3399
Moving the A2J Conversation to a New Level | NSRLP
<< “It’s really time to admit that we have allowed tremendously complex legal processes to develop that exploit the fact that the vast majority of people cannot manage tremendously complex legal processes.” >>
coevolution-of-rulesets-and-their-adversaries  law  justice 
13 days ago
A cleansing fire: Moral outrage alleviates guilt and buffers threats to one’s moral identity (PDF Download Available)
<<Why do people express moral outrage? While this sentiment often stems from a perceived violation of some moral principle, we test the counter-intuitive possibility that moral outrage at third-party transgressions is sometimes a means of reducing guilt over one’s own moral failings and restoring a moral identity.>>

ha ha! no shit sherlock
moral-panics  psychology  social-media 
17 days ago
WeWork: The Perfect Manifestation of the Millennial Id - The Atlantic
Terrible, awful headline, but some interesting discussion of WeWork's business model & risks thereof.
real-estate  business 
17 days ago
Google Cloud Platform Blog: 12 best practices for user account, authorization and password management
not really specific to services built on GCP, but worth reading what Google is recommending to developers
security  identity  software-as-a-service  google 
21 days ago
you’re not alone in everything✨ Saw TLJ x4 on Twitter: "Rey & Kylo Ren fighting the Praetorian Guards to _____ 🎈a thread🎈"
The fact that so many of these work so well is a testament to how well paced, sharply edited, and kinetic this action sequence was. The director of TLJ understands the basic fact that fight sequences are a form of dance, & a great fight sequence must nearly always borrow from the cinematic language of Hollywood dance numbers. This seems like such an elementary point --- isn't the pleasure of a fight sequence ultimately just the pleasure of watching human bodies in motion? --- yet most action directors completely fail to grasp it.

I would lay money that you could not do this (well) with any action sequence from the prequels. N.b. that I hated Phantom Menace so much that I have refused even to watch the other two, which hasn't kept me from seeing bits and pieces on the screens of neighboring seats on airplanes and the like.
music  film  dance  star-wars 
27 days ago
Conflict Vs. Mistake | Slate Star Codex
SSC has its justified detractors, & I have many, many beefs with things that Alexander has written, but this is actually a good and sympathetic treatment of a difference in people's intellectual/emotional temperaments that is rarely spelled out (& yet you almost surely recognize both sides). If you cannot see the merits of both ways of looking at the world then you are missing a big instrument in your intellectual toolkit.
psychology  rhetoric  technocracy-and-its-discontents  social-justice 
28 days ago
Why I left Google to join Grab – Steve Yegge – Medium
Yegge-esque, although lacking a certain something, probably due to its sales-pitchiness. I will excerpt just one joke that made me laugh out loud:

<<
I am tempted to make fun of Jeff Bezos here, but I’ve heard that nobody has done that three times in a row and lived to tell the tale, so I’ll, ah, quit while I’m ahead.
>>
google  steve-yegge  rants 
29 days ago
How Do Individuals Repay Their Debt? The Balance-Matching Heuristic
<<
We study how individuals repay their debt using linked data on multiple credit cards from five major issuers. We find that individuals do not allocate repayments to the higher interest rate card, which would minimize the cost of borrowing. Instead, individuals allocate repayments using a balance-matching heuristic under which the share of repayments on each card is matched to the share of balances on each card. We show that balance matching captures more than half of the predictable variation in repayments, performs substantially better than other models, and is highly persistent within individuals over time. Consistent with these findings, we show that machine learning algorithms attribute the greatest variable importance to balances and the least variable importance to interest rates in predicting repayment behavior.
>>

Argh! Horrible.
finance  psychology 
29 days ago
A Conversation about Teaching Software Engineering – Embedded in Academia
a great list of stuff that budding software engineers should be taught is important
software-development  education 
4 weeks ago
Latacora | Security Teams For Startups
<<
Latacora does just one kind of engagement: we join your engineering team virtually and run security, for about a year. Then we help you hire someone full-time to replace us.
>>

this is a brilliant idea, they should become gigantic so that every post-series-A startup can be their client
security  engineering  management  startups 
4 weeks ago
[1801.02774] Adversarial Spheres
<<
State of the art computer vision models have been shown to be vulnerable to small adversarial perturbations of the input. In other words, most images in the data distribution are both correctly classified by the model and are very close to a visually similar misclassified image. Despite substantial research interest, the cause of the phenomenon is still poorly understood and remains unsolved. We hypothesize that this counter intuitive behavior is a naturally occurring result of the high dimensional geometry of the data manifold. As a first step towards exploring this hypothesis, we study a simple synthetic dataset of classifying between two concentric high dimensional spheres. For this dataset we show a fundamental tradeoff between the amount of test error and the average distance to nearest error. In particular, we prove that any model which misclassifies a small constant fraction of a sphere will be vulnerable to adversarial perturbations of size O(1/d−−√). Surprisingly, when we train several different architectures on this dataset, all of their error sets naturally approach this theoretical bound. As a result of the theory, the vulnerability of neural networks to small adversarial perturbations is a logical consequence of the amount of test error observed. We hope that our theoretical analysis of this very simple case will point the way forward to explore how the geometry of complex real-world data sets leads to adversarial examples.
>>
computer-vision  security  via:cshalizi 
4 weeks ago
Commentary: Trump's Harmful Stereotyping of Korean Americans | Fortune
I actually missed this in the general avalanche of outrage. I still laugh inwardly at all the Republicans who have been saying for years that Asian- Americans are naturally a Republican consistency.
trump  korea  asian-american-issues 
4 weeks ago
Smaller crowds outperform larger crowds and individuals in realistic task conditions. Galesic, Mirta,Barkoczi, Daniel,Katsikopoulos, Konstantinos Decision, Vol 5(1), Jan 2018, 1-15
<<
Decisions about political, economic, legal, and health issues are often made by simple majority voting in groups that rarely exceed 30–40 members and are typically much smaller. Given that wisdom is usually attributed to large crowds, shouldn’t committees be larger? In many real-life situations, expert groups encounter a number of different tasks. Most are easy, with average individual accuracy being above chance, but some are surprisingly difficult, with most group members being wrong. Examples are elections with surprising outcomes, sudden turns in financial trends, or tricky knowledge questions. Most of the time, groups cannot predict in advance whether the next task will be easy or difficult. We show that under these circumstances moderately sized groups, whose members are selected randomly from a larger crowd, can achieve higher average accuracy across all tasks than either larger groups or individuals. This happens because an increase in group size can lead to a decrease in group accuracy for difficult tasks that is larger than the corresponding increase in accuracy for easy tasks. We derive this nonmonotonic relationship between group size and accuracy from the Condorcet jury theorem and use simulations and further analyses to show that it holds under a variety of assumptions. We further show that situations favoring moderately sized groups occur in a variety of real-life situations including political, medical, and financial decisions and general knowledge tests. These results have implications for the design of decision-making bodies at all levels of policy. (PsycINFO Database Record (c) 2018 APA, all rights reserved)
>>
cognition  social-organization  via:marginalrevolution 
4 weeks ago
Jason M. Lemkin 🦄 on Twitter: "What I learned from 5 weeks in Beijing + Shanghai: - startup creation + velocity dwarfs anything in SF - no one in China I met is remotely worried about U.S. or possibly even cares - access to capital is crazy - scale fee
Just an anecdotal tweet but IMO this kind of sentiment is widely corroborated by China visitors from the US computing industry. Anybody else remember about a decade ago when Thomas Friedman was constantly pronouncing that the US was going to stay ahead of China by being innovative to a degree which, for some mysterious reason *cough* *old white mustached dude racism* *cough* the Chinese would never match?

Note also reply down the thread:
<<
@ryanwaggoner
How can small-time developers / founders / investors in the US capitalize on what’s happening in China?

@swyx
1. go there
2. see what works
3. come back
4. clone what’s missing

repeat until hit
>>

This is literally how businesses in East Asian developing economies (Japan, then Korea/Taiwan, then China itself) approached catch-up growth to the West.
twitter  gossip  china  silicon-valley 
5 weeks ago
What We Talk About When We Talk About Computation
via a former colleage.

i'm honestly somewhat puzzled by Krishnaswamy's position here. you can obviously implement a turing machine in the lambda calculus, and therefore any computation that is possible on turing machines can be encoded (inefficiently) in the lambda calculus; therefore parallel or can be implemented in the lambda calculus. the CCT says nothing about efficiency of the computational encoding; it is a statement about computability, not complexity. thus the CCT is true for the "counterexample" he gives. yes, it's true that if you simply disallow the encoding that i describe, then the lambda calculus cannot emulate por, but i could similarly "disprove" that *any* computational model is universal by disallowing encodings.
computation  formal-semantics  programming-languages 
5 weeks ago
Book-Smart, Not Street-Smart: Blockchain-Based Smart Contracts and The Social Workings of Law | Levy | Engaging Science, Technology, and Society
<<
This paper critiques blockchain-based “smart contracts,” which aim to automatically and securely execute obligations without reliance on a centralized enforcement authority. Though smart contracts do have some features that might serve the goals of social justice and fairness, I suggest that they are based on a thin conception of what law does, and how it does it. Smart contracts focus on the technical form of contract to the exclusion of the social contexts within which contracts operate, and the complex ways in which people use them. In the real world, contractual obligations are enforced through all kinds of social mechanisms other than formal adjudication—and contracts serve many functions that are not explicitly legal in nature, or even designed to be formally enforced. I describe three categories of contracting practices in which people engage (the inclusion of facially unenforceable terms, the inclusion of purposefully underspecified terms, and willful nonenforcement of enforceable terms) to illustrate how contracts actually “work.” The technology of smart contracts neglects the fact that people use contracts as social resources to manage their relations. The inflexibility that they introduce, by design, might short-circuit a number of social uses to which law is routinely put. Therefore, I suggest that attention to the social and relational contexts of contracting are essential considerations for the discussion, development, and deployment of smart contracts.
>>

duh. OTOH there are probably things that "smart contracts" are good for; they are just not a superset of the things that ordinary contracts are good for.
blockchains  law  papers 
5 weeks ago
New Bay Area housing trend: Living in the backyard
This is a totally fine part of the solution to the housing problem. The Bay Area has a gigantic amount of single-family zoned land area that isn't going to be transformed into real apartment buildings overnight. We need a lot more liberalization of this type of development. On the other hand, the usual suspects are pushing back:

<<
In-law units can be an important tool for cities to address their housing shortages, said Jason Rhine, a lobbyist for the League of California Cities, which opposed Wieckowski’s 2016 pro-in-law unit bill. “Where we become concerned,” he said, “is when the state is trying to micromanage where and how.”

If the state prevents local governments from collecting fees, Rhine asked, how will a city pay for the services those new residents need?

The League urged Gov. Jerry Brown to veto Wieckowski’s first in-law unit bill, arguing that easing restrictions could lead to “impaired neighborhood character,” increase competition for parking spots, and threaten the privacy of existing homeowners.
>>

Literal NIMBYism. "We support in-law units, just not in this spot, and not in this way."
real-estate  housing  bay-area  economics  public-policy 
5 weeks ago
2018 AMS Short Course on Discrete Differential Geometry
<<
This page contains supplemental information from the AMS Short Course on Discrete Differential Geometry (DDG) held from January 8–9, 2018 in San Diego, CA. Course speakers provided an introduction to the emerging field of discrete differential geometry, which studies discrete analogs of smooth geometric objects, and provides essential links between analytical descriptions of geometry and computational algorithms. Course participants came from a variety of areas across pure & applied mathematics, ranging from undergraduate to PhD to junior and senior faculty (about 87 in all). Many thanks to our participants for making it a great event!
>>
math  computational-geometry  online-courses 
5 weeks ago
@ErrataRob on the Uber "bug bounty"
1/ As much as I defend Uber, this NYTimes story is wrong: ...
2/ The "vuln" was leaving login credentials exposed on GitHub, which the person used to steal customer account information. This is worth only their max bounty of $10k.
3/ Using stolen credentials to hack in and steal customer account information, then demand money before deleting such information is indeed "extortion".
uber  security  bug-bounty-programs 
5 weeks ago
[1712.09665] Adversarial Patch
<<
We present a method to create universal, robust, targeted adversarial image patches in the real world. The patches are universal because they can be used to attack any scene, robust because they work under a wide variety of transformations, and targeted because they can cause a classifier to output any target class. These adversarial patches can be printed, added to any scene, photographed, and presented to image classifiers; even when the patches are small, they cause the classifiers to ignore the other items in the scene and report a chosen target class.
>>
computer-vision  machine-learning  security 
5 weeks ago
What Self-Awareness Really Is (and How to Cultivate It)
<<
We’ve found that even though most people believe they are self-aware, self-awareness is a truly rare quality: We estimate that only 10%–15% of the people we studied actually fit the criteria.
...
Contrary to popular belief, studies have shown that people do not always learn from experience, that expertise does not help people root out false information, and that seeing ourselves as highly experienced can keep us from doing our homework, seeking disconfirming evidence, and questioning our assumptions.
...
the more power a leader holds, the more likely they are to overestimate their skills and abilities. One study of more than 3,600 leaders across a variety of roles and industries found that, relative to lower-level leaders, higher-level leaders more significantly overvalued their skills (compared with others’ perceptions). In fact, this pattern existed for 19 out of the 20 competencies the researchers measured, including emotional self-awareness, accurate self-assessment, empathy, trustworthiness, and leadership performance
>>
psychology  business  management  possible-bullshit 
6 weeks ago
Free Food for Millionaires, by Min Jin Lee (@Kindle)
Finished 2018-01-05. Recommended. Remarkably rich, thematically and characterologically, for a first novel. On the other hand, I found it a little longer and more meandering than I wanted it to be, and your enjoyment may be modulated by your tolerance for detailed portraiture of the manners of upper-middle-class people in the orbits of Ivy League alumni networks and New York finance.

I'm a Korean-American from the New York area myself, and via my family and acquaintances thereof from my youth I have had some very attenuated contact with the world that Lee describes here. It wasn't much, but even that small dose ultimately filled me with disgust and resentment, and I'm pretty glad that I bailed out and went West to become a Bay Area computer geek. By all rights, I should find this novel annoying as fuck, just because of what it reminds me of. But I enjoyed it! So probably you will too.

One problem with writing novels set in contemporary America is that Americans have shed much of the social constraint that provides the potential energy for novels of manners set in past times or foreign cultures. In the 1990s Korean-American community in New York, Lee locates a subculture that is as repressed and status-obsessed as the English were a century earlier, and mines this fertile ground, obtaining in the process a surprising and unique window into the American psyche more generally (for the manners and attitudes of class that Lee describes are hardly limited to Korean-Americans).

Also Lee writes with exceptional compassion for every single one of her characters, even the choicest douchebags of New York banking, and I think you have to read the novel to believe it.
booklog  finished:2018  fiction 
6 weeks ago
A Browser You’ve Never Heard of Is Dethroning Google in Asia - WSJ
Tell me again how Chrome is verging on monopoly?

<<
UC Browser, which has more than 430 million users world-wide, accounted for 51% of India’s mobile browser market over the past year, compared with 30% for Chrome, according to web analytics firm StatCounter. In Indonesia, UC Browser led Chrome by 41% to 32% during the period.

Chrome has more than a billion users world-wide, according to Google, and it has some 47% global market share versus UC Browser’s 16%, according to StatCounter. In the U.S., Chrome has 39% market share, behind Apple Inc.’s Safari browser, which has 52%.
>>
web-browsers  asia  internet 
7 weeks ago
An epic treatise on scheduling, bug tracking, and triage
Bookmarks are not endorsements.

The main issue I have with this article is its claim that stories and bugs are fairly constant size on average and thus estimable with low variance. I don't know if people are just working on much more straightforward stuff than I am used to, or if my teams have just been really bad at project management, but in my experience it is not uncommon to have 5x variance in work items relative even to fairly careful estimates, and furthermore in any system of sufficient size and sophistication there some simple sounding user stories that take 10x or 20x longer than others because apparent simplicity for the user may cut against the grain of architectural assumptions or constraints in the implementation.
software-development  project-management 
7 weeks ago
The New Education: How to Revolutionize the University to Prepare Students for a World In Flux, by Cathy N. Davidson (@Kindle)
Finished 2017-12-30. OK, I suppose. Many suggestions overlap with those of other education reformers but this is a good overview of some of the major issues with American higher education. Most of her recommendations are pretty remote from typical practice and sadly I do not have high hopes that they will be realized in the near future across large swathes of the higher education landscape.
booklog  finished:2017  education  higher-education 
7 weeks ago
Solid
<<
Solid (derived from "social linked data") is a proposed set of conventions and tools for building decentralized social applications based on Linked Data principles. Solid is modular and extensible and it relies as much as possible on existing W3C standards and protocols.
>>
redecentralize 
7 weeks ago
Pachinko, by Min Jin Lee (@Kindle)
Finished 2017-12-23. Highly recommended. This is a deeply moving, meticulously crafted family epic set in the 20th century in Korea and Japan, written in the realist tradition of 19th century European novelists, but oriented in a fashion that foregrounds a 21st century sensibility and concerns including women's rights, immigrants' rights, and the experience of marginalized people. Also a page-turner without resorting to cheap cliffhangers or other excessively mechanical plottiness! Nicely done.

It is hard to separate my reaction to this from the ongoing hype (it is making numerous best-of-the-year lists, and indeed that is why I finally made the leap and read it). Do I like this so much because it is very much the kind of fiction that fits the current cultural moment, or is it as enduringly good as I perceive it to be? Regardless, it seems excellent, maybe good enough to place Lee in the running for the best living American novelist; at least, it is hard to think of someone who is clearly better.
booklog  finished:2017  fiction  historical-fiction  korea  japan  immigration  globalization  feminism  religion 
8 weeks ago
Silicon Valley Is Turning Into Its Own Worst Fear
<<
I used to find it odd that these hypothetical AIs were supposed to be smart enough to solve problems that no human could, yet they were incapable of doing something most every adult has done: taking a step back and asking whether their current course of action is really a good idea. Then I realized that we are already surrounded by machines that demonstrate a complete lack of insight, we just call them corporations. Corporations don’t operate autonomously, of course, and the humans in charge of them are presumably capable of insight, but capitalism doesn’t reward them for using it. On the contrary, capitalism actively erodes this capacity in people by demanding that they replace their own judgment of what “good” means with “whatever the market decides.”
>>

as the tweet says, corporations: the original paperclip maximizer
singularity  cognition  capitalism 
9 weeks ago
MIT 4.605x - Week 12 - Lecture 24.2 - Intentional Ruin & The Ruin as Monument
Contains hilarious quote about reconstruction of Dresden Frauenkirche: "CATIA was used to rebuild the building, because it's Baroque, and has a lot of curves, and no one knows how to make curves anymore except by computers, because we're, sort of, half illiterate. The fact that you could do it in the 18th century and can't do it now seems to strike no one as particularly odd."
mit-4.605x  edx  online-courses  architecture  history 
9 weeks ago
MIT 4.605x - Week 12 - Lecture 23.7 - Commodity Culture & Its Cost
in which Mark changes the way you will feel about drinking your mocha latte
mit-4.605x  edx  online-courses  architecture  history 
9 weeks ago
« earlier      
academia advice america apple architecture art artificial-intelligence asian-american-issues bay-area biology blog-posts blogs book-reviews booklog books bullshit business calendars california capitalism career-advice catosphere cats child-development china civil-liberties climate-change coding-boot-camps comics comics-strips computational-geometry computer-graphics computer-science computer-vision computing confederate-states-of-america conservatism corruption crime cryptocurrency cryptography css culture data-structures databases demographics design devops digital-art distributed-systems docker ebooks economic-inequality economics education edx environment evolution exploits facebook fantasy-fiction feminism fiction film finance finished:2006 finished:2007 finished:2009 finished:2011 finished:2012 finished:2013 finished:2014 finished:2016 finished:2017 fonts food foreign-policy free-as-in-beer free-as-in-speech free-software funny futurism game-design game-theory games genetics git google government graphics hacks hardware health higher-education hillary-clinton hiring history housing humor intellectual-property internet interviews javascript journalism korea labor language law law-enforcement leftism libertarianism linux luggage machine-learning management maps math media microsoft military mit-4.605x mobile-computing music music-recommendations musiclog new-york-city nonfiction nutrition online-courses operating-systems papers people performance philosophy photography photos plutocracy police politics poverty privacy productivity programming programming-languages propaganda protocols psychology public-policy publishing python racism real-estate redecentralize religion rent-seeking republicans research rhetoric rust san-francisco science science-fiction security security-state security-theater sexism sexual-equality sexual-inequality shopping silicon-valley social-engineering social-formation-of-belief social-inequality social-networks social-organization social-science social-software sociopathy software software-architecture software-development startups statistics strange-loop surveillance-state talks technological-progress technology technology-industry to-blog to-buy to-read to-read-maybe to-watch todo tools transit trump tweet-threads-that-should-be-blog-posts twitter uber united-states unix urbanism user-interface via:cshalizi via:hackernews via:marginalrevolution via:metafilter via:reddit via:twitter video videos visualization war web-development welfare white-people writing

Copy this bookmark:



description:


tags: