The Value of a Liberal Arts Education

With a rather evocative headline, "How a half-educated tech elite delivered us into evil," John Naughton explains how the people involved in tech these days are profoundly and deeply ignorant and incurious about the potential effects of what they are doing.

The Germans have a word for this, "Fachidiot," and Japanese word for this is "専門バカ":

One of the biggest puzzles about our current predicament with fake news and the weaponisation of social media is why the folks who built this technology are so taken aback by what has happened. Exhibit A is the founder of Facebook, Mark Zuckerberg, whose political education I recently chronicled. But he’s not alone. In fact I’d say he is quite representative of many of the biggest movers and shakers in the tech world. We have a burgeoning genre of “OMG, what have we done?” angst coming from former Facebook and Google employees who have begun to realise that the cool stuff they worked on might have had, well, antisocial consequences.

Put simply, what Google and Facebook have built is a pair of amazingly sophisticated, computer-driven engines for extracting users’ personal information and data trails, refining them for sale to advertisers in high-speed data-trading auctions that are entirely unregulated and opaque to everyone except the companies themselves.

The purpose of this infrastructure was to enable companies to target people with carefully customised commercial messages and, as far as we know, they are pretty good at that. (Though some advertisers are beginning to wonder if these systems are quite as good as Google and Facebook claim.) And in doing this, Zuckerberg, Google co-founders Larry Page and Sergey Brin and co wrote themselves licences to print money and build insanely profitable companies.

It never seems to have occurred to them that their advertising engines could also be used to deliver precisely targeted ideological and political messages to voters. Hence the obvious question: how could such smart people be so stupid? The cynical answer is they knew about the potential dark side all along and didn’t care, because to acknowledge it might have undermined the aforementioned licences to print money. Which is another way of saying that most tech leaders are sociopaths. Personally I think that’s unlikely, although among their number are some very peculiar characters: one thinks, for example, of Paypal co-founder Peter Thiel – Trump’s favourite techie; and Travis Kalanick, the founder of Uber.

I would actually argue that some in the tech field are willfully blind because their paycheck depends on this lack of awareness, while others are blind because they feel that they are somehow above such "mundane" concerns.

In either case, they aren't people that we can trust with our future.

People Are Beginning to Come to My Point of View

For some time now, I have said that offensive cyber operations are a bad idea, because in order to hack someone, you are sending them a copy of your payload, and they can then use it themselves.

Ryan Cooper has come to this position as well:

Since August 2016, the National Security Agency has suffered a continual stream of devastating failures. Their internal hacking group, known as Tailored Access Operations (TAO), was breached 15 months ago by hackers calling themselves the "Shadow Brokers," which has been dribbling out the contents of the NSA's most prized hacking tools. The result has been a wave of internet crime — ransomware, lost files, and network attacks that disrupted businesses and cost hundreds of millions of dollars.

And as this New York Times story illustrates, the agency has been completely incapable of figuring out how the breach happened. Their computer networks could have been penetrated, or they could have someone on the inside leaking the tools. But after more than a year, they have not been able to plug the leak. It's long past time the NSA was forced to stop hacking, and to start protecting the American people from the sort of tools they create.

At the time of the leak last year, I speculated that the NSA was exposing the American people to online attack, but I was not prepared for how bad it would be. Several huge ransomware attacks (in which a computer is infiltrated, its hard drive encrypted, and the de-encrypt key held for a bitcoin ransom) using NSA hacking tools have swept the globe, hitting companies like FedEx, Merck, and Mondelez International, as well as hospitals and telecoms in 99 countries.

Cyber weapons are different, because they are implicitly revealed, and available for manufacture and deployment, by the target once they are years.

Would it make sense to send drones after ISIS/ISIL/Daesh/Whatever if in so doing, they would be able to deploy drones against targets in the US?

This is how cyber works.

About F%$#ing Time

It appears that the hagiography of brogrammers and Silicon Valley "Unicorns" has hit its expiration date:

When Danny Greg first moved to San Francisco to work at Github in 2012, he used to get high-fives in the street from strangers when he wore his company hoodie.

These days, unless he’s at an investor event, he’s cautious about wearing branded clothing that might indicate he’s a techie. He’s worried about the message it sends.

Greg is one of many people working in tech who are increasingly self-conscious about how the industry – represented by consumer-facing tech titans like Google, Facebook, Amazon, Apple, Twitter and Uber – is perceived: as underregulated, overly powerful companies filled with wealthy tech bros and “brilliant assholes” with little regard for the local communities they occupy. Silicon Valley has taken over from Wall Street as the political bogeyman of choice, turning tech workers – like it or not – into public ambassadors for the 1%.

“I would never say I worked at Facebook,” said one 30-year-old software engineer who left the company last year to pursue an alternative career. Instead, at dinner parties he would give purposefully vague responses and change the subject. “There’s this song and dance you learn to play because people are quick to judge.”

Like Wall Street before, the tech industry is a justifiable punchbag. “MBA jerks used to go and work for Wall Street, now wealthy white geeks go to Stanford and then waltz into a VC or tech firm.”

Gee, my heart bleeds for these guys.

What Could Possibly Go Wrong?

It appears that tech companies are looking to running portions of towns.

Why do I NOT find this reassuring?

It sounds almost dystopian: One of North America's largest cities handing over a key neighborhood to a tech giant so it can be rebuilt from scratch. That, however, may be exactly what is about to happen. Recently, it was announced that Sidewalk Labs, a division of Google-parent Alphabet, will (pending approval) lead the creation of about a 12-acre district called Quayside in a prime area of Toronto's newly revitalized waterfront area.

But if the plan immediately sounds like the beginning of a sci-fi movie in which things are about to go very wrong, Sidewalk Labs' vision at least is distinctly utopian. Simply put, Google's urban spinoff wants to build a laboratory for how a city should be run, a smart city predicated on tech, data, and everything from self-driving shuttles to garbage sorted by robots — all while shifting its Canadian headquarters to the new district.

This has fail written all over it.

After all, these are the folks who can't write a calculator app that adds 1+2+3.

Consider the Source

Andrew Ross Sorkin, the New York Times's cheerleader for Wall Street and its ilk, has finally noticed that Silicon Valley unicorns valuations are 6 pounds of sh%$ in a 5 pound bag:

Uber is said to be worth $62.5 billion. Airbnb is valued at $31 billion. Elon Musk’s SpaceX Technologies is valued at $21 billion, and Pinterest at $12.3 billion.

Those eye-popping valuations regularly fill articles and water-cooler conversations in Silicon Valley, all under the umbrella of “unicorn” companies — a term for private companies that are said to be worth more than $1 billion. That moniker now applies to at least 135 businesses, making the descriptions of them as unicorns, well, less apt. (Maybe donkeys?) Early investors and employees spend countless hours calculating and recalculating how much their stake is worth.

Here is some bad news for them: Those valuations may be a bit of myth — or perhaps wishful thinking.

In Palo Alto, Calif., just down the road from many of the biggest tech companies and the most influential venture capitalists, a professor at Stanford University has quietly been working on a project to crunch the valuation numbers behind some of these private companies.

Ilya A. Strebulaev and another professor working with him, Will Gornall of the University of British Columbia, have come to a startling conclusion: The average unicorn is worth half the headline price tag that is put out after each new valuation.

And if the special side deals that most unicorn companies offer to certain investors — more on this sleight of hand in a moment — are taken into account, almost half of the companies would fall below the $1 billion threshold.

Of course, Sorkin still cannot bring himself to drop the F-bomb (Fraud), but this is systemic fraud.

I don't just want to see sanity and honesty in Silicon Valley, I want to see venture capitalists frog-marched out of their offices in handcuffs.

An Old Hollywood Story

Here is a gem buried in a discussion about plans for the tech giants to take over the entertainment world:

Several times in conversations with people in Hollywood, I heard the tech people referred to as “dumb money” — the sort of outsiders (in the past, they came from oil, then from finance) who parade through town looking to call the shots. One Hollywood executive who has worked often with tech companies told me: “I wouldn’t say we’ve looked at them with fear, no.”

Yeah, pretty much.

Snark of the Day

Nothing Could Possibly Go Wrong If Elon Musk Is Chosen To Rebuild Puerto Rico’s Power Grid

—Dealbreaker

Count me as dubious as the folks at Dealbreaker.

Another Step Forward for Reaction Engines

Proposed Test Stand


Heat exchanger with miles of tiny tubes

Reaction engines, whose SSTO Skylon concept has been making the rounds for nearly a decade, seems to be getting more support, specifically, it has scored a large contract with DARPA to demonstrate its precooler engine technology:

Reaction Engines has achieved a major breakthrough in the U.S. market with a contract from the U.S. Defense Advanced Research Projects Agency (DARPA). The award covers conducting high-temperature testing of the precooler technology at the heart of its proposed hypersonic air-breathing, combined-cycle Sabre rocket concept.

Reaction has seen interest grow in the U.S. about elements of Sabre, particularly the precooling heat exchanger, ever since the concept was first independently validated by the U.S. Air Force Research Laboratory in 2015 (AW&ST Aug. 3-16, 2015, p. 63). Designed to chill airflow from over 1,800F to −240 F (1,000C to −150C) in less than 1/20th of a second, the heat exchanger is pivotal to the process of extracting oxygen from the air for use by the rocket.

However, the heat exchanger can also be used more generically to precool other engine cycles and reduce heating on engine components in high-speed flight. According to Reaction, the design, dubbed HTX, “could enable new classes of vehicles and operational possibilities.” The precooler will be tested at speeds up to Mach 5 in a new high-temperature airflow evaluation facility to be built in Castle Rock, Colorado, as a base for REI (Reaction Engines Inc.), the U.S. subsidiary of the UK-based company.

It does appear that their technology is getting funding and support, so I'm hoping to see a flight test article in the next 5 years or so.

Previous posts about Reaction Engines here.

Well, This is Great

Did you know that Equifax runs the My Social Security and is responsible for verifying data for Obamacare exchanges for the US government?

You know, that whole, "Reinventing Government", thing that Bill Clinton put forward in the 1990s, when critical government functions were outsourced to private for-profit operators, is looking to be an even worse deal than when it was first implemented in the mid-1990s.

Of course, efficiency and savings were never really the goals: It was a depressingly successful attempt to subvert the civil service laws and to return to the spoils system.

Just ask President Garfield how well that worked out.

I Need to Make My Web Site Tougher More Confusing and More Complex

A study has been made about so-called "Flat" user interfaces, and it turns out that they are 22% slower for users to navigate a site.

You are no doubt aware of it the trend. It's the latest user interface fashion, favored by Apple and Google, among others.

It does boggle the mind that an interface created by Microsoft for its Zune^® music player has taken user interface design by storm.

First, even by the standards of Microsoft, the Zune^® was a miserable failure, and second, who in their right mind would steal user interface ideas from Microsoft?

The only explanation I have for this is a conspiracy theory:  I know from my Google^™ analytics account that the amount of time the average user spends on the site is closely tracked. (They call it "Engagement".)

This means that for advertising supported media, a second wasted is a second monetized.

I'm not actually going to make my site less intuitive, I rely on my ordinary thought processses to confuse my reader(s). It's a matter of pride.

Neither would I suggest that you leave a tab open to my site in the background, as my making sush a suggestion would be a violation of the terms of service of Google^™ Adsense^™.

Libel Troll Fraudster Gets Case Thrown Out of Court

Shiva Ayyadurai claimed to have created email in 1978.

The facts, of course speak otherwise.

Email predates his high school freshman programming exercise by at least 10 years, email actually accounted for over half of all ARPANet traffic two years before he wrote his program, but that didn't stop him from attempting to sue Techdirt out of existence, possibly in collusion with wannabe Bond villain and literal vampire Peter Thiel.

Well, the judge just threw out his whole case.

It's not a complete win for the defendant, because the federal judge did not strike the case under California's anti-SLAPP law, which would have have allowed them to sue for legal fees and penalties, but this is still an unambiguous win:

As you likely know, for most of the past nine months, we've been dealing with a defamation lawsuit from Shiva Ayyadurai, who claims to have invented email. This is a claim that we have disputed at great length and in great detail, showing how email existed long before Ayyadurai wrote his program. We pointed to the well documented public history of email, and how basically all of the components that Ayyadurai now claims credit for preceded his own work. We discussed how his arguments were, at best, misleading, such as arguing that the copyright on his program proved that he was the "inventor of email" -- since patents and copyrights are very different, and just because Microsoft has a copyright on "Windows" it does not mean it "invented" the concept of a windowed graphical user interface (because it did not). As I have said, a case like this is extremely draining -- especially on an emotional level -- and can create massive chilling effects on free speech.

A few hours ago, the judge ruled and we prevailed. The case has been dismissed and the judge rejected Ayyadurai's request to file an amended complaint. We are certainly pleased with the decision and his analysis, which notes over and over again that everything that we stated was clearly protected speech, and the defamation (and other claims) had no merit. This is, clearly, a big win for the First Amendment and free speech -- especially the right to call out and criticize a public figure such as Shiva Ayyadurai, who is now running for the US Senate in Massachusetts. We're further happy to see the judge affirm that CDA Section 230 protects us from being sued over comments made on the blog, which cannot be attributed to us under the law. We talk a lot about the importance of CDA 230, in part because it protects sites like our own from these kinds of lawsuits. This is just one more reason we're so concerned about the latest attempt in Congress to undermine CDA 230. While those supporting the bill may claim that it only targets sites like Backpage, such changes to CDA 230 could have a much bigger impact on smaller sites like our own.

We are disappointed, however, that the judge denied our separate motion to strike under California's anti-SLAPP law. For years, we've discussed the importance of strong anti-SLAPP laws that protect individuals and sites from going through costly legal battles. Good anti-SLAPP laws do two things: they stop lawsuits early and they make those who bring SLAPP suits -- that is, lawsuits clearly designed to silence protected speech -- pay the legal fees. The question in this case was whether or not California's anti-SLAPP law should apply to a case brought in Massachusetts. While other courts have said that the state of the speaker should determine which anti-SLAPP laws are applied (even in other states' courts), it was an issue that had not yet been ruled upon in the First Circuit where this case was heard. While we're happy with the overall dismissal and the strong language used to support our free speech rights, we're nevertheless disappointed that the judge chose not to apply California's anti-SLAPP law here.

This guy is running for Senate in Massachusetts, as a Republican, and he gave a speech at the recent white supremacist rally in Boston.

He also claims that anyone who knows the history of email is a racist.

What a lovely fellow.

Why We Hate Them

I am referring to Comcast, of course, who is now suing the state of Vermont for the audacity to insist that they keep their promises:

Comcast has sued the state of Vermont to try to avoid a requirement to build 550 miles of new cable lines.

Comcast's lawsuit against the Vermont Public Utility Commission (VPUC) was filed Monday in US District Court in Vermont and challenges several provisions in the cable company's new 11-year permit to offer services in the state. One of the conditions in the permit says that "Comcast shall construct no less than 550 miles of line extensions into un-cabled areas during the [11-year] term."

Comcast would rather not do that. The company's court complaint says that Vermont is exceeding its authority under the federal Cable Act while also violating state law and Comcast's constitutional rights:

The VPUC claimed that it could impose the blanket 550-mile line extension mandate on Comcast because it is the "largest" cable operator in Vermont and can afford it. These discriminatory conditions contravene federal and state law, amount to undue speaker-based burdens on Comcast's protected speech under the First Amendment of the United States Constitution... and deprive Comcast and its subscribers of the benefits of Vermont law enjoyed by other cable operators and their subscribers without a just and rational basis, in violation of the Common Benefits Clause of the Vermont Constitution.

Rival providers Charter and Burlington Telecom don't have to comply with these special requirements, Comcast said. Instead, the other companies "need only comply with the non-discriminatory line extension policies" established in a VPUC rule.

Comcast's complaint also objected to several other requirements in the permit, including "unreasonable demands" for upgrades to local public, educational, and governmental (PEG) access channels and the building of "institutional networks ("I-Nets") to local governmental and educational entities upon request and on non-market based terms."

………

Comcast often refuses to extend its network to customers outside its existing service area unless the customers pay for Comcast's construction costs, which can be tens of thousands of dollars.

Comcast is regularly at the top of the list on the most loathed company in America, but somehow or other, they continue to flourish.

There is some sort of profound market failure going on.

You Ever Notice What Happens When the Silicon Valley Mindset Meets the Real World?

First Theranos, and now Juicero.

In all fairness, Theranos has a harder job, they were actually in the healthcare business, which involved deceiving regulators, which is hard, while Juicero was just a f%$#ing juice machine:

It sounds like America’s favorite $400 juice machine will be no longer.

“After selling over a million Produce Packs, we must let you know that we are suspending the sale of the Juicero Press and Produce Packs immediately,” reads the company blog post.

Juicero will also be giving people money back. “For the next 90 days, we are offering refunds for your purchase of the Juicero Press,” according to the note.

Founded by Doug Evans, San Francisco-based Juicero had raised more than $118 million in funding from prominent VCs like Google Ventures and Kleiner Perkins. Carmelo Anthony also invested through his Melo7 Tech venture fund. Even The Campbell Soup Company threw money at it. Juicero started raising funding in 2013 and launched 16 months ago.

The company was subject to mockery, particularly after a Bloomberg piece showed that the juice packets could be squeezed by hand and did not require a fancy machine.

The emperor has no clothes.

If regulators and prosecutors did a deep dive on Silicon Valley, those that weren't in prison would be asking, "Do you want fries with that?"

I'm Sorry Dave, I'm Afraid I Can't Do That

What a surprise.

It turns that it is trivial to hack the most sophisticated Artificial Intelligence (AI) systems by simply training them poorly:

If you don't know what your AI model is doing, how do you know it's not evil?

Boffins from New York University have posed that question in a paper at arXiv, and come up with the disturbing conclusion that machine learning can be taught to include backdoors, by attacks on their learning data.

The problem of a “maliciously trained network” (which they dub a “BadNet”) is more than a theoretical issue, the researchers say in this paper: for example, they write, a facial recognition system could be trained to ignore some faces, to let a burglar into a building the owner thinks is protected.

The assumptions they make in the paper are straightforward enough: first, that not everybody has the computing firepower to run big neural network training models themselves, which is what creates an “as-a-service” market for machine learning (Google, Microsoft and Amazon all have such offerings in their clouds); and second, that from the outside, there's no way to know a service isn't a “BadNet”.

Note that current high end AI models are not so much programmed as trained, and it appears that this provides an unprecedented opportunity to develop malicious software.

I'm thinking that you might see an AI drone that gets the whole Manchurian Candidate treatment in the not too distant future.

Seriously Neat Tech

Spider silk has some remarkable mechanical properties. Its low weight, high strength, an low modulus make it a potential break through in body armor.

The problem is spiders, which produce the material in very limited quantities, and their cannibalistic proclivities mitigate against high intensity farming.

Well, it looks like the DoD is trying to get silk genetically engineered worms to produce spider silk an alternative:

The U.S. Army is upping its investment in genetically engineered spider silk for body armor. Last year, the service paid almost $100,000 to Kraig Biocraft Laboratories, which makes spider silk that can be produced at scale — with silkworms. On Wednesday, the company announced that the Army will move to the second phase of the contract and will look to Kraig to produce a customized strain of the silk for “high-performance fibers for protective apparel applications.” That is: flexible body armor made from genetically engineered spider silk. The total contract amount would reach $900,000 if parameters are met. Army representatives said that interested in the material purely from a research perspective, for now.

Kraig Biocraft injects spider DNA into silkworm eggs, enabling the worms to produce its custom silk. The researchers describe the process in this 2011 PNAS paper.

Spider silk is much tougher than regular worm silk, and about half as tough as Kevlar. But it’s far more flexible, (3 percent elasticity for kevlar versus nearly 40 percent for spider silk.) The Army believes that the energy absorption of the material could be much higher than kevlar (as determined by multiplying the strength of the fiber by the elongation.)

It’s also much more elastic and flexible than kevlar. But getting enough spider silk to clothe an Army is a tall order. The crawly arachnids don’t produce silk in high volume and when you crowd spiders too close together, they eat each other. The quest to produce spider silk in hosts other than spiders has led researchers to use a variety of other methods such as yeast, e. coli bacteria and mammalian cells. 

There is also the fact that the techniques for handling silkworms, and harvesting the silk, have been known for thousands of years.

It is also far less alarming than the prospect of an escape of motherf%$#ing mutant spiders.

Like This Guy is Going to Save the World

For 12 years Elon Musk had a personal assistant. When she asked for a raise, he told her to take a vacation so he could see how essential her services were.

When she came back, he he told her that her services were not required, and he fired her. (See also here and here)

Elon Musk is a busy billionaire who you might imagine has a lot of people working for him to keep all of his various projects afloat, but Business Insider brought up an anecdote from a biography of Musk today about how he once fired his long-time assistant for asking for a raise. (Update: Elon Musk has denied the biography’s account on Twitter.)

Look, obviously being the super rich guy behind some of the most influential American companies in recent history is going to make you feel important and probably turn into kind of a jerk. But this anecdote is almost something else, so here it is, as summarized by Business Insider from Musk’s biography, Elon Musk: Tesla, SpaceX, and the Quest for a Fantastic Future, by author Ashlee Vance:

According to Vance, the assistant, Mary Beth Brown, asked Musk for a significant raise after she’d been working with him for 12 years. In response, Musk told Brown to take two weeks off, during which he would assume her responsibilities and see whether she was critical to his success.

When Brown returned, Musk told her he didn’t need her anymore.

Musk also told Vance that he offered Brown another position at the company but that she never returned to the office after that.

Now, of course, this is just an anecdote in a biography, so the telling of what happened may not be exactly as it actually happened. Does it sound like this is something Musk would do? Yeah. Does anyone think Musk can actually operate without an assistant? You probably shouldn’t.

But! After some brief digging to see if I could find something else, I found a Quora posting from the verified profile of Justine Musk, ex-wife of Elon, that practically doubles down on Mary Beth Brown’s story, as well as how insane it is to have any sort of relationship with Elon Musk:

Mary Beth Brown started working for Elon soon after we moved to LA twelve or thirteen years ago (Elon and I were still married then). MB was an exceptional and devoted employee of Elon’s and lovely to deal with on a personal level. She gave her life to the job — and to our family — and the news of her departure was a shock to me.

Apparently (according, I believe, to Ashlee Vance, who wrote the book on Elon), MB asked for a raise. E told her that if she was truly critical to SpaceX, it should not be able to operate in her absence (or something to that effect). He suggested a 3-week experiment to test this hypothesis/her worth. This reminds me of something similar he once said to me, many years ago, after I came back from a week’s visit with my family in Canada — that his life had operated quite smoothly in my absence. He was letting me know that I was an incompetent house manager. (He was not wrong.) So of the different stories I have heard behind MB’s departure from SpaceX, this is the one that resonates with me. (Although you would *never* use the word ‘incompetent’ in association with MB!)

Wherever MB is and whatever she is doing, I hope her life is fabulous. She and I have not always been on the same side, but I have — and will always have — tremendous respect for that remarkable woman.

Abraham Lincoln once said , "I care not much for a man's religion whose dog and cat are not the better for it."

By this standard, Elon Musk's cult of well ……… Elon Musk  ……… severely lacking.

This is Profoundly Weird

Marcus Hutchins, a white hat hacker who shut down the WannaCry ransomware, was just arrested by the FBI, and charged with creating and distributing a banking Trojan 3 years ago:

On Wednesday, US authorities detained a researcher who goes by the handle MalwareTech, best known for stopping the spread of the WannaCry ransomware virus.

In May, WannaCry infected hospitals in the UK, a Spanish telecommunications company, and other targets in Russia, Turkey, Germany, Vietnam, and more. Marcus Hutchins, a researcher from cybersecurity firm Kryptos Logic, inadvertently stopped WannaCry in its tracks by registering a specific website domain included in the malware's code.

Hutchins was arrested for allegedly creating the Kronos banking malware.

Motherboard verified that a detainee called Marcus Hutchins, 23, was being held at the Henderson Detention Center in Nevada early on Thursday. A few hours after, Hutchins was moved to another facility, according to a close personal friend.

The friend told Motherboard they "tried to visit him as soon as the detention centre opened but he had already been transferred out." Motherboard granted the source anonymity due to privacy concerns.

"I've spoken to the US Marshals again and they say they have no record of Marcus being in the system. At this point we've been trying to get in contact with Marcus for 18 hours and nobody knows where he's been taken," the person added. "We still don't know why Marcus has been arrested and now we have no idea where in the US he's been taken to and we're extremely concerned for his welfare."

So, they have arrested him, and are holding him incommunicado, and at this time it appears that he has not been allowed to talk to a lawyer.

Also note that "MalWareTech" seemed to confirm that the WannaCry code originated with the NSA, which might imply that there some institutional imperative to go after him that was not strictly judicial.

Also, at the time of the Kronos release, Marcus Hutchins was casting about on Twitter for a copy of the code, which seems to an awfully odd thing to do if he wrote the code in the first place:

Anyone got a kronos sample?

— MalwareTech (@MalwareTechBlog) July 13, 2014

Marcy Wheeler also noticed an odd coincidence that corresponded to his arrest:

In remarkably timed news, between 3:10 and 3:25 AM UTC this morning (8 PM last night Mountain Time), someone emptied out all the WannaCry accounts.

So, while Hutchins was detained, someone took all the ransom money that 

This is all profoundly odd.

Throw Your Amazon Echo out the Window Now

Amazon is looking at giving app developers access to audio recordings from its Alexa.

Such a good idea to give access to every conversation in your room to Russian hackers:

The data is also kept in the event it’s request by law enforcement, however Amazon fought police over what it saw as an overly broad request for audio logs on a murder suspect last year. (The company relented in April of this year and handed over the logs when the suspect voluntarily said he was willing to provide them.)

Amazon does not hand this data over to developers, The Information says, because such a move would undermine Amazon’s commitment to user privacy. However, because Google, which makes the most popular Echo competitor currently on the market, does give developers access to this data, Amazon’s Echo and Alexa divisions feel they are at a disadvantage, the report states. Google did not immediately respond to a request for comment on its data-sharing policies for the Home speaker.

For instance, some developers fear that without audio logs of requests, like those related to a food delivery order, they won’t know exactly what went wrong if the order is ultimately incorrect and the customer unhappy. According to The Information, Amazon does give some data over to a select few “whitelisted” developers, though how that system works is unclear. Amazon is considering granting third-party app developers access to transcripts of audio recordings saved by Alexa-powered devices, according to a report from The Information today. The change would be aimed at enticing developers to continue investing in Alexa as a voice assistant platform, by giving those app makers more data that could help improve their software over time. Amazon’s goal, according to The Information, is to stay competitive with more recent entrants in the smart speaker market, like Apple and Google.

Amazon declined to comment on its future plans for Alexa data-sharing policies. However, a company spokesperson told The Verge, “When you use a skill, we provide the developer the information they need to process your request. We do not share customer identifiable information to third-party skills without the customer’s consent. We do not share audio recordings with developers.”

As it stands today, Amazon records audio through Alexa devices like the Echo home speaker and the new Echo Look camera and Echo Show monitor, however only after a “wake word” like “Hey Alexa” is used to prime the software. These devices send these audio clips to an Amazon-owned server where they’re analyzed to produce a near-instantaneous response from Alexa, but where they’re also stored so Amazon can improve its digital assistant through artificial intelligence training techniques.

………

Amazon does not hand this data over to developers, The Information says, because such a move would undermine Amazon’s commitment to user privacy. However, because Google, which makes the most popular Echo competitor currently on the market, does give developers access to this data, Amazon’s Echo and Alexa divisions feel they are at a disadvantage, the report states. Google did not immediately respond to a request for comment on its data-sharing policies for the Home speaker.

For instance, some developers fear that without audio logs of requests, like those related to a food delivery order, they won’t know exactly what went wrong if the order is ultimately incorrect and the customer unhappy. According to The Information, Amazon does give some data over to a select few “whitelisted” developers, though how that system works is unclear.

Yeah, throw out Google Home as well.

Orwell in a f%$#ing box.

This Is What Comes from Mixing Ambien and Alcohol

A little red wine, vintage record, some Ambien ... and magic!

— Elon Musk (@elonmusk) June 7, 2017

Like Keith Moon, and not in a good way

The increasingly erratic Elon Musk is now claiming that he got verbal governmental approval to construct a Hyperloop transit system from Washington, DC and New York City.

Rather unsurprisingly all manner of public officials over the proposed route have no recollection of any such discussion:

Elon Musk does not have government approval to build a Hyperloop tunnel from New York City to Washington DC.

The Tesla executive took to Twitter this morning to tantalize his legion of fans and the tech press with the “news” that he had “just received verbal govt approval for The Boring Company to build an underground NY-Phil-Balt-DC Hyperloop. NY-DC in 29 mins … City center to city center in each case, with up to a dozen or more entry/exit elevators in each city.”

Lest any billionaires need to brush up on civics 101: the US system of government does not operate on “verbal government approvals”.

Musk walked back his claim about 90 minutes later, tweeting: “Still a lot of work needed to receive formal approval, but am optimistic that will occur rapidly”.

A lot of work is needed to receive formal approval, indeed.

Musk was received with typical credulity by the tech press, and considerable consternation by various government agencies. Several spokespeople who answered the phones at relevant city, state and federal government bodies laughed upon hearing of the claim that an interstate transit project with a significant street-level footprint in four of the east coast’s largest cities could be approved verbally.

Gee, you think?

I'm not sure if Musk has gone over the edge because of drug abuse, Ambien and alcohol is addictive and dangerous, or if it's because he has become convinced of his own epic awesomeness from sycophants and the, "Typical credulity by the tech press," but Elon Musk is well into delusional, and seems to be heading to Bond villain territory.

I'm kind of surprised.  Of all the PayPal mafia, I would have figured that would be vampire Peter Thiel would go over the edge first.

If You Can't Beat Them, Join Them

Reverse flow operation (starts at 45s)


See also this diagram


The GE Engine

GE will be attempting to challenge Pratt & Whitney in the turboprop market, where the PT6 turboprop completely dominates the market.

Of interest to me is that GE will be copying the basic operating principals of the PT6,(paid subscription required) P&W's reverse flow operation.

The inlet is at the back of the engine, and air flows forward. This allows the compressor and the compressor turbine to be completely separate from the power turbine and prop.

While the air flow is rather more circuitous than that of a straight through engine, it has a number of significant advantages:

• No need for concentric shafts while still maintaining a two spool compressor and turbine.
• A smaller and lighter starter motor.
• More easily adopted to different power levels.
• Greater simplicity and reliability.

The PT6 has used this formula to completely dominate the market, and it looks like GE will be aping their approach, with a lot of additive manufacturing throw into the mix:

………

Then there’s the ATP, GE’s Advanced Turboprop engine (see photo, above). This is a very big deal in terms of technology and targeting.

For those cloistered monks among you, some background: Pratt & Whitney Canada’s PT6 family has reigned supreme among turboprops since, well, forever. And for good reason. The type ranges in power from 500-2,000 shp and has demonstrated rock-solid reliability through decades of operation. Its bulletproof reputation is the reason almost all single-turboprop-powered aircraft—from the Piper’s owner-flown M600 to Beechcraft’s PT-6 Texan II military trainer to the do-it-all Pilatus PC-12—are fitted with a PT-6. More than 51,000 PT6s have been produced since the engine’s introduction in the 1960s. It has been expanded to include 69 versions that power some 100 different aircraft models, including all production King Airs.

GE hopes the ATP will break Pratt’s near monopoly. Developed at the company’s “turboprop center of excellence” in Prague with a $400 million investment, this, the world’s most “printed” engine (additive manufacturing has replaced 855 parts with a mere dozen 3D-printed components)features a single-lever integrated engine and propeller control, 16:1 pressure ratio, reverse-flow combustor and output of 850-1,650 shp. The design promises 20% better fuel burn, 10% more power and longer maintenance intervals than you know what.

This is, in its own way, a tribute to the genius of the design team that first devised the PT6.