Menu
"numbers" entries
Jeff Jonas, Siri and Mesos, YouTube's Bandwidth Bill, and AWS Numbers
-
Decoding Jeff Jonas (National Geographic) — “He thinks in three—no, four dimensions,” Nathan says. “He has a data warehouse in his head.” And that’s where the work takes place—in his head. Not on paper. Not on a computer. He resorts to paper only to work the details out. When asked about his thought process, Jonas reaches for words, then says: “It’s like a Rubik’s Cube. It all clicks into place. “The solution,” he says, is “simply there to find.” Jeff’s a genius and has his own language for explaining what he does. This quote goes a long way to explaining it.
-
How Apple Uses Mesos for Siri — great to see not only some details of the tooling that Apple built, but also their acknowledgement of the open source foundations and ongoing engagement with those open source communities. There have been times in the past when Apple felt like a parasite on the commons rather than a participant.
-
Cheaper Bandwidth or Bust: How Google Saved YouTube (ArsTechnica) — Remember YouTube’s $2 million-a-month bandwidth bill before the Google acquisition? While it wasn’t an overnight transition, apply Google’s data center expertise, and this cost drops to about $666,000 a month.
-
AWS Business Numbers — Amazon Web Services generated $5.2 billion over the past four quarters, and almost $700 million in operating income. During the first quarter of 2015, AWS sales reached $1.6 billion, up 49% year-over-year, and roughly 7% of Amazon’s overall sales.
Open Data, Packet Dumping, GPU Deep Learning, and Genetic Approval
-
Wiki New Zealand — open data site, and check out the chart builder behind the scenes for importing the data. It’s magic.
-
stenographer (Google) — open source packet dumper for capturing data during intrusions.
-
Which GPU for Deep Learning? — a lot of numbers. Overall, I think memory size is overrated. You can nicely gain some speedups if you have very large memory, but these speedups are rather small. I would say that GPU clusters are nice to have, but that they cause more overhead than the accelerate progress; a single 12GB GPU will last you for 3-6 years; a 6GB GPU is plenty for now; a 4GB GPU is good but might be limiting on some problems; and a 3GB GPU will be fine for most research that looks into new architectures.
-
23andMe Wins FDA Approval for First Genetic Test — as they re-enter the market after FDA power play around approval (yes, I know: one company’s power play is another company’s flouting of safeguards designed to protect a vulnerable public).
Security Videos, Network Simulation, UX Book, and Profit in Perspective
-
ShmooCon 2015 Videos — videos to security talks from ShmooCon 2015.
-
Comcast (Github) — Comcast is a tool designed to simulate common network problems like latency, bandwidth restrictions, and dropped/reordered/corrupted packets. On BSD-derived systems such as OSX, we use tools like ipfw and pfctl to inject failure. On Linux, we use iptables and tc. Comcast is merely a thin wrapper around these controls.
-
The UX Reader — This ebook is a collection of the most popular articles from our [MailChimp] UX Newsletter, along with some exclusive content.
-
Bad Assumptions — Apple lost more money to currency fluctuations than Google makes in a quarter.
Mousey PC, Sad G+, Medium Data, and Upgraded DARPA Contest Robot
-
PC in a Mouse — 80s = PC in a keyboard. 90s = PC in a box. 2000s = PC in the screen. 2015 we get PC in a mouse. By 2020 will circuitry be inline in the cable or connector?
-
Estimating G+ Usage (BoingBoing) — of 2.2B profiles, 6.6M have made new public posts in 2015. Yeesh.
-
Medium Data — too big for one machine, but barely worth the overhead of high-volume data processing.
-
New Hardware for the DARPA Robotics Challenge Finals (IEEE) — in the future, we’ll all have a 3.7 kwh battery and a wireless router in our heads.
Click to Captcha, Managing Hackers, Easy Ordering, and Inside Ad Auctions
-
One Click Captcha (Wired) — Google’s new Captcha tech is just a checkbox: “I am not a robot”. Instead of depending upon the traditional distorted word test, Google’s “reCaptcha” examines cues every user unwittingly provides: IP addresses and cookies provide evidence that the user is the same friendly human Google remembers from elsewhere on the Web. And Shet says even the tiny movements a user’s mouse makes as it hovers and approaches a checkbox can help reveal an automated bot.
-
The Responsive Enterprise: Embracing the Hacker Way (ACM) — Letting developers wander around without clear goals in the vastness of the software universe of all computable functions is one of the major reasons why projects fail, not because of lack of process or planning. I like all of this, although at times it can be a little like what I imagine it would be like if Cory Doctorow wrote a management textbook. (via Greg Linden)
-
Pizza Hut Tests Ordering via Eye-Tracking — The digital menu shows diners a canvas of 20 toppings and builds their pizza, from one of 4,896 combinations, based on which toppings they looked at longest.
-
How Browsers Get to Know You in Milliseconds (Andy Oram) — breaks down info exchange, data exchange, timing, even business relationships for ad auctions. Augment understanding of the user from third-party data (10 milliseconds). These third parties are the companies that accumulate information about our purchasing habits. The time allowed for them to return data is so short that they often can’t spare time for network transmission, and instead co-locate at the AppNexus server site. In fact, according to Magnusson, the founders of AppNexus created a cloud server before opening their exchange.
Metastable Failures, Static Python Analysis, Material Desktop, and AWS Scale Numbers
-
Metastable Failure State (Facebook) — very nice story about working together to discover the cause of one of those persistently weird problems.
-
Bandit — static security analysis of Python code.
-
Quantum OS — Linux desktop based on Google’s Material Design. UI guidelines fascinate me: users love consistency, designers and brands hate that everything works the same.
-
Inside AWS — Every day, AWS installs enough server infrastructure to host the entire Amazon e-tailing business from back in 2004, when Amazon the retailer was one-tenth its current size at $7 billion in annual revenue. “What has changed in the last year,” Hamilton asked rhetorically, and then quipped: “We have done it 365 more times.” That is another way of saying that in the past year AWS has added enough capacity to support a $2.55 trillion online retailing operation, should one ever be allowed to exist.
Security and Privacy, ISP Measurement, Github for Education, and Mobile Numbers
-
A Critique of the Balancing Metaphor in Privacy and Security — The arguments presented by this paper are built on two underlying assertions. The first is that the assessment of surveillance measures often entails a judgement of whether any loss in privacy is legitimised by a justifiable increase in security. However, one fundamental difference between privacy and security is that privacy has two attainable end-states (absolute privacy through to the absolute absence of privacy), whereas security has only one attainable end-state (while the absolute absence of security is attainable, absolute security is a desired yet unobtainable goal). The second assertion, which builds upon the first, holds that because absolute security is desirable, new security interventions will continuously be developed, each potentially trading a small measure of privacy for a small rise in security. When assessed individually each intervention may constitute a justifiable trade-off. However, when combined together, these interventions will ultimately reduce privacy to zero. (via Alistair Croll)
-
ISP Interconnection and its Impact on Consumer Internet Performance (Measurement Lab) — In researching our report, we found clear evidence that interconnection between major U.S. access ISPs (AT&T, Comcast, CenturyLink, Time Warner Cable, and Verizon) and transit ISPs Cogent, Level 3, and potentially XO was correlated directly with degraded consumer performance throughout 2013 and into 2014 (in some cases, ongoing as of publication). Degraded performance was most pronounced during peak use hours, which points to insufficient capacity and congestion as a causal factor. Further, by noting patterns of performance degradation for access/transit ISP pairs that were synchronized across locations, we were able to conclude that in many cases degradation was not the result of major infrastructure failures at any specific point in a network, but rather connected with the business relationships between ISPs.
-
The Emergence of Github as Collaborative Platform for Education (PDF) — We argue that GitHub can support much of what traditional learning systems do, as well as go beyond them by supporting collaborative activities.
-
Mobile is Eating the World (A16Z) — mobile becoming truly ubiquitous, bringing opportunities to use the construct “X is eating Y.”
Distributed Systems Design 101, Patent Trolls, Intel's Half a Billion from IoT, and Google's Project Zero.
-
Inside bit.ly’s Distributed Systems — this is a 101 for modern web distributed systems design.
-
Patent Trolls are Now 67% of New Patent Lawsuits in USA (WaPo) — data from PwC.
-
Intel Made Half a Billion from Internet of Things Last Year (Quartz) — half a billion here, half a billion there, pretty soon it adds up to real money.
-
Google’s Project Zero (Wired) — G pays a team to attack common software and report the bugs to the manufacturer. Interesting hypothesis about how the numbers inbalance between Every Russian 14 Year Old and this small team doesn’t matter: modern hacker exploits often chain together a series of hackable flaws to defeat a computer’s defenses. Kill one of those bugs and the entire exploit fails. That means Project Zero may be able to nix entire collections of exploits by finding and patching flaws in a small part of an operating system, like the “sandbox” that’s meant to limit an application’s access to the rest of the computer. ”On certain attack surfaces, we’re optimistic we can fix the bugs faster than they’re being introduced,” Hawkes says. “If you funnel your research into these limited areas, you increase the chances of bug collisions.”
Available Data, Goal Setting, Real Tech, and Gamification Numbers
-
Dynamo and BigTable — good preso overview of two approaches to solving availability and consistency in the event of server failure or network partition.
-
Goals Gone Wild (PDF) — In this article, we argue that the beneficial effects of goal setting have been overstated and that systematic harm caused by goal setting has been largely ignored. We identify specific side effects associated with goal setting, including a narrow focus that neglects non-goal areas, a rise in unethical behavior, distorted risk preferences, corrosion of organizational culture, and reduced intrinsic motivation.
-
Tech Isn’t All Brogrammers (Alexis Madrigal) — a reminder that there are real scientists and engineers in Silicon Valley working on problems considerably harder than selling ads and delivering pet food to one another. (via Brian Behlendorf)
-
Numbers from 90+ Gamification Case Studies — cherry-picked anecdata for your business cases.
Machine Learning Mistakes, Recommendation Bandits, Droplet Robots, and Plain English
-
Machine Learning Done Wrong — [M]ost practitioners pick the modeling algorithm they are most familiar with rather than pick the one which best suits the data. In this post, I would like to share some common mistakes (the don’t-s).
-
Bandits for Recommendations — A common problem for internet-based companies is: which piece of content should we display? Google has this problem (which ad to show), Facebook has this problem (which friend’s post to show), and RichRelevance has this problem (which product recommendation to show). Many of the promising solutions come from the study of the multi-armed bandit problem.
-
Droplets — the Droplet is almost spherical, can self-right after being poured out of a bucket, and has the hardware capabilities to organize into complex shapes with its neighbors due to accurate range and bearing. Droplets are available open-source and use cheap vibration motors and a 3D printed shell. (via Robohub)
-
Apple’s App Store Approval Guidelines — some of the plainest English I’ve seen, especially the Introduction. I can only aspire to that clarity. If your App looks like it was cobbled together in a few days, or you’re trying to get your first practice App into the store to impress your friends, please brace yourself for rejection. We have lots of serious developers who don’t want their quality Apps to be surrounded by amateur hour.