Pinboard (jm) https://pinboard.in/u:jm/public/ recent bookmarks from jm Amazon Connect Talent vs. bias law 2026-04-30T10:04:52+00:00 https://www.linkedin.com/posts/coquinn_a-note-from-amazon-web-services-awss-share-7454936476359819264-8DqQ/?rcm=ACoAAAD7ciEBGVCp3m50-5sdPXL70GJw7TDNHVE jmAmazon Connect Talent was just announced. It conducts AI-powered conversational interviews with candidates, generates "anonymized competency scores," and surfaces ranked candidates to recruiters who "make the call." Fun fact: in New York City, that is an Automated Employment Decision Tool under Local Law 144. AEDTs require an annual independent bias audit with publicly posted results, plus at least ten business days of notice to candidates before use. Illinois, Colorado, and the EU AI Act impose adjacent obligations. The launch materials mention none of this. The compliance posture appears to be: candidate names are stripped from recruiter dashboards, therefore bias is solved. That is not how any of this works. Proxies for protected class -- speech patterns, zip codes, education history, the resume already sitting in your ATS -- are exactly what bias audits exist to measure. I don't think the product is bad. I think the announcement is conspicuously missing the guidance customers need before they can deploy it in NYC without violating Local Law 144 on day one. (The day's other news so far: Amazon Connect now ships as a four-SKU family, and there is a new design philosophy called "humorphism" with its own .com. Both feel small next to the above.) If you're selling automated hiring decisions in 2026, the bias-audit conversation belongs in the launch. ]]> bias law amazon aws recruiting regulation automation ai https://pinboard.in/ https://pinboard.in/u:jm/b:af7440a2d620/ Former Uber self-driving chief crashes his Tesla on FSD 2026-03-18T12:59:14+00:00 https://electrek.co/2026/03/17/former-uber-self-driving-chief-tesla-fsd-crash-supervision-problem/ jmTesla is asking humans to supervise a system that is specifically designed to make supervision feel pointless. As he puts it, an unreliable machine keeps you alert, and a perfect machine needs no oversight, but one that works almost perfectly creates a trap where drivers trust it just enough to stop paying attention. The research backs this up. Psychologists call it the “vigilance decrement”, monitoring a nearly perfect system is boring, boredom leads to mind-wandering, and drivers need 5 to 8 seconds to mentally reengage after an automated system hands control back. But emergencies unfold faster than that. Krikorian cites an Insurance Institute for Highway Safety study showing that after just one month of using adaptive cruise control, drivers were more than six times as likely to look at their phones. Tesla’s own website warns FSD users not to become complacent, but the system’s smooth performance actively trains that complacency. He points to two well-known crashes to illustrate the impossible math. In the 2018 Mountain View accident that killed Apple engineer Walter Huang, the driver had six seconds before his Tesla steered into a concrete median. He never touched the wheel. In the 2018 Uber crash in Tempe, Arizona, sensors detected a pedestrian with 5.6 seconds of warning, but the safety driver looked up with less than a second remaining. In Krikorian’s own case, he did take action, but he was asked to snap from passenger back to pilot in a fraction of a second, overriding months of conditioning. The logs show he turned the wheel. They don’t show the impossible math of that transition. The pattern Krikorian describes should sound familiar to anyone who has followed Tesla’s FSD controversies: condition the driver to rely on the system, erode their vigilance through months of smooth performance, then point to the terms of service and blame them when something breaks. When FSD works, Tesla gets credit. When it doesn’t, the driver gets blamed. ]]> fsd tesla risk attention supervision liability driving safety vigilance automation https://pinboard.in/ https://pinboard.in/u:jm/b:18572fcf8fa3/ The Computer Disease 2026-01-26T17:33:11+00:00 https://x.com/Swizec/status/2004633162522263987 jm "Well, Mr. Frankel, who started this program, began to suffer from the computer disease that anybody who works with computers now knows about. It's a very serious disease and it interferes completely with the work. The trouble with computers is you *play* with them. They are so wonderful. You have these switches - if it's an even number you do this, if it's an odd number you do that - and pretty soon you can do more and more elaborate things if you are clever enough, on one machine. After a while the whole system broke down. Frankel wasn't paying any attention; he wasn't supervising anybody. The system was going very, very slowly - while he was sitting in a room figuring out how to make one tabulator automatically print arc-tangent X, and then it would start and it would print columns and then bitsi, bitsi, bitsi, and calculate the arc-tangent automatically by integrating as it went along and make a whole table in one operation. Absolutely useless. We *had* tables of arc-tangents. But if you've ever worked with computers, you understand the disease - the *delight* in being able to see how much you can do. But he got the disease for the first time, the poor fellow who invented the thing." - Richard P. Feynman, _Surely You're Joking, Mr. Feynman!: Adventures of a Curious Character_ (via Swizec Teller)]]> automation fun computers richard-feynman the-computer-disease arc-tangents enjoyment hacking via:swizec-teller https://pinboard.in/ https://pinboard.in/u:jm/b:a74c508a5c4c/ RGG Studio’s test automation setup 2025-05-06T09:17:57+00:00 https://automaton-media.com/en/interviews/how-do-like-a-dragon-games-come-out-so-fast-one-of-rgg-studios-secrets-is-a-highly-efficient-testing-and-debugging-cycle-that-starts-as-soon-as-development-does/ jmAt first, we only dabbled in automated packaging and automated error detection, but we made the tools we needed to go further during the development of Yakuza 6, when we started automating the analysis of in-game logs and the issue tracking system for keeping track of bugs and tasks. Then, by the time Yakuza: Like a Dragon was released in 2020, we created the catchy sounding “fully automated bug detection system” (laughs). This is how it works – the history of actions you performed when playing the game manually (where you travelled, who you talked to, what items you used, etc.) is converted into commands and recorded, then automatically output as replay data (scripts) which you can edit manually and run as automated tests. Replay data continues to be recorded when running automated tests, and if a bug occurs during an automated test, the replay data gets saved, so you can run it back later to encounter the bug yourself. It often happens that you can’t reproduce a bug just by warping to its coordinates. This is because you also need to recreate the steps leading up to it – that’s why it’s important to record each step. Also, I’d like to mention that just implementing automated testing doesn’t mean much on its own, because you won’t know what the results of the tests are. That’s why we needed a crash report function to detect bugs. There’s also a function that records information needed to investigate detected bugs, as well as a way to check the status of successful tests. Then, by implementing a system that gives us a visualization of performance, we were able to make iteration more efficient, increasing the overall efficiency of the development process. ]]> automation testing yakuza games coding tests test-automation https://pinboard.in/ https://pinboard.in/u:jm/b:6337ae2e5588/ llms and humans unite, you have nothing to lose but your chores 2025-03-12T11:54:40+00:00 https://www.oblomovka.com/wp/2025/03/12/llms-and-humans-unite-you-have-nothing-to-lose-but-your-chores/ jm vibe-coding tools automation one-offs scripting web cdp google-chrome playwright claude hacks llms ai browsers https://pinboard.in/ https://pinboard.in/u:jm/b:dc01fa4629b8/ Arguments about AI summarisation 2025-03-11T12:52:57+00:00 https://lists.w3.org/Archives/Public/w3c-wai-gl/2025JanMar/0089.html jmSure I'm excited about new tech as the next person, but I want to express my concerns (sorry to point out some elephants in the room): 1. Ethics - major large language models rely on stolen training data, and they use low wage workers to 'train' at the expense of the well being of those workers. 2. Environment - Apart from raw material usage that comes with increase in processing power, LLMs uses a lot more energy and water than human scribes and summarisers do (both during training and at point of use). Magnitudes more, not negligible, such that major tech cos are building/buying nuclear power plants and areas near data centres suffer from water shortages and price hikes. Can we improve disability rights while disregarding environmental effects? 3. Quality - we've got a lot of experts in our group: who are sometimes wrong, sure, but it seems like a disservice to their input, knowledge and expertise to pipe their speech through LLMs. From the couple of groups I've been in that used AI summaries, I've seen them: - a. miss the point a lot of the time; it looks reasonable but doesn't match up with what people said/meant; - b. 'normalise' what was said to what most people would say, so it biases towards what's more common in training data, rather than towards the smart things individuals in this group often bring up. Normalising seems orthogonal to innovation? - c. create summaries that are either very long and wooly, with many unnecessary words, or short but incorrect. If we're considering if it's technically possible, I'd urge us to consider the problems with these systems too, including in ethics, environmental impact and quality. The "normalising" risk is one that hadn't occurred to me, but it makes perfect sense given how LLMs operate.]]> llms ai summarisation w3c discussion meetings automation transcription https://pinboard.in/ https://pinboard.in/u:jm/b:cb76582c6be9/ The state of Tomi's Home Assistant in 2024 2024-12-12T11:12:26+00:00 https://blog.rozman.info/the-state-of-my-home-assistant-in-2024/ jm home-automation ha home home-assistant hacks automation https://pinboard.in/ https://pinboard.in/u:jm/b:ba9cef057c5d/ Ethical Applications of AI to Public Sector Problems 2024-10-07T16:03:50+00:00 https://jacobian.org/2024/oct/1/ethical-public-sector-ai/ jmThere have been massive developments in AI in the last decade, and they’re changing what’s possible with software. There’s also been a huge amount of misunderstanding, hype, and outright bullshit. I believe that the advances in AI are real, will continue, and have promising applications in the public sector. But I also believe that there are clear “right” and “wrong” ways to apply AI to public sector problems. He breaks down AI usage into "Assistive AI", where AI is used to process and consume information (in ways or amounts that humans cannot) to present to a human operator, versus "Automated AI", where the AI both processes and acts upon information, without input or oversight from a human operator. The latter is unethical to apply in the public sector.]]> ai ethics llm genai public-sector government automation https://pinboard.in/ https://pinboard.in/u:jm/b:b6bce7069d26/ Modal interfaces considered harmful 2024-09-19T13:13:11+00:00 https://99percentinvisible.org/episode/children-of-the-magenta-automation-paradox-pt-1/ jmWhen one of the co-pilots hauled back on his stick, he pitched the plane into an angle that eventually caused the stall. [...] it’s possible that he didn’t understand that he was now flying in a different mode, one which would not regulate and smooth out his movements. This confusion about what how the fly-by-wire system responds in different modes is referred to, aptly, as “mode confusion,”  and it has come up in other accidents. ]]> automation aviation flying modal-interfaces ui ux interfaces modes mode-confusion air-france-447 disasters https://pinboard.in/ https://pinboard.in/u:jm/b:464d3b2a60c7/ Nevada's genAI-driven unemployment benefits system 2024-09-18T09:56:28+00:00 https://futurism.com/the-byte/nevada-unemployment-claims-ai jm "There’s no AI [written decisions] that are going out without having human interaction and that human review," DETR's director told the website. "We can get decisions out quicker so that it actually helps the claimant." [...] "The time savings they’re looking for only happens if the review is very cursory," explained Morgan Shah, the director of community engagement for Nevada Legal Services. "If someone is reviewing something thoroughly and properly, they’re really not saving that much time." Ultimately, Shah said, workers using the system to breeze through claims may end up "being encouraged to take a shortcut." [...] As with most attempts at using this still-nascent technology in the public sector, we probably won't know how well the Nevada unemployment AI works unless it's shown to be doing a bad job — which feels like an experiment being conducted on some of the most vulnerable members of society without their consent. Of course, the definition of a "bad job" depends who's defining it. If the system is processing a high volume of applications, it may not matter to its operators if it's processing them _correctly_ or not.]]> generative-ai ai racism bias nevada detr benefits automation https://pinboard.in/ https://pinboard.in/u:jm/b:82956e21b5a4/ remuslazar/homeassistant-carwings 2024-07-23T08:49:30+00:00 https://github.com/remuslazar/homeassistant-carwings jm todo homeassistant nissan leaf cars automation monitoring smarthome home https://pinboard.in/ https://pinboard.in/u:jm/b:9bd9c5690f8b/ AI and Israel’s Dystopian Promise of War without Responsibility 2024-04-15T08:54:27+00:00 https://internationalpolicy.org/publications/ai-and-israels-dystopian-promise-of-war-without-responsibility/ jm In Gaza we see an “indiscriminate” and “over the top” bombing campaign being actively rebranded by Israel as a technological step up, when in actuality there is currently no evidence that their so-called Gospel has produced results qualitatively better than those made by minds of flesh and blood. Instead, Israel’s AI has produced an endless list of targets with a decidedly lower threshold for civilian casualties. Human eyes and intelligence are demoted to rubber stamping a conveyor belt of targets as fast they can be bombed. It’s a path that the US military and policy makers should not only be wary of treading, but should reject loudly and clearly. In the future we may develop technology worthy of the name Artificial Intelligence, but we are not there yet. Currently the only promise a system such as Gospel AI holds is the power to occlude responsibility, to allow blame to fall on the machine picking the victims instead of the mortals providing the data. ]]> ai war grim-meathook-future israel gaza automation war-crimes lavender gospel https://pinboard.in/ https://pinboard.in/u:jm/b:4cb06082d4c6/ How to set up a Zappi to avoid draining solar batteries 2024-04-04T10:55:11+00:00 https://support.myenergi.com/hc/en-gb/articles/4716922623633-Hybrid-PV-Battery-Set-up-Avoid-Draining jm solar-pv sustainability home zappi power hacks automation https://pinboard.in/ https://pinboard.in/u:jm/b:ac855881fc6f/ The Mechanical Turk of Amazon Go 2024-01-31T17:27:15+00:00 https://pluralistic.net/2024/01/31/neural-interface-beta-tester/ jmA reader wrote to me this week. They're a multi-decade veteran of Amazon who had a fascinating tale about the launch of Amazon Go, the "fully automated" Amazon retail outlets that let you wander around, pick up goods and walk out again, while AI-enabled cameras totted up the goods in your basket and charged your card for them. According to this reader, the AI cameras didn't work any better than Tesla's full-self driving mode, and had to be backstopped by a minimum of three camera operators in an Indian call center, "so that there could be a quorum system for deciding on a customer's activity – three autopilots good, two autopilots bad." Amazon got a ton of press from the launch of the Amazon Go stores. A lot of it was very favorable, of course: Mister Market is insatiably horny for firing human beings and replacing them with robots, so any announcement that you've got a human-replacing robot is a surefire way to make Line Go Up. But there was also plenty of critical press about this – pieces that took Amazon to task for replacing human beings with robots. What was missing from the criticism? Articles that said that Amazon was probably lying about its robots, that it had replaced low-waged clerks in the USA with even-lower-waged camera-jockeys in India. Which is a shame, because that criticism would have hit Amazon where it hurts, right there in the ole Line Go Up. Amazon's stock price boost off the back of the Amazon Go announcements represented the market's bet that Amazon would evert out of cyberspace and fill all of our physical retail corridors with monopolistic robot stores, moated with IP that prevented other retailers from similarly slashing their wage bills. That unbridgeable moat would guarantee Amazon generations of monopoly rents, which it would share with any shareholders who piled into the stock at that moment. ]]> mechanical-turk amazon-go fakes amazon call-centers absent-indian ai fakery line-go-up automation capitalism https://pinboard.in/ https://pinboard.in/u:jm/b:8508b08c7940/ alienatedsec/solis-ha-modbus-cloud 2023-10-09T13:51:36+00:00 https://github.com/alienatedsec/solis-ha-modbus-cloud jm home-assistant solis solar-pv automation rs485 modbus https://pinboard.in/ https://pinboard.in/u:jm/b:cc88648c8c49/ Automation Bias 2023-08-08T10:39:14+00:00 https://en.wikipedia.org/wiki/Automation_bias jm automation bias complacency future ai ml tech via:etienneshrdlu https://pinboard.in/ https://pinboard.in/u:jm/b:760d2ba11e5d/ The CNET Fake News Fiasco, Autopilot, and the Uncanny Cognitive Valley 2023-02-07T11:31:15+00:00 https://garymarcus.substack.com/p/the-cnet-fake-news-fiasco-autopilot jmComplacency and overtrust. That brings us back to ChatGPT, CNET, and the present. Mackworth was prescient. Automation is a double-edged sword, and there is a kind of uncanny valley. We know perfectly well not to trust lousy systems; and wouldn’t need to pay attention to truly reliable systems, but the closer they get to perfect, the easier it is for mere mortals to space out. The CNET editors probably took a quick look at the large-language model generated prose, and thought it looked good enough; “complacency and over-trust” to their own detriment. If the autogenerated articles had been riddled with spelling errors and grammatical errors, the editors might not have trusted them. But at a superficial level the essays looked good enough, and so they editors trusted them, and a bunch of false information slipped through. That’s going to happen a whole lot more this year, with synthetic articles. Too many people are starting to count on systems that inherently can’t keep track of the truth. Few people are likely ever to pay enough attention. The stakes may get even higher if people start to rely on immature AI—which is all we have got, so far—for legal or medical advice. ]]> vigilance complacency ai machine-learning norman-mackworth cognition work automation uncanny-valley https://pinboard.in/ https://pinboard.in/u:jm/b:139c8f43fce4/ A healthcare algorithm started cutting care, and no one knew why 2023-01-09T14:44:28+00:00 https://www.theverge.com/2018/3/21/17144260/healthcare-medicaid-algorithm-arkansas-cerebral-palsy jm Legal Aid filed a federal lawsuit in 2016, arguing that the state had instituted a new [healthcare] policy without properly notifying the people affected about the change. There was also no way to effectively challenge the system, as they couldn’t understand what information factored into the changes, De Liban argued. No one seemed able to answer basic questions about the process. “The nurses said, ‘It’s not me; it’s the computer,’” De Liban says. When they dug into the system, they discovered more about how it works. Out of the lengthy list of items that assessors asked about, only about 60 factored into the home care algorithm. The algorithm scores the answers to those questions, and then sorts people into categories through a flowchart-like system. It turned out that a small number of variables could matter enormously: for some people, a difference between a score of a three instead of a four on any of a handful of items meant a cut of dozens of care hours a month. (Fries didn’t say this was wrong, but said, when dealing with these systems, “there are always people at the margin who are going to be problematic.”) [...] From the state’s perspective, the most embarrassing moment in the dispute happened during questioning in court. Fries was called in to answer questions about the algorithm and patiently explained to De Liban how the system works. After some back-and-forth, De Liban offered a suggestion: “Would you be able to take somebody’s assessment report and then sort them into a category?” [...] Fries said he could, although it would take a little time. He looked over the numbers for Ethel Jacobs. After a break, a lawyer for the state came back and sheepishly admitted to the court: there was a mistake. Somehow, the wrong calculation was being used. They said they would restore Jacobs’ hours. “Of course we’re gratified that DHS has reported the error and certainly happy that it’s been found, but that almost proves the point of the case,” De Liban said in court. “There’s this immensely complex system around which no standards have been published, so that no one in their agency caught it until we initiated federal litigation and spent hundreds of hours and thousands of dollars to get here today. That’s the problem.” ]]> algorithms government health healthcare automation grim-meathook-future future https://pinboard.in/ https://pinboard.in/u:jm/b:738238124fc3/ BirdNET-Pi 2022-04-19T09:21:24+00:00 https://birdnetpi.com/ jm audio birds birdnet automation raspberry-pi cool identification https://pinboard.in/ https://pinboard.in/u:jm/b:34e78fcfa03f/ We won! UK Home Office to stop using racist visa algorithm 2022-02-10T10:54:11+00:00 https://www.jcwi.org.uk/news/we-won-home-office-to-stop-using-racist-visa-algorithm jmThe visa algorithm discriminated on the basis of nationality - by design. Applications made by people holding ‘suspect’ nationalities received a higher risk score. Their applications received intensive scrutiny by Home Office officials, were approached with more scepticism, took longer to determine, and were much more likely to be refused. We argued this was racial discrimination and breached the Equality Act 2010. Entrenched bias and racism in the visa system breaks hearts and tears families apart, like the four siblings from Nigeria unable to travel to the UK for their sister's wedding, or the countless skilled professionals refused unable to contribute to conferences and events in the UK just because they don't come from a rich white country - including scores of African academics and artists denied entry for no good reason. The streaming tool was opaque. Aside from admitting the existence of a secret list of suspect nationalities, the Home Office refused to provide meaningful information about the algorithm. It remains unclear what other factors were used to grade applications. The algorithm suffered from a feedback loop — a vicious circle in which biased enforcement and visa statistics reinforce which countries stay on the list of suspect nationalities. In short, applicants from suspect nationalities were more likely to have their visa application rejected. These visa rejections then informed which nationalities appeared on the list of ‘suspect’ nations. This error, combined with the pre-existing bias in Home Office enforcement (in which some nationalities are targeted for enforcement because they are believed to be easier to remove), accelerated bias in the Home Office’s visa process. Such feedback loops are a well-documented problem with automated decision systems. ]]> algorithms racism uk immigration automation home-office https://pinboard.in/ https://pinboard.in/u:jm/b:7f3d2b387e37/ FIND 2021-11-24T15:25:40+00:00 https://www.internalpositioning.com/ jm location home-assistant home automation tracking devices https://pinboard.in/ https://pinboard.in/u:jm/b:9457f1dfb5a3/ IRONIES OF AUTOMATION - ScienceDirect 2021-08-04T13:54:48+00:00 https://www.sciencedirect.com/science/article/pii/B9780080293486500269 jm via:abebab hci human-computer-interaction interfaces automation papers ai https://pinboard.in/ https://pinboard.in/u:jm/b:1d4ea8ab9534/ Fan-tastic misuse of Raspberry Pi's GPIO 2021-04-12T14:17:50+00:00 https://hackaday.com/2021/04/06/fan-tastic-misuse-of-raspberry-pi-gpio/ jm home automation gpio raspberry-pi hacks hardware wireless https://pinboard.in/ https://pinboard.in/u:jm/b:93ac2157998f/ Hammerspoon 2020-12-03T11:43:32+00:00 https://www.hammerspoon.org/ jm via:fanf automation osx mac lua scripting hammerspoon https://pinboard.in/ https://pinboard.in/u:jm/b:2eb669dd1863/ Charles proxy for web scraping 2020-11-13T18:09:17+00:00 https://www.scrapingbee.com/blog/charles-proxy/ jm scraping mitm charles web http proxies web-scraping automation https://pinboard.in/ https://pinboard.in/u:jm/b:7a89e2e70395/ Pasco’s sheriff created a futuristic program to stop crime before it happens. It monitors and harasses families. | Investigations | Tampa Bay Times 2020-09-04T09:23:04+00:00 https://projects.tampabay.com/projects/2020/investigations/police-pasco-sheriff-targeted/intelligence-led-policing/ jm civil-rights police algorithms policing automation florida us-politics https://pinboard.in/ https://pinboard.in/u:jm/b:ddce9f4b4de9/ Automating safe, hands-off deployments 2020-06-23T22:53:17+00:00 https://aws.amazon.com/builders-library/automating-safe-hands-off-deployments/ jmEach team needs to balance the safety of small-scoped deployments with the speed at which we can deliver changes to customers in all Regions. Deploying changes to 24 Regions or 76 Availability Zones through the pipeline one at a time has the lowest risk of causing broad impact, but it could take weeks for the pipeline to deliver a change to customers globally. We have found that grouping deployments into “waves” of increasing size, as seen in the previous sample prod pipeline, helps us achieve a good balance between deployment risk and speed. Each wave’s stage in the pipeline orchestrates deployments to a group of Regions, with changes being promoted from wave to wave. New changes can enter the production phase of the pipeline at any time. After a set of changes is promoted from the first step to the second step in wave 1, the next set of changes from gamma is promoted into the first step of wave 1, so we don’t end up with large bundles of changes waiting to be deployed to production. The first two waves in the pipeline build the most confidence in the change: The first wave deploys to a Region with a low number of requests to limit the possible impact of the first production deployment of the new change. The wave deploys to only one Availability Zone (or cell) at a time within that Region to cautiously deploy the change across the Region. The second wave then deploys to one Availability Zone (or cell) at a time in a Region with a high number of requests where it is highly likely that customers will exercise all the new code paths and where we get good validation of the changes. After we have higher confidence in the safety of the change from the initial pipeline waves’ deployments, we can deploy to more and more Regions in parallel in the same wave. For example, the previous sample prod pipeline deploys to three Regions in wave 3, then to up to 12 Regions in wave 4, then to the remaining Regions in wave 5. The exact number and choice of Regions in each of these waves and the number of waves in a service team’s pipeline depend on the individual service’s usage patterns and scale. The later waves in the pipeline still help us achieve our objective to prevent negative impact to multiple Availability Zones in the same Region. When a wave deploys to multiple Regions in parallel, it follows the same cautious rollout behavior for each Region that was used in the initial waves. Each step in the wave only deploys to a single Availability Zone or cell from each Region in the wave. ]]> automation ops devops amazon aws deployment waves az multi-region ci cd https://pinboard.in/ https://pinboard.in/u:jm/b:7ecaa0ea15ea/ Google’s medical AI was super accurate in a lab. Real life was a different story. | MIT Technology Review 2020-04-28T15:55:43+00:00 https://www.technologyreview.com/2020/04/27/1000658/google-medical-ai-accurate-lab-real-life-clinic-covid-diabetes-retina-disease/?truid=8c8f2699f50eb3b9985a111121cfee47 jmWhen it worked well, the AI did speed things up. But it sometimes failed to give a result at all. Like most image recognition systems, the deep-learning model had been trained on high-quality scans; to ensure accuracy, it was designed to reject images that fell below a certain threshold of quality. With nurses scanning dozens of patients an hour and often taking the photos in poor lighting conditions, more than a fifth of the images were rejected. Patients whose images were kicked out of the system were told they would have to visit a specialist at another clinic on another day. If they found it hard to take time off work or did not have a car, this was obviously inconvenient. Nurses felt frustrated, especially when they believed the rejected scans showed no signs of disease and the follow-up appointments were unnecessary. They sometimes wasted time trying to retake or edit an image that the AI had rejected. Because the system had to upload images to the cloud for processing, poor internet connections in several clinics also caused delays. “Patients like the instant results, but the internet is slow and patients then complain,” said one nurse. “They’ve been waiting here since 6 a.m., and for the first two hours we could only screen 10 patients.” The Google Health team is now working with local medical staff to design new workflows. For example, nurses could be trained to use their own judgment in borderline cases. The model itself could also be tweaked to handle imperfect images better.  ]]> google health medicine ai automation software internet developing-world real-world images scanning https://pinboard.in/ https://pinboard.in/u:jm/b:5622d91f3d62/ Why cancer-spotting AI needs to be handled with care 2020-01-29T16:13:32+00:00 https://www.theverge.com/2020/1/27/21080253/ai-cancer-diagnosis-dangers-mammography-google-paper-accuracy jm“There’s this idea in society that finding more cancers is always better, but it’s not always true,” Adewole Adamson, a dermatologist and assistant professor at Dell Medical School, tells The Verge. “The goal is finding more cancers that are actually going to kill people.” But the problem is “there’s no gold standard for what constitutes cancer.” As studies have found, you can show the same early-stage lesions to a group of doctors and get completely different answers about whether it’s cancer. And even if they do agree that that’s what a lesion shows — and their diagnoses are right — there’s no way of knowing whether that cancer is a threat to someone’s life. This leads to overdiagnosis, says Adamson: “Calling things cancer that, if you didn’t go looking for them, wouldn’t harm people over their lifetime.” As soon as you do call something cancer, it triggers a chain of medical intervention that can be painful, costly, and life-changing. In the case of breast cancer, that might mean radiation treatments, chemotherapy, the removal of tissue from the breast (a lumpectomy), or the removal of one or both breasts entirely (a mastectomy). These aren’t decisions to be rushed. Overdiagnosis, he says, “is a problem for a lot of different cancers; for prostate, melanoma, breast cancer, thyroid. And if AI systems become better and better at finding smaller and smaller lesions you will manufacture a lot of pseudo-patients who have a ‘disease’ that won’t actually kill them.” ]]> overdiagnosis health medicine cancer computer-vision automation ai google diagnosis https://pinboard.in/ https://pinboard.in/u:jm/b:2a73fa29b63c/ Ironies of automation 2020-01-09T17:32:47+00:00 https://blog.acolyer.org/2020/01/08/ironies-of-automation/ jmIn an automated system, two roles are left to humans: monitoring that the automated system is operating correctly, and taking over control if it isn’t. An operator that doesn’t routinely operate the system will have atrophied skills if ever called on to take over. Unfortunately, physical skills deteriorate when they are not used, particularly the refinements of gain and timing. This means that a formerly experienced operator who has been monitoring an automated process may now be an inexeperienced one. Not only are the operator’s skills declining, but the situations when the operator will be called upon are by their very nature the most demanding ones where something is deemed to be going wrong. Thus what we really need in such a situation is a more, not a lesser skilled operator! To generate successful strategies for unusual situtations, an operator also needs good understanding of the process under control, and the current state of the system. The former understanding develops most effectively through use and feedback (which the operator may no longer be getting the regular opportunity for), the latter takes some time to assimilate. (via John Allspaw)]]> via:allspaw automation software reliability debugging ops design failsafe failure human-interfaces ui ux outages https://pinboard.in/ https://pinboard.in/u:jm/b:8dde64ad2f41/ Home Automation Without The Cloud 2019-12-02T14:14:59+00:00 https://aaronparecki.com/home-automation/ jm via:nelson house home automation iot cloud-free https://pinboard.in/ https://pinboard.in/u:jm/b:286a34c5d4eb/ Stop explaining black box machine learning models for high stakes decisions and use interpretable models instead 2019-10-28T17:32:31+00:00 https://blog.acolyer.org/2019/10/28/interpretable-models/ jmBlack box machine learning models are currently being used for high stakes decision-making throughout society, causing problems throughout healthcare, criminal justice, and in other domains. People have hoped that creating methods for explaining these black box models will alleviate some of these problems, but trying to _explain_ black box models, rather than creating models that are _interpretable_ in the first place, is likely to perpetuate bad practices and can potentially cause catastrophic harm to society. There is a way forward -- it is to design models that are inherently interpretable. This manuscript clarifies the chasm between explaining black boxes and using inherently interpretable models, outlines several key reasons why explainable black boxes should be avoided in high-stakes decisions, identifies challenges to interpretable machine learning, and provides several example applications where interpretable models could potentially replace black box models in criminal justice, healthcare, and computer vision. I wholeheartedly support this idea, it makes a lot of sense to me in terms of producing ML/AI that can be supported operationally.]]> machine-learning ai ops support transparency papers black-box models computer-says-no automation explainability https://pinboard.in/ https://pinboard.in/u:jm/b:bb9229d3cbeb/ Operating Apache Kafka Clusters 24/7 Without A Global Ops Team 2019-10-02T10:00:01+00:00 https://eng.lyft.com/operating-apache-kafka-clusters-24-7-without-a-global-ops-team-417813a5ce70 jm autoremediation failures ops kafka scalability automation https://pinboard.in/ https://pinboard.in/u:jm/b:457b3e9b4a93/ Crash Course | The New Republic 2019-09-23T11:50:52+00:00 https://newrepublic.com/article/154944/boeing-737-max-investigation-indonesia-lion-air-ethiopian-airlines-managerial-revolution jm[Boeing] engineers devised a software fix called MCAS, which pushed the nose down in response to an obscure set of circumstances in conjunction with the “speed trim system,” which Boeing had devised in the 1980s to smooth takeoffs. Once the 737 MAX materialized as a real-life plane about four years later, however, test pilots discovered new realms in which the plane was more stall-prone than its predecessors. So Boeing modified MCAS to turn down the nose of the plane whenever an angle-of-attack (AOA) sensor detected a stall, regardless of the speed. That involved giving the system more power and removing a safeguard, but not, in any formal or genuine way, running its modifications by the FAA, which might have had reservations with two critical traits of the revamped system: Firstly, that there are two AOA sensors on a 737, but only one, fatefully, was programmed to trigger MCAS. The former Boeing engineer Ludtke and an anonymous whistle-blower interviewed by 60 Minutes Australia both have a simple explanation for this: Any program coded to take data from both sensors would have had to account for the possibility the sensors might disagree with each other and devise a contingency for reconciling the mixed signals. Whatever that contingency, it would have involved some kind of cockpit alert, which would in turn have required additional training—probably not level-D training, but no one wanted to risk that. So the system was programmed to turn the nose down at the feedback of a single (and somewhat flimsy) sensor. And, for still unknown and truly mysterious reasons, it was programmed to nosedive again five seconds later, and again five seconds after that, over and over ad literal nauseam.
 And then, just for good measure, a Boeing technical pilot emailed the FAA and casually asked that the reference to the software be deleted from the pilot manual.
 So no more than a handful of people in the world knew MCAS even existed before it became infamous. Here, a generation after Boeing’s initial lurch into financialization, was the entirely predictable outcome of the byzantine process by which investment capital becomes completely abstracted from basic protocols of production and oversight: a flight-correction system that was essentially jerry-built to crash a plane. “If you’re looking for an example of late stage capitalism or whatever you want to call it,” said longtime aerospace consultant Richard Aboulafia, “it’s a pretty good one.”
 ]]> boeing business capitalism engineering management fail disasters automation cost-control stock-market fly-by-wire https://pinboard.in/ https://pinboard.in/u:jm/b:147b64ad3d93/ Climate change: 'We've created a civilisation hell bent on destroying itself – I'm terrified', writes Earth scientist 2019-05-28T15:43:36+00:00 https://theconversation.com/climate-change-weve-created-a-civilisation-hell-bent-on-destroying-itself-im-terrified-writes-earth-scientist-113055 jm'At some point in the future, the technosphere could even function without humans. We worry about robots taking over human’s jobs. Perhaps we should be more concerned with them taking over our role as apex consumers.' ]]> technology ecology technosphere extinction er fear grim future automation climate-crisis climate-change https://pinboard.in/ https://pinboard.in/u:jm/b:939c9ea0c3ab/ When License-Plate Surveillance Goes Horribly Wrong - The New York Times 2019-04-24T10:21:43+00:00 https://www.nytimes.com/2019/04/23/opinion/when-license-plate-surveillance-goes-horribly-wrong.html jm“They built a system to mitigate harm, and yet I ended up with guns pulled on me due to faulty data,” he said. “And it’s more proof that we’ve built this invisible layer behind the scenes that leads to real-world consequences.” This is the common thread between automated surveillance systems -- false positives happen, but the systems are designed to assume this is harmless.]]> false-positives surveillance anpr license-plates automation https://pinboard.in/ https://pinboard.in/u:jm/b:9d122cf1cae5/ Taming the Beast: How Scylla Leverages Control Theory to Keep Compactions Under Control - ScyllaDB 2018-06-14T10:29:17+00:00 https://www.scylladb.com/2018/06/12/scylla-leverages-control-theory/ jm At any given moment, a database like ScyllaDB has to juggle the admission of foreground requests with background processes like compactions, making sure that the incoming workload is not severely disrupted by compactions, nor that the compaction backlog is so big that reads are later penalized. In this article, we showed that isolation among incoming writes and compactions can be achieved by the Schedulers, yet the database is still left with the task of determining the amount of shares of the resources incoming writes and compactions will use. Scylla steers away from user-defined tunables in this task, as they shift the burden of operation to the user, complicating operations and being fragile against changing workloads. By borrowing from the strong theoretical background of industrial controllers, we can provide an Autonomous Database that adapts to changing workloads without operator intervention. ]]> scylladb storage settings compaction automation thresholds control-theory ops cassandra feedback https://pinboard.in/ https://pinboard.in/u:jm/b:0924f70f896e/ Austerity is an Algorithm 2018-04-09T15:54:46+00:00 https://logicmag.io/03-austerity-is-an-algorithm/ jmThings changed in December 2016, when the government announced that the system had undergone full automation. Humans would no longer investigate anomalies in earnings. Instead, debt notices would be automatically generated when inconsistencies were detected. The government’s rationale for automating the process was telling. “Our aim is to ensure that people get what they are entitled to—no more and no less,” read the press release. “And to crack down hard when people deliberately defraud the system.” The result was a disaster. I’ve had friends who’ve received an innocuous email urging them to check their MyGov account—an online portal available to Australian citizens with an internet connection to access a variety of government services—only to log in and find they’re hundreds or thousands of dollars in arrears, supposedly because they didn’t accurately report their income. Some received threats from private debt collectors, who told them their wages would be seized if they didn’t submit to a payment plan. Those who wanted to contest their debts had to lodge a formal complaint, and were subjected to hours of Mozart’s Divertimento in F Major before they could talk to a case worker. Others tried taking their concerns directly to the Centrelink agency on Twitter, where they were directed to calling Lifeline, a 24-hour hotline for crisis support and suicide prevention. At the end of 2015, my friend Chloe received a notice claiming she owed $20,000 to the government. She was told that she had reported her income incorrectly while on Youth Allowance, which provides financial assistance to certain categories of young people. The figure was shocking and, like others in her position, she grew suspicious. She decided to contest the debt: she contacted all of her previous employers so she could gather pay slips, and scanned them into the MyGov app. “I gave them all of my information to prove that there was no way I owed them $20,000,” she says. The bean counters were unmoved. They maintained that Chloe had reported her after-tax income instead of her before-tax income. As a result, they increased the amount she owed to $30,000. She agreed to a payment plan, which will see her pay off the debt in fortnightly installments of $50 over the course of two decades. “I even looked into bankruptcy because I was so stressed by it,” she says. “All I could think about was the Centrelink debt, and once they upped it to 30k, I was so ashamed and sad and miserable,” she says. ]]> austerity algorithms automation dystopia australia government debt-collectors robo-debt dole benefit grim-meathook-future https://pinboard.in/ https://pinboard.in/u:jm/b:48d3f90d157a/ Sonarr 2017-12-01T15:49:08+00:00 https://sonarr.tv/ jm pvr tv automation usenet bittorrent https://pinboard.in/ https://pinboard.in/u:jm/b:2579be14b458/ Something is wrong on the internet – James Bridle – Medium 2017-11-06T21:19:41+00:00 https://medium.com/@jamesbridle/something-is-wrong-on-the-internet-c39c471271d2 jm internet youtube children web automation violence horror 4chan james-bridle https://pinboard.in/ https://pinboard.in/u:jm/b:1fb2f7c526d7/ Automated unemployment insurance fraud detection system had a staggering 93% error rate in production 2017-03-28T10:31:46+00:00 https://twitter.com/elizabeth_joh/status/838967852476485632 jm State officials have said that between Oct. 1, 2013, when the MiDAS [automated unemployment insurance fraud detection] system came on line, and Aug. 7, 2015, when the state halted the auto-adjudication of fraud determinations and began to require some human review of MiDAS findings, the system had a 93% error rate and made false fraud findings affecting more than 20,000 unemployment insurance claims. Those falsely accused of fraud were subjected to quadruple penalties and aggressive collection techniques, including wage garnishment and seizure of income tax refunds. Some were forced into bankruptcy. The agency is now reviewing about 28,000 additional fraud determinations that were made during the relevant period, but which involved some human review. An unknown number of those fraud findings were also false.
]]> fraud broken fail michigan detroit social-welfare us-politics computer-says-no automation discrimination fraud-detection https://pinboard.in/ https://pinboard.in/u:jm/b:053098e43fed/ Chatbot that overturned 160,000 parking fines now helping refugees claim asylum | Technology | The Guardian 2017-03-09T13:57:09+00:00 https://www.theguardian.com/technology/2017/mar/06/chatbot-donotpay-refugees-claim-asylum-legal-aid jmThe original DoNotPay, created by Stanford student Joshua Browder, describes itself as “the world’s first robot lawyer”, giving free legal aid to users through a simple-to-use chat interface. The chatbot, using Facebook Messenger, can now help refugees fill in an immigration application in the US and Canada. For those in the UK, it helps them apply for asylum support.
]]> government technology automation bots asylum forms facebook https://pinboard.in/ https://pinboard.in/u:jm/b:69e6fff8a2e9/ DST breaks everything 2016-11-02T10:56:40+00:00 http://www.telegraph.co.uk/travel/news/half-of-heathrow-noise-complaints-made-by-just-10-people/ jmIn January last year the airport unearthed a scheme whereby campaigners were using automated software to generate complaints against the airport. Officials caught out the set-up when the two anti-Heathrow enthusiasts forgot to take into account the hour going back in October, and began complaining about flights that had not yet taken off or arrived. ]]> bugs dst daylight-savings-time funny heathrow complaints automation noise https://pinboard.in/ https://pinboard.in/u:jm/b:202113a95287/ Google Intrusion Detection Problems 2016-08-24T10:42:47+00:00 http://www.fredtrotter.com/2016/08/22/google-intrusion-detection-problems/ jm'We have lost access to multiple critical data stores because Google has an automated threat detection system that is incapable of handling false positives.' ]]> google security cloud false-positives intrusion-detection automation fail https://pinboard.in/ https://pinboard.in/u:jm/b:58c560e976c2/ IMDB on automation, pt 2 2016-07-05T09:20:41+00:00 http://imgs.xkcd.com/comics/is_it_worth_the_time.png jm quotes time automation hacks life imdb productivity efficiency https://pinboard.in/ https://pinboard.in/u:jm/b:5fd965e6ee34/ IMDB on automation 2016-07-05T09:19:37+00:00 http://imgs.xkcd.com/comics/automation.png jm ifttt quotes automation coding hacks reality https://pinboard.in/ https://pinboard.in/u:jm/b:d79605ba66e1/ The tyranny of the algorithm yet again... 2016-06-15T10:09:17+00:00 https://b2fxxx.blogspot.ie/2016/05/the-tyranny-of-algorithm-yet-again.html jmThat these place names exist won't be a surprise to anyone familiar with English limnology - the study of rivers and inland waters. As Wikipedia helpfully tells us, "The Isis is the name given to the part of the River Thames above Iffley Lock which flows through the university city of Oxford". In at least one local primary school I'm familiar with, the classes are called Windrush, Cherwell, Isis and Thames. [...] Now PayPal has decided that they are not prepared to facilitate payments for goods to be delivered to an address which includes the word "Isis". An Isis street resident ran into some unexpected difficulties when attempting to purchase a small quantity of haberdashery on the internet with the aid of a PayPal account. The transaction would not process. In puzzlement she eventually got irritated enough to brave the 24/7 customer support telephone tag labyrinth. The short version of the response from the eventual real person she managed to get through to was that PayPal have blacklisted addresses which include the name "Isis". They will not process payments for goods to be delivered to an Isis related address, whatever state of privileged respectability the residents of such properties may have earned or inherited in their lifetimes to this point. One has to wonder if this also brings the risk of adding the user to a secret list, somewhere. Trial by algorithm.]]> isis algorithms automation fail law-enforcement paypal uk rivers https://pinboard.in/ https://pinboard.in/u:jm/b:50bf0ccbb6fd/ Revealed: How copyright law is being misused to remove material from the internet 2016-05-23T12:46:50+00:00 https://www.theguardian.com/technology/2016/may/23/copyright-law-internet-mumsnet?CMP=share_btn_tw jmIn fact, no copyright infringement had occurred at all. Instead, something weirder had happened. At some point after Narey posted her comments on Mumsnet, someone had copied the entire text of one of her posts and pasted it, verbatim, to a spammy blog titled “Home Improvement Tips and Tricks”. The post, headlined “Buildteam interior designers” was backdated to September 14 2015, three months before Narey had written it, and was signed by a “Douglas Bush” of South Bend, Indiana. The website was registered to someone quite different, though: Muhammed Ashraf, from Faisalabad, Pakistan. Quite why Douglas Bush or Muhammed Ashraf would be reviewing a builder based in Clapham is not explained in “his” post. BuildTeam says it has no idea why Narey’s review was reposted, but that it had nothing to do with it. “At no material times have we any knowledge of why this false DCMA take down was filed, nor have we contracted any reputation management firms, or any individual or a group to take such action on our behalf. Finally, and in conjunction to the above, we have never spoken with a ‘Douglas Bush,’ or a ‘Muhammed Ashraf.’” ]]> fraud censorship mumsnet dmca takedowns google automation copyright https://pinboard.in/ https://pinboard.in/u:jm/b:6965697186a8/ Jenkins 2.0 2016-03-27T10:20:43+00:00 https://jenkins.io/2.0/#pipelines jm via:eric jenkins ci cd deployment pipelines testing automation build https://pinboard.in/ https://pinboard.in/u:jm/b:435381483deb/ Diffy: Testing services without writing tests 2015-09-03T22:58:30+00:00 https://blog.twitter.com/2015/diffy-testing-services-without-writing-tests jm http testing thrift automation twitter diffy diff soa tests https://pinboard.in/ https://pinboard.in/u:jm/b:aebc474734fc/ One Direction Offers Remix Competition, Then Sony/Soundcloud Punish The Entrants As Copyright Infringers | Techdirt 2015-07-20T21:16:56+00:00 https://www.techdirt.com/articles/20150719/08200431691/one-direction-offers-remix-competition-then-sonysoundcloud-punish-entrants-as-copyright-infringers.shtml jmTorrentFreak has the story of a UK-producer and songwriter named Lee Adams who took part in an official remix competition of boy band One Direction's music, put on by the band and its label, Sony Music. The stems for remixing were released on Soundcloud. The rules of the contest required entrants to upload their remixes on Soundcloud... and that's exactly what Adams did. And yet those works still got taken down via copyright claims from Sony Music as infringing. ]]> sony soundcloud anti-piracy automation piracy stems remixing one-direction lee-adams https://pinboard.in/ https://pinboard.in/u:jm/b:647b33a4a826/ President's message gets lost in (automated) translation 2015-03-18T14:34:57+00:00 http://www.irishexaminer.com/ireland/presidents-message-gets-lost-in-translation-319077.html jmIn a series of bizarre translations, YouTube’s automated translation service took artistic licence with the [President's] words of warmth. When the head of state sent St Patrick’s Day greetings to viewers, the video sharing site said US comedian Tina Fey was being “particular with me head”. As President Higgins spoke of his admiration for Irish emigrants starting new communities abroad, YouTube said the President referenced blackjack and how he “just couldn’t put the new iPhone” down. And, in perhaps the most unusual moment, as he talked of people whose hearts have sympathy, the President “explained” he was once on a show “that will bar a gift card”. (via Daragh O'Brien)]]> lol president ireland michael-d-higgins automation translation machine-learning via:daraghobrien funny blackjack iphone tina-fey st-patrick fail https://pinboard.in/ https://pinboard.in/u:jm/b:441f44ce8890/ Automating Tinder with Eigenfaces 2015-02-11T17:12:40+00:00 http://crockpotveggies.com/2015/02/09/automating-tinder-with-eigenfaces.html jmWhile my friends were getting sucked into "swiping" all day on their phones with Tinder, I eventually got fed up and designed a piece of software that automates everything on Tinder. This is awesome. (via waxy) ]]> via:waxy tinder eigenfaces machine-learning k-nearest-neighbour algorithms automation ai https://pinboard.in/ https://pinboard.in/u:jm/b:2bf3ed9b9679/ Amazon sellers hit by nightmare before Christmas as glitch cuts prices to 1p | Technology | The Guardian 2014-12-14T19:39:32+00:00 http://www.theguardian.com/money/2014/dec/14/amazon-glitch-prices-penny-repricerexpress jmFrom 7-8pm on Friday, [RepricerExpress] software, used by third-party sellers to ensure their products are the cheapest on the market, went a bit haywire and reduced prices to as little as 1p. ]]> 1p amazon resellers repricer-express fail price-cutting automation risks undercutting https://pinboard.in/ https://pinboard.in/u:jm/b:10f434e3efd7/ Should Airplanes Be Flying Themselves? 2014-11-14T14:57:02+00:00 http://www.vanityfair.com/business/2014/10/air-france-flight-447-crash# jm airlines automation flight flying accidents post-mortems af447 air-france autopilot alerts pilots team-leaders clipper-skippers alternate-law https://pinboard.in/ https://pinboard.in/u:jm/b:404d98e6a267/ Pillar 2014-06-16T12:56:53+00:00 https://github.com/comeara/pillar jmManages migrations for your Cassandra data stores. Pillar grew from a desire to automatically manage Cassandra schema as code. Managing schema as code enables automated build and deployment, a foundational practice for an organization striving to achieve Continuous Delivery. Pillar is to Cassandra what Rails ActiveRecord migrations or Play Evolutions are to relational databases with one key difference: Pillar is completely independent from any application development framework. ]]> migrations database ops pillar cassandra activerecord scala continuous-delivery automation build https://pinboard.in/ https://pinboard.in/u:jm/b:acc70894611d/ fcron 2014-04-17T14:51:58+00:00 http://fcron.free.fr/ jmFcron is a scheduler. It aims at replacing Vixie Cron, so it implements most of its functionalities. But contrary to Vixie Cron, fcron does not need your system to be up 7 days a week, 24 hours a day : it also works well with systems which are running only occasionnally (contrary to anacrontab). In other words, fcron does both the job of Vixie Cron and anacron, but does even more and better :)) ... Thanks Craig! ]]> via:chughes cron fcron unix linux ops scheduler automation scripts https://pinboard.in/ https://pinboard.in/u:jm/b:d26ef4ca83ce/ Huginn 2014-04-14T15:39:50+00:00 https://github.com/cantino/huginn/blob/master/README.md jma system for building agents that perform automated tasks for you online. They can read the web, watch for events, and take actions on your behalf. Huginn's Agents create and consume events, propagating them along a directed event flow graph. Think of it as Yahoo! Pipes plus IFTTT on your own server. You always know who has your data. You do. MIT-licensed open source, built on Rails.]]> ifttt automation huginn ruby rails open-source agents https://pinboard.in/ https://pinboard.in/u:jm/b:45473ceb4e29/ Google: Our Robot Cars Are Better Drivers Than Puny Humans | MIT Technology Review 2013-10-26T22:27:23+00:00 http://www.technologyreview.com/news/520746/data-shows-googles-robot-cars-are-smoother-safer-drivers-than-you-or-i/ jmOne of those analyses showed that when a human was behind the wheel, Google’s cars accelerated and braked significantly more sharply than they did when piloting themselves. Another showed that the cars’ software was much better at maintaining a safe distance from the vehicle ahead than the human drivers were. “We’re spending less time in near-collision states,” said Urmson. “Our car is driving more smoothly and more safely than our trained professional drivers.” ]]> google cars driving safety roads humans robots automation https://pinboard.in/ https://pinboard.in/u:jm/b:7aeed9b5001b/ Juniper Adds Puppet support 2013-08-27T14:46:59+00:00 http://packetpushers.net/pull-my-strings-im-your-puppet-juniper-bringing-devops-to-networking/ jm ssh cli automation networking networks puppet ops juniper cisco https://pinboard.in/ https://pinboard.in/u:jm/b:dd0799ce4597/ When 'Smart Homes' Get Hacked: I Haunted A Complete Stranger's House Via The Internet - Forbes 2013-07-27T17:37:05+00:00 http://www.forbes.com/sites/kashmirhill/2013/07/26/smart-homes-hack/ jm[Trustwave's Crowley] found security flaws that would allow a digital intruder to take control of a number of sensitive devices beyond the Insteon systems, from the Belkin WeMo Switch to the Satis Smart Toilet. Yes, they found that a toilet was hackable. You only have to have the Android app for the $5,000 toilet on your phone and be close enough to the toilet to communicate with it. “It connects through Bluetooth, with no username or password using the pin ‘0000’,” said Crowley. “So anyone who has the application on their phone and was connected to the network could control anyone else’s toilet. You could turn the bidet on while someone’s in there.” ]]> home automation insteon security hardware fail attacks bluetooth han trustwave belkin satis https://pinboard.in/ https://pinboard.in/u:jm/b:812c51732106/ ssh - fabric appears to start apache2 but doesn't - Stack Overflow 2013-07-08T13:13:12+00:00 http://stackoverflow.com/questions/6379484/fabric-appears-to-start-apache2-but-doesnt jm fabric fail bugs pty ssh automation ops https://pinboard.in/ https://pinboard.in/u:jm/b:f6bbd2033cb6/ Testing Your Automation [slides] 2013-04-08T09:50:19+00:00 https://speakerdeck.com/nathenharvey/testing-your-automation jm devops ops chef automation testing tdd infrastructure provisioning deployment https://pinboard.in/ https://pinboard.in/u:jm/b:a89320ec7528/ KDE's brush with git repository corruption: post-mortem 2013-03-24T20:26:16+00:00 http://jefferai.org/2013/03/24/too-perfect-a-mirror/ jm while we planned for the case of the server losing a disk or entirely biting the dust, or the total loss of the VM’s filesystem, we didn’t plan for the case of filesystem corruption, and the way the corruption affected our mirroring system triggered some very unforeseen and pathological conditions. [...] the corruption was perfectly mirrored... or rather, due to its nature, imperfectly mirrored. And all data on the anongit [mirrors] was lost. One risk demonstrated: by trusting in mirroring, rather than a schedule of snapshot backups covering a wide time range, they nearly had a major outage. Silent data corruption, and code bugs, happen -- backups protect against this, but RAID, replication, and mirrors do not. Another risk: they didn't have a rate limit on project-deletion, which resulted in the "anongit" mirrors deleting their (safe) data copies in response to the upstream corruption. Rate limiting to sanity-check automated changes is vital. What they should have had in place was described by the fix: 'If a new projects file is generated and is more than 1% different than the previous file, the previous file is kept intact (at 1500 repositories, that means 15 repositories would have to be created or deleted in the span of three minutes, which is extremely unlikely).']]> rate-limiting case-studies post-mortems kde git data-corruption risks mirroring replication raid bugs backups snapshots sanity-checks automation ops https://pinboard.in/u:jm/b:6ba6f59d2552/ OmniTI's Experiences Adopting Chef 2013-01-14T14:01:53+00:00 http://omniti.com/seeds/seeds-our-experiences-with-chef-adoption-challenges jm chef deployment ops omniti systems vagrant automation https://pinboard.in/ https://pinboard.in/u:jm/b:07d0ae11b7e7/ Shell Scripts Are Like Gremlins 2012-12-14T14:55:24+00:00 http://blog.brattyredhead.com/blog/2012/12/13/shell-scripts-are-like-gremlins/ jmShell Scripts are like Gremlins. You start out with one adorably cute shell script. You commented it and it does one thing really well. It’s easy to read, everyone can use it. It’s awesome! Then you accidentally spill some water on it, or feed it late one night and omgwtf is happening!? +1. I have to wean myself off the habit of automating with shell scripts where a clean, well-unit-tested piece of code would work better.]]> shell-scripts scripting coding automation sysadmin devops chef deployment https://pinboard.in/ https://pinboard.in/u:jm/b:3ed33353d153/ What can data scientists learn from DevOps? 2012-11-08T15:29:33+00:00 http://redmonk.com/dberkholz/2012/11/06/what-can-data-scientists-learn-from-devops/ jm via:fintanr data-science data automation devops analytics analysis https://pinboard.in/ https://pinboard.in/u:jm/b:87adce97bc7f/ High-frequency trading: The fast and the furious | The Economist 2012-08-03T15:34:00+00:00 http://www.economist.com/node/21547988 jm "The NYMEX panel found that Infinium had finished writing the algorithm only the day before it introduced it to the market, and had tested it for only a couple of hours in a simulated trading environment to see how it would perform. The firm's normal testing processes take six to eight weeks. When the algorithm started its frenetic buying spree, the measures designed to shut it down automatically did not work. One was supposed to turn the system off if a maximum order size was breached, but because the machine was placing lots of small orders rather than a single big one the shutdown was not triggered. The other measure was meant to prevent Infinium from selling or buying more than a certain number of contracts, but because of an error in the way the rogue algorithm had been written, this, too, failed to spot a problem."
]]> hft automation trading markets stocks nymex bugs software https://pinboard.in/ https://pinboard.in/u:jm/b:647442de3e15/ Air France 447 Flight-Data Recorder Transcript - What Really Happened Aboard Air France 447 - Popular Mechanics 2011-12-08T15:55:06+00:00 http://www.popularmechanics.com/technology/aviation/crashes/what-really-happened-aboard-air-france-447-6611877 jm aviation crash flight flying autopilot stalls warnings alarms ui af447 risks automation https://pinboard.in/ https://pinboard.in/u:jm/b:cc02b8859dac/ Gerrit, Git and Jenkins 2011-02-10T14:09:47+00:00 http://alblue.bandlem.com/2011/02/someday.html jm git coding code-review workflows jenkins gerrit c-i testing automation https://pinboard.in/u:jm/b:e011af356bf3/