Watch, a story about a retraction is really a story about Nature’s failure to publish a refutation:
This PLoS One study attempted to look at how pressure to publish might influence the prevalence of positive results. The author, Daniele Fanelli, made an odd choice by using ‘papers per capita by state’ as the measure of pressure to publish. The state seems way too macro a level to look at. I would expect pressure to vary strongly between schools and between departments within schools. There’s no reasoning for this given in the paper - I suspect that it was just simpler to use this measure, already provided by the NSF, then to come up with a method for looking at individual institutions.
This article about Bionimbus, about an effort to efficiently and securely share cancer genomics data, mentions the Open Science Data Cloud. The OSDC is a platform for sharing large datasets, and although their current set of projects, including Bionimbus, are mostly genomics-focused, they seem to be open to datasets of all types.
A colleague referenced this 2010 paper which measures proportion of positive results by scientific discipline.
A quick note on an interesting point raised by Roger Peng:
A paper and associated web-based tool have been proposed to detect publication bias and p-hacking in the literature. From the paper:
I spent three years working in fMRI labs. To this day it’s not clear to me if the field has exceptionally ambiguous standards, or if it’s only one of many scientific subfields based around new technology struggling to define good practice. Whether it’s got company or not, neuroimaging certainly has issues.
There’s an op ed in the New York Times about how a coding error in an economics article may have had profoundly negative influence on economic policy:
There’s a new article out in Nature Reviews Neuroscience about the failure of scientific studies in general (and neuroscience and fMRI studies in particular) to adequately power their studies. The NRN paper isn’t open access, but you can email the authors for a pre-print. There’s a good write-up at National Geographic.
An article in the Washington Post details an alleged case of Fraud which led to an as-yet-uncorrected Nature paper and a suicide:
An opinion article from Iain Chalmers (of the very well regarded though not fully open access Cochrane Collaboration) and Paul Glasziou (of the Center for Evidence-based Medicine) discusses the high potential for waste in much research.
The Obama administration has announced that all research funded by federal agencies (with R&D budgets greater than $100 million) should be made open access within one year of publication:
An interesting article from Accident Analysis and Prevention from 2004 goes over three case studies where Null Hypothesis Significance Testing may have cost lives.
On the PLoS blog, an article from Jack Andraka, winner of the 2012 Intel Science prize for the invention of a dip stick test for pancreatic cancer, on trying to reach behind paywalls as a 14 year old:
Found this interesting paper from a couple years ago: Detecting and Correcting the Lies that Data Tellby F Schmidt.
This post at the Scholarly Kitchen touches on a number of arguments against mandatory CC-BY licensing. They’re all worth talking about, although none seem worth keeping science behind closed doors.
A new PLoS one paper looked at the outcome of 1054 trials submitted to the ethics committee of a major Spanish hospital from 1997-2004. This flowchart captures the basic results:
It’s been my experience that I wildly disagree with some of the fundamental values of the folks at Scholarly Kitchen, but I get a lot out of their posts and arguments. Their most recent post on a possible scandal at Pub Med Central is a case in point:
A relevant follow up to the last post. Science critic Ben Goldacre has a new project: a petition to get all trials registered and to enforce reporting deadlines.
An article in BMJ explores how well registered trials adhere to reporting deadlines.
A new PLoS Medicine report investigated how pharmaceutical promotions and advertising influence prescriptions:
Back in November, Perspectives on Psychological Science put out a special issue on replicability. I’ll be attempting to summarize each of the articles in that issue for this tumblr. See this post for links to all article summaries, or use this tag to browse.
A brief comment points out a major flaw in a published study, in part by using simulations:
A friend forwarded me this paper, “Empirical estimates suggest most published medical research is true”, perhaps in an attempt to challenge my cynicism. I like to believe I am open to being challenged, and I do recognize that purposefully cataloging problems with research leaves me biased. But I don’t think this paper is the best counterpoint.
A post at Retraction Watch summarizes the situation:
A recent paper did an analysis of breast cancer studies published over the last 16 years. They evaluated 164 trials and looked at whether results re: the drug’s toxicity or overall survival rate was reported prominently in the abstract, within the article, or at all. They looked at who funded the work, the impact factor of the journal the work was published in, and most interestingly, whether the trial found positive or negative results.
Back in November, Perspectives on Psychological Science put outa special issue on replicability. I’ll be attempting to summarize each of the articles in that issue for this tumblr. See this post for links to all article summaries, or use this tag to browse.
Perspectives on Psychological Science recently came out with a special issue on replicability in psychology research and its implications for the reliability and health of the field as a whole. They cite as motivation the last two years, which have been filled with fraud, admissions of questionable research practices, a published paper in a major paper claiming evidence of ESP, reports that researchers cannot or will not share data or disclose the full extent of their analyses, and the failure to replicate several major findings in the field.
Back in November, Perspectives on Psychological Science put out a special issue on replicability. I’ll be attempting to summarize each of the articles in that issue for this tumblr. See this post for links to all article summaries, or use this tag to browse.
Science Fraud faces legal threats
This review of Diederik Stapel’s autobiography quotes this interesting passage:
I found this 2004 article through the footnotes of Nate Silver’s book. Here are the highlights (but I recommend reading it all the way through.)
This article raises, for me at least, the interesting question of whether the large amount of research that remains uncited in medical/scientific literature is a good thing, a bad thing, or a neutral thing. I did a little more reading, and found some other posts on the topic.
Scientists Propose Central Database for Disclosing Conflicts of Interest
This article (warning: pdf) talks in detail about how the psychology community frequently fails to account for the great diversity of cultures in our world.
From this post:
Article Level Metadata
This article focuses on making study results available to patients who are providing the medical data. But it touches on more general themes - on the right of patients not just to control their own data but to access the results of research into their illnesses, and to be empowered to do some of the research themselves.
Stapel Report
I normally do not post about sexism, racism, and other forms of bigotry in the scientific community here. While I obviously think they exist and am against them, it is ambiguous how these “isms” impact/bias the scientific literature. (I am sure they do. But perhaps in a less concrete way than, say, the file-drawer phenomenon, or clear misuse of statistics.)
From the copy on their website:
Retractions stigmatize scientific fields, study finds
Quoting my own write up:
Vaughn Bell of Mind Hacks interviews researchers about the problems with fMRI:
Dorothy Bishop critiques a 2003 neuroimaging paper:
From the article:
“A Systematic Review and Meta-Analysis of Survey Data”
(Lots of other interesting non-dismal-sciencey results here.)
A pair of articles in Infection and Immunity. In addition to summarizing problems, they offer the following reforms (all greatly abridged in the excerpts below):
As described in this Nature article:
From the article:
Reproducibility Project
Why Most Published Research Findings Are False
Surely You're Joking... : The Oil Drop Experiment
Contradicted and Initially Stronger Effects in Highly Cited Clinical Research
402 Citations Questioning the Indiscriminate Use of Null Hypothesis Significance Tests in Observational Studies
The decline effect and the scientific method
Walking Fast and Slow
We Must Stop The Avalanche of Low Quality Research
If you want reproducible science, the software needs to be open source.
Solutions: Dryad Data Repository
Journal Disavows Study Touted By Abortion Foes
How Corporations Corrupt Science at the Public's Expense