Follow-up: Neil Trevett and Tom Olson from Khronos Group Discuss OpenCL and Vulkan Roadmap | PC Perspective
this is maybe the biggest current gaping hole in my systems programming knowledge. going to become more important. should digest this more thoroughly as well as the current state of opencl/CUDA
programming  gpu-programming  to-read 
4 days ago
Bad Blood: Secrets and Lies in a Silicon Valley Startup, by John Carreyrou (@Kindle)
Finished 2018-06-10. Recommended. A brisk read, funny and maddening.

A consistent theme here is very successful old men deciding to rely too much on social proof and gut instinct over due diligence, physical evidence, and the advice of more conscientious but lower status people around them.

Also, to be frank, Stanford doesn't come off looking too great, particularly the Hoover Institution, although I guess anyone with a clue already knew that Hoover is a pernicious parasite.
booklog  nonfiction  finished:2018  silicon-valley  biotechnology  venture-capital  stanford  conservatism 
7 days ago
Opinion | Don’t Blame Silicon Valley for Theranos - The New York Times
Theranos did make presentations to many, if not most, of the top life sciences firms. Part of the company’s appeal was the familiar origin myth of Theranos’s founder, Elizabeth Holmes, who, like Bill Gates and Mark Zuckerberg before her, dropped out of college in order to found her company.

That might impress some social media investors, but in life sciences, everyone puts in years of formal study just to earn a seat at the table. For example, at MPM Capital, a venture firm that invests in life sciences, almost every one of its 20 investing directors and partners has either a Ph.D. or M.D., and one has both. Even the general counsel has a Ph.D. in cell, molecular and developmental biology.

GV, formerly Google Ventures, has a five-person investment team for Life Science & Health that includes two members with Ph.D.s in bioengineering; another with both an M.D. and a Ph.D. in biophysics; and a partner who, unlike Ms. Holmes, finished at Stanford, then went on to earn an M.D. and M.B.A. at Harvard.

Theranos approached GV twice and was turned down twice because of what one partner called “so much hand-waving.” People I have talked to at other investment firms said they turned down Theranos for similar reasons, unsatisfied with Theranos’s attempt to substitute its intangible “coolness” in place of technical details needed to validate its diagnostic technology.

Another tipoff? Theranos wouldn’t publish in peer-reviewed journals. Guy Cavet, chief technology officer for the biotech firm Atreca, said: “Every smart prospective partner of a life sciences start-up looks for strong peer-reviewed publications. It’s a way of getting expert due diligence at zero cost.”

Experience in health care is critical for a company like Theranos, which has to comply with government regulations. Instead, even the board of directors was weighted during most of the company’s life with older political figures like George P. Shultz and Henry A. Kissinger.

Luke Evnin, a co-founder at MPM Capital, said he had never met with Theranos or Ms. Holmes, but he found the makeup of the board puzzling: “It is pretty weird that if you look at her board, there’s not a single person who knows what they’re doing in the business.”

The first million dollars that the company received was from Tim Draper, a venture capitalist who became a venture capitalist through a very un-Silicon Valley-like route: His father was one (as was his grandfather). Mr. Draper had known Ms. Holmes as a childhood neighbor and playmate. The investors that followed Mr. Draper are a motley group, at least the ones visible in S.E.C. filings: a tiny firm named ATA Ventures; Continental Properties, a real estate company; and Donald L. Lucas, whose claim to fame was having invested in Oracle Corporation early.

But while Silicon Valley Proper wasn’t interested, the media was. Ms. Holmes was on the covers of Fortune, Forbes, Inc., and T: The New York Times Style Magazine. “The Next Steve Jobs” promised the cover of Inc. Richard Kovacevich, then a board member and a former Wells Fargo C.E.O., crowed, “We didn’t need advertising.”

No, they needed results. Theranos might still prove viable. But if Walgreens ends up with swampland, it’s not Silicon Valley’s fault.

It has been amusing to watch the media, which hyped Theranos far harder than the actual Silicon Valley venture scene, rush to use Theranos as an object lesson in the corruption of Silicon Valley.
silicon-valley  venture-capital  nepotism  journalism 
8 days ago
[1711.01254] Automated Detection, Exploitation, and Elimination of Double-Fetch Bugs using Modern CPU Features
Double-fetch bugs are a special type of race condition, where an unprivileged execution thread is able to change a memory location between the time-of-check and time-of-use of a privileged execution thread. If an unprivileged attacker changes the value at the right time, the privileged operation becomes inconsistent, leading to a change in control flow, and thus an escalation of privileges for the attacker. More severely, such double-fetch bugs can be introduced by the compiler, entirely invisible on the source-code level.
We propose novel techniques to efficiently detect, exploit, and eliminate double-fetch bugs. We demonstrate the first combination of state-of-the-art cache attacks with kernel-fuzzing techniques to allow fully automated identification of double fetches. We demonstrate the first fully automated reliable detection and exploitation of double-fetch bugs, making manual analysis as in previous work superfluous. We show that cache-based triggers outperform state-of-the-art exploitation techniques significantly, leading to an exploitation success rate of up to 97%. Our modified fuzzer automatically detects double fetches and automatically narrows down this candidate set for double-fetch bugs to the exploitable ones. We present the first generic technique based on hardware transactional memory, to eliminate double-fetch bugs in a fully automated and transparent manner. We extend defensive programming techniques by retrofitting arbitrary code with automated double-fetch prevention, both in trusted execution environments as well as in syscalls, with a performance overhead below 1%.
security  program-analysis  research  papers 
10 days ago
Dinosaur Comics - June 4th, 2018 - awesome fun times!
a solid and novel science fictional premise, and here executed with almost borgesian finesse
humor  science-fiction 
12 days ago
Revisiting the Marshmallow Test: A Conceptual Replication Investigating Links Between Early Delay of Gratification and Later Outcomes - Tyler W. Watts, Greg J. Duncan, Haonan Quan, 2018
Entirely predictably, this study is being misreported as completely debunking the original study, whereas (if I understand the abstract), in fact it merely shows an effect size that is smaller:

<< Concentrating on children whose mothers had not completed college, we found that an additional minute waited at age 4 predicted a gain of approximately one tenth of a standard deviation in achievement at age 15. But this bivariate correlation was only half the size of those reported in the original studies and was reduced by two thirds in the presence of controls for family background, early cognitive ability, and the home environment. >>
psychology  child-development  research 
13 days ago
Against trendism: how to defang the social media disinformation complex
I have thought that (5) (hide favorite counts, retweet counts, & follower counts) would improve Twitter considerably for some time now; of course they will never do it.
social-media  social-software  social-formation-of-belief 
13 days ago
Is the use of Facebook turning into a social class indicator? - Femke Goedhart | Tableau Public
maybe FB really is turning into a declasse social network for middle aged people, as I said years ago (& later changed my mind about), rather than becoming the identity layer for all human interaction? but it will be interesting to see what happens as this cohort ages and finds that it has to engage with all the real-world social networks that are stuck on FB through groups etc.
visualization  facebook  social-media  class 
13 days ago
Trade sanctions against America won't work. Sanctioning Trump himself might. - Macleans.ca
accurate. I fully endorse this course of action for foreign countries that want to bring pressure against this President. (Note that he would be immune to these tactics if he had divested on assuming the office.)
international-relations  united-states  canada  trump 
15 days ago
Marketing Technology Landscape Supergraphic (2018): Martech 5000 (actually 6,829) - Chief Marketing Technologist
surely every one of these firms is backed by VCs and led by an executive team that displays the utmost concern for protecting the privacy of individuals
marketing  internet  privacy  lol-what-am-i-saying 
15 days ago
One year of C
I disagree with a lot of this and suspect that this does not scale to large team projects but it advocates a style of C programming that I had not seen before which is interesting.
programming  experience-reports 
15 days ago
Why Doesn't Anyone Answer the Phone Anymore? - The Atlantic
<<in the last couple years, there is a more specific reason for eyeing my phone’s ring warily. Perhaps 80 or even 90 percent of the calls coming into my phone are spam of one kind or another.>>
culture  communication 
16 days ago
Moving Fast and Securing Things – Several People Are Coding
sigh. file under "stuff your startup can do when investors have given you astonishing amounts of resources".
security  software-development 
6 weeks ago
PostgreSQL's fsync() surprise [LWN.net]
when the PostgreSQL community found out that the way the kernel handles I/O errors could result in data being lost without any errors being reported to user space, a fair amount of unhappiness resulted. The problem, which is exacerbated by the way PostgreSQL performs buffered I/O, turns out not to be unique to Linux, and will not be easy to solve even there.
Craig Ringer first reported the problem to the pgsql-hackers mailing list at the end of March. In short, PostgreSQL assumes that a successful call to fsync() indicates that all data written since the last successful call made it safely to persistent storage. But that is not what the kernel actually does. When a buffered I/O write fails due to a hardware-level error, filesystems will respond differently, but that behavior usually includes discarding the data in the affected pages and marking them as being clean. So a read of the blocks that were just written will likely return something other than the data that was written.

What about error status reporting? One year ago, the Linux Filesystem, Storage, and Memory-Management Summit (LSFMM) included a session on error reporting, wherein it was described as "a mess"; errors could easily be lost so that no application would ever see them.

databases  linux  operating-systems 
6 weeks ago
xkcd: Python Environment
sigh, so true, it's gotten to the point where I just try to do Python coding entirely in a Docker container whenever I can
python  programming  humor  or-is-it 
6 weeks ago
Kengo Kuma to build aquatics centre and harbour bath in Copenhagen
I will be very impressed if they manage to execute on this design. Even if they do, it will be jam packed on any nice day and look nothing like this. Still, it's not often I see architecture this striking.
architecture  design 
7 weeks ago
Where Countries Are Tinderboxes and Facebook Is a Match - The New York Times
A reconstruction of Sri Lanka’s descent into violence, based on interviews with officials, victims and ordinary users caught up in online anger, found that Facebook’s newsfeed played a central role in nearly every step from rumor to killing. Facebook officials, they say, ignored repeated warnings of the potential for violence, resisting pressure to hire moderators or establish emergency points of contact.
. . .
Mr. Gunawardana, the public information head, said that with Facebook unresponsive, he used the platform’s reporting tool. He, too, found that nothing happened.

“There needs to be some kind of engagement with countries like Sri Lanka by big companies who look at us only as markets,” he said. “We’re a society, we’re not just a market.”
facebook  social-media  sri-lanka 
7 weeks ago
algorithm - Skip List vs. Binary Search Tree - Stack Overflow
nice, must read the dean & jones paper someday at least.

it is a telling indictment of SO that this awesome answer is not the top-voted one for this question
data-structures  papers  to-read 
10 weeks ago
Confessions of an Unreal Engine 4 Engineering Firefighter
better and more broadly relevant than the title suggests, one small excerpt:

One example of this was a company's project that was taking much longer to develop than it should, and I was called to either speed up development through engineering, or to figure out what the issue was. Immediately after arriving on-site, and before I had been greeted by the person responsible for hiring me, several engineers were telling me to run away as fast as I could because the producer who had complete authority over engineering has no engineering knowledge, and kept mixing up and reassigning tickets to the wrong engineers; network engineers getting gameplay fixes, gameplay engineers getting rendering fixes, etc.. Because of these mix-ups, frequent meetings were held to fix these mix-ups, and if engineers hate one thing, it is frequent meetings. Every one of those engineers had claimed to have brought up this issue with the producer and upper management, and they had even shown me emails of this communication, without any prior checking or confirmation that I had to be given access to such information. They were so exhausted of the situation that they just simply started doing what their tickets said. One of the network engineers spent 3 months learning and working on animation graphs because of this. In this case, engineers found a way to improve themselves but unfortunately without improving the company. Because I was able to discover this before even taking the company tour, I was able to lay this all out in front of upper management immediately. They immediately asked who was giving up this information and wanted to crack down on this perceived insubordination, rather than trying to address the issue. I was then dismissed for not specifying names, and my services were no longer required. A month later, their engineering team went from 7 engineers to 2 as they were abandoning ship. A month afterward, I got a call saying they no longer had an engineering team and asked my rates to finish the engineering effort. Their offer was nowhere near the needed budget to hit the required engineering milestones. They ended up hiring a whole new team of lesser skilled engineers, and the process repeated itself.
software-development  management  business  labor  via:hackernews 
10 weeks ago
Conflict-free Replicated Data Types
Replicating data under Eventual Consistency (EC) allows any replica to accept updates without remote synchronisation. This ensures performance and scalability in large-scale distributed systems (e.g., clouds). However, published EC approaches are ad-hoc and error-prone. Under a formal Strong Eventual Consistency (SEC) model, we study sufficient conditions for convergence. A data type that satisfies these conditions is called a Conflict-free Replicated Data Type (CRDT). Replicas of any CRDT are guaranteed to converge in a self-stabilising manner, despite any number of failures. This paper formalises two popular approaches (state- and operation-based) and their relevant sufficient conditions. We study a number of useful CRDTs, such as sets with clean semantics, supporting both add and remove operations, and consider in depth the more complex Graph data type. CRDT types can be composed to develop large-scale distributed applications, and have interesting theoretical properties.
papers  data-structures  algorithms  concurrency  to-read 
11 weeks ago
A comprehensive study of Convergent and Commutative Replicated Data Types
Eventual consistency aims to ensure that replicas of some mutable shared object converge without foreground synchronisation. Previous approaches to eventual consistency are ad-hoc and error-prone. We study a principled approach: to base the design of shared data types on some simple formal conditions that are sufficient to guarantee eventual consistency. We call these types Convergent or Commutative Replicated Data Types (CRDTs). This paper formalises asynchronous object replication, either state based or operation based, and provides a sufficient condition appropriate for each case. It describes several useful CRDTs, including container data types supporting both \add and \remove operations with clean semantics, and more complex types such as graphs, montonic DAGs, and sequences. It discusses some properties needed to implement non-trivial CRDTs.
data-structures  concurrency  algorithms  papers  to-read 
11 weeks ago
Calvin Walton on Twitter: "Fun fact: @rustlang nightly can bootstrap on an i586 box. Unfun fact: Build completed successfully in 3 days, 0:56:30 500MHz K6-2+ w/ 512MB ram is not recommended build environment."
From time to time I wonder why it took the global discipline of computer science so long to create a safe systems programming language. It is possible that neither the theory *nor* the hardware was up to the task until the 2010s.
rust  programming-languages  programming-systems  performance 
11 weeks ago
Information flow reveals prediction limits in online social activity | the morning paper
<<…we estimate that approximately 95% of the potential predictive accuracy attainable for an individual is available within the social ties of that individual only, without requiring the individual’s data.>>
facebook  social-networks  privacy  papers 
11 weeks ago
GitHub - cupslab/password_meter
This project implements a data-driven password meter. Its effects on password security and usability were evaluated in the following publication: http://www.blaseur.com/papers/CHI17meter.pdf and a demo is available at: https://cups.cs.cmu.edu/meter/
papers  security  to-read 
12 weeks ago
The Intellectual We Deserve | Current Affairs
Until I read this, I had remained about as ignorant about Peterson as I was about any other random B-list YouTube celebrity, which is to say that I had a vague idea about his brand (a professor giving life advice to conservative-leaning young people), but I had no idea what he stood for or whatever. I had naively assumed that he was a more media-savvy and self-promoting version of David Gelernter, which is to say an academic technician, accomplished in his field, who had found a popular audience by dressing up his extra-academic reactionary gestures in a layer of pop erudition. I had forgotten to account for the fact that conservative media is literally a kakistocracy, and nobody so benign as that could possibly rise to the heights that Peterson has. In fact, this article makes plain that Peterson is a poisonous bullshit artist of the highest order, not even a halfway competent practitioner of philosophy; of course his utter incompetence at reasoning or communicating clearly is entirely necessary to his appeal. It is telling that even the glib blowhard Sam Harris cannot sanction this man's buffoonery, yet Peterson's star has risen far above those of Harris and Dawkins and their ilk; they are constrained by at least a passing respect for reason and a desire to be lucid, whereas Peterson has completely liberated himself of such limitations and can thus craft his message purely for psychological impact on an undiscerning audience.

Contests I would like to see:

+ a debate between Peterson and Deepak Chopra to see whose eternal fountain of woo-woo can spurt the highest and longest.

+ a debate between Peterson and Stephen Wolfram to see who has the more grandiose ego (Wolfram probably wins this one; to be fair Wolfram is actually an extremely proficient mathematician and hacker so he has a more justifiable reason for his ego, which isn't to say that it is entirely justifiable).
conservatism  media  academia  youtube  via:cshalizi 
march 2018
[1803.01307] Angora: Efficient Fuzzing by Principled Search
Fuzzing is a popular technique for finding software bugs. However, the performance of the state-of-the-art fuzzers leaves a lot to be desired. Fuzzers based on symbolic execution produce quality inputs but run slow, while fuzzers based on random mutation run fast but have difficulty producing quality inputs. We propose Angora, a new mutation-based fuzzer that outperforms the state-of-the-art fuzzers by a wide margin. The main goal of Angora is to increase branch coverage by solving path constraints without symbolic execution. To solve path constraints efficiently, we introduce several key techniques: scalable byte-level taint tracking, context-sensitive branch count, search based on gradient descent, and input length exploration. On the LAVA-M data set, Angora found almost all the injected bugs, found more bugs than any other fuzzer that we compared with, and found eight times as many bugs as the second-best fuzzer in the program who. Angora also found 103 bugs that the LAVA authors injected but could not trigger. We also tested Angora on eight popular, mature open source programs. Angora found 6, 52, 29, 40 and 48 new bugs in file, jhead, nm, objdump and size, respectively. We measured the coverage of Angora and evaluated how its key techniques contribute to its impressive performance.
security  tools  programming  program-analysis 
march 2018
Land value taxation, Prop 13 reform and single family homeowners – Lisa Schweitzer
your yearly tax bill could basically stay the same for the lifetime of your tenure in the house and the accumulated value of the tax, or some portion, gets extracted at point of sale. That is, when you move and sell, the state captures the public increment out of the sale price. (Ouch, still, but you aren’t being taxed out of anything; if the home value wealth is “paper” wealth until it is made liquid, then it’s merely a “paper” loss to hand over that increment at the end. We’ll see the double-standard in play as people react to that statement).

+1, yes please, can we have this now
housing  california  taxation  georgism 
march 2018
[1802.07068] Talent vs Luck: the role of randomness in success and failure
The largely dominant meritocratic paradigm of highly competitive Western cultures is rooted on the belief that success is due mainly, if not exclusively, to personal qualities such as talent, intelligence, skills, efforts or risk taking. Sometimes, we are willing to admit that a certain degree of luck could also play a role in achieving significant material success. But, as a matter of fact, it is rather common to underestimate the importance of external forces in individual successful stories. It is very well known that intelligence or talent exhibit a Gaussian distribution among the population, whereas the distribution of wealth - considered a proxy of success - follows typically a power law (Pareto law). Such a discrepancy between a Normal distribution of inputs, with a typical scale, and the scale invariant distribution of outputs, suggests that some hidden ingredient is at work behind the scenes. In this paper, with the help of a very simple agent-based model, we suggest that such an ingredient is just randomness. In particular, we show that, if it is true that some degree of talent is necessary to be successful in life, almost never the most talented people reach the highest peaks of success, being overtaken by mediocre but sensibly luckier individuals. As to our knowledge, this counterintuitive result - although implicitly suggested between the lines in a vast literature - is quantified here for the first time. It sheds new light on the effectiveness of assessing merit on the basis of the reached level of success and underlines the risks of distributing excessive honors or resources to people who, at the end of the day, could have been simply luckier than others. With the help of this model, several policy hypotheses are also addressed and compared to show the most efficient strategies for public funding of research in order to improve meritocracy, diversity and innovation.

I believe this just tries to model mathematically something that I thought most people already felt intuitively, although maybe I'm giving people too much credit.

Also, if the distribution is log-normal rather than power law (& it is reasonably hard to distinguish between them without careful statistical tests), then wouldn't a Gaussian distribution of intelligence, talent, & other factors simply imply a multiplicative effect among them, rather than the existence of a hidden variable?
economics  meritocracy-and-its-discontents  statistics 
march 2018
« earlier      
academia advice algorithms 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-science computer-vision computing concurrency confederate-states-of-america conservatism corruption crime cryptocurrency cryptography culture data-structures databases demographics design devops digital-art distributed-systems docker economic-inequality economics education edx environment 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 free-as-in-beer free-as-in-speech funny futurism game-theory games genetics google government graphics hacks hardware health higher-education 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 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 programming programming-languages propaganda psychology public-policy publishing python racism real-estate redecentralize religion rent-seeking republicans research rhetoric rust san-francisco science science-fiction security security-state sexism sexual-equality sexual-inequality shopping silicon-valley social-engineering social-formation-of-belief social-inequality social-networks social-organization social-science social-software software software-architecture software-development startups statistics strange-loop surveillance-state talks technological-progress technology technology-industry to-blog to-read to-read-maybe to-watch todo tools transit trump tweet-threads-that-should-be-blog-posts twitter uber united-states urbanism user-interface via:cshalizi via:hackernews via:marginalrevolution via:metafilter via:twitter video videos visualization war web-development welfare white-people

Copy this bookmark: