10 September 2019

Apple’s new A13 chip is faster according to charts with no numbers


Apple is announcing new iPhone models today. The iPhone 11 uses an Apple A13 Bionic system-on-a-chip. It is faster than the A12 Bionic in the iPhone XR, XS and XS Max.

But how much faster exactly? According to Apple, Apple is making the fastest GPU and CPU for a smartphone.

Worse, the company showed two charts with no X-axis. With this chart, CPU performance of the A13 Bionic could be 2% faster or 248% faster than CPU performance of the A12 Bionic. The same thing applies for GPU performance. I guess we’ll have to wait for benchmarks.

It reminds me of another technology company that is well-known for its charts with no numbers…


Read Full Article

Apple introduces the Apple Watch Series 5 with always-on display


Apple introduced some new Apple Watch models at a press conference. The Apple Watch Series 5 has an always-on display. It seems to look just like the Apple Watch Series 4.

“Apple Watch puts groundbreaking health, fitness and communication capabilities on the wrist of millions and millions of people,” Apple CEO Tim Cook said. He then introduced a video segment showing how Apple Watch users are healthier.

The Apple Watch automatically adjusts the brightness of the new always-on display. When you lower your wrist, the brightness goes down. It features an LTPO display with an adaptive refresh rate. It can go down to 1Hz, or one screen refresh per second. That’s how Apple can reach 18 hours of battery life with a display that stays on.

The new Apple Watch also features a built-in compass. There’s a new app that tells you your latitude, longitude and direction. It could be particularly useful when you’re hiking.

When it comes to emergency calling, Apple is extending emergency calling to 150 countries. When you press and hold down the side button, it automatically calls local emergency services.

Aluminum models come in silver, gold and space gray. Those cases are now made from recycled aluminum. Stainless steel models come in gold, space black and and polished.

And finally, there are two new titanium models (brushed and brushed space black) and a ceramic model. Apple is refreshing special editions of the Apple Watch with Nike and Hermès.

Apple Watch Series 5 with a GPS starts at $399. For $499, you also get a cellular modem. Pre-orders start today and they will be available on September 20. The Apple Watch Series 3 first introduced in 2017 now starts at $199.

Apple also announced three new health research studies with health facilities. Apple is starting a hearing study with the World Health Organization and the University of Michigan, a women’s health study with NIH and Harvard thanks to the new cycle tracking feature, and a heart and movement study with the American Heart Association and Brigham and Women's Hospital.


Read Full Article

The Polaroid Lab uses the light from your phone’s screen to turn digital photos into Polaroids


When all of us are carrying phones that can snap a thousand photos a minute and are connected to cloud systems that can store millions, there’s an undeniable charm to physical photos. The ones deemed worthy; the ones so special that they must be transformed from bit to atom.

While photo printers are nothing new, Polaroid is twisting up the concept (and rebooting an idea from a few years back) with the “Polaroid Lab”. It’s a $129 tower that uses the light from your phone’s screen, bounced off a series of mirrors, to make a proper Polaroid from the photos you’ve already taken.

Open your photo in Polaroid’s companion app, place your phone (any iPhone after the 6S, and ‘current models of Samsung, Huawei, Google Pixel, and One Plus’ Android handsets) on top of the tower, and push the red button. A few seconds later, out pops a grey Polaroid. Did it work? You’ll have to wait a few minutes for it to develop, just like the good (?) ol’ days.

Is using light and mirrors better than just sending a picture to a printer over Bluetooth or WiFi and blasting the ink out from a cartridge? Maybe not. But it’s neat! It’s physical and sciencey and fun — and, arguably, as close as you can get to having a “true” Polaroid picture of a moment that’s already happened.

The company says that the Polaroid Lab works with its existing I-Type and 600 series films… which, as any enthusiast could tell you, doesn’t come cheap. Expect each photo printed here to cost you a buck or two. That’s a bit steeper than many at-home printers and definitely pricier than just blasting out some 4x6s at Costco, but this thing will almost certainly still find its audience amongst those going for a certain look.

There’s also a way to “blow up” one photo across a bunch of Polaroids, if you’ve got the film to spare. Here’s a demo video of what that looks like:

If the whole concept seems familiar, you might be remembering the Impossible Instant Lab — a product of a veeeery similar vein that raised over half a million dollars on Kickstarter back in 2012. The Impossible Instant Lab was discontinued in July of 2017… just a few months after the team behind it acquired the rights to the Polaroid brand. This seems to be a reboot of the concept, now with the added weight and officialness of the Polaroid name thrown behind it.

The team behind the Polaroid Lab says it should hit the shelves by October 10th.


Read Full Article

Google brings Cloud Dataproc to Kubernetes


Cloud Dataproc is probably one of the lesser-known products in Google Cloud’s portfolio, but it’s a powerful tool for data wranglers who are looking for a fully managed cloud service that lets them run Apache Spark and Hadoop clusters without having to worry about managing the underlying infrastructure. Today. Google announced that it is launching the alpha of Cloud Dataproc to Kubernetes — and while that, too, may not sound all that interesting at first, it’s an important step for Google Cloud as it works to adapt more of its products to a hybrid cloud model.

The general idea here is to give enterprise customers (and make no mistake, enterprise customers are the main focus of Google Cloud these days) the ability to run Apache Spark jobs on Google Kubernetes Engine (GKE) clusters. With products like Anthos now making GKE available virtually anywhere, this means customers can now also take Cloud Dataproc to their own data centers. Right now, the service only supports Apache Spark, but Google plans to support other open-source projects, too.

“Enterprises are increasingly looking for products and services that support data processing across multiple locations and platforms,” said Matt Aslett, Research Vice President at 451 Research. “The launch of Cloud Dataproc on Kubernetes is significant in that it provides customers with a single control plane for deploying and managing Apache Spark jobs on Google Kubernetes Engine in both public cloud and on-premises environments.”

Typically, Spark applications run on Hadoop YARN clusters. Google notes that the Cloud Dataproc on Kubernetes will free users from having to use two cluster management systems and will give them a single view across both YARN and Kubernetes clusters. “Supporting both YARN and Kubernetes can bring your enterprise the needed flexibility to modernize certain hybrid workloads while continuing to monitor YARN-based workloads,” the company writes in today’s announcement.

The new service is now available as an alpha. If you want to give it a try, you’ll have to apply for access by emailing Google.


Read Full Article

Recursive Sketches for Modular Deep Learning




Much of classical machine learning (ML) focuses on utilizing available data to make more accurate predictions. More recently, researchers have considered other important objectives, such as how to design algorithms to be small, efficient, and robust. With these goals in mind, a natural research objective is the design of a system on top of neural networks that efficiently stores information encoded within—in other words, a mechanism to compute a succinct summary (a “sketch”) of how a complex deep network processes its inputs. Sketching is a rich field of study that dates back to the foundational work of Alon, Matias, and Szegedy, which can enable neural networks to efficiently summarize information about their inputs.

For example: Imagine stepping into a room and briefly viewing the objects within. Modern machine learning is excellent at answering immediate questions, known at training time, about this scene: “Is there a cat? How big is said cat?” Now, suppose we view this room every day over the course of a year. People can reminisce about the times they saw the room: “How often did the room contain a cat? Was it usually morning or night when we saw the room?”. However, can one design systems that are also capable of efficiently answering such memory-based questions even if they are unknown at training time?

In “Recursive Sketches for Modular Deep Learning”, recently presented at ICML 2019, we explore how to succinctly summarize how a machine learning model understands its input. We do this by augmenting an existing (already trained) machine learning model with “sketches” of its computation, using them to efficiently answer memory-based questions—for example, image-to-image-similarity and summary statistics—despite the fact that they take up much less memory than storing the entire original computation.

Basic Sketching Algorithms
In general, sketching algorithms take a vector x and produce an output sketch vector that behaves like x but whose storage cost is much smaller. The fact that the storage cost is much smaller allows one to succinctly store information about the network, which is critical for efficiently answering memory-based questions. In the simplest case, a linear sketch x is given by the matrix-vector product Ax where A is a wide matrix, i.e., the number of columns is equal to the original dimension of x and the number of rows is equal to the new reduced dimension. Such methods have led to a variety of efficient algorithms for basic tasks on massive datasets, such as estimating fundamental statistics (e.g., histogram, quantiles and interquartile range), finding popular items (known as frequent elements), as well as estimating the number of distinct elements (known as support size) and the related tasks of norms and entropy estimation.
A simple method to sketch the vector x is to multiply it by a wide matrix A to produce a lower-dimensional vector y.
This basic approach works well in the relatively simple case of linear regression, where it is possible to identify important data dimensions simply by the magnitude of weights (under the common assumption that they have uniform variance). However, many modern machine learning models are actually deep neural networks and are based on high-dimensional embeddings (such as Word2Vec, Image Embeddings, Glove, DeepWalk and BERT), which makes the task of summarizing the operation of the model on the input much more difficult. However, a large subset of these more complex networks are modular, allowing us to generate accurate sketches of their behavior, in spite of their complexity.

Neural Network Modularity
A modular deep network consists of several independent neural networks (modules) that only communicate via one’s output serving as another’s input. This concept has inspired several practical architectures, including Neural Modular Networks, Capsule Neural Networks and PathNet. It is also possible to split other canonical architectures to view them as modular networks and apply our approach. For example, convolutional neural networks (CNNs) are traditionally understood to behave in a modular fashion; they detect basic concepts and attributes in their lower layers and build up to detecting more complex objects in their higher layers. In this view, the convolution kernels correspond to modules. A cartoon depiction of a modular network is given below.
This is a cartoon depiction of a modular network for image processing. Data flows from the bottom of the figure to the top through the modules represented with blue boxes. Note that modules in the lower layers correspond to basic objects, such as edges in an image, while modules in upper layers correspond to more complex objects, like humans or cats. Also notice that in this imaginary modular network, the output of the face module is generic enough to be used by both the human and cat modules.
Sketch Requirements
To optimize our approach for these modular networks, we identified several desired properties that a network sketch should satisfy:
  • Sketch-to-Sketch Similarity: The sketches of two unrelated network operations (either in terms of the present modules or in terms of the attribute vectors) should be very different; on the other hand, the sketches of two similar network operations should be very close.
  • Attribute Recovery: The attribute vector, e.g., the activations of any node of the graph can be approximately recovered from the top-level sketch.
  • Summary Statistics: If there are multiple similar objects, we can recover summary statistics about them. For example, if an image has multiple cats, we can count how many there are. Note that we want to do this without knowing the questions ahead of time.
  • Graceful Erasure: Erasing a suffix of the top-level sketch maintains the above properties (but would smoothly increase the error).
  • Network Recovery: Given sufficiently many (input, sketch) pairs, the wiring of the edges of the network as well as the sketch function can be approximately recovered.
This is a 2D cartoon depiction of the sketch-to-sketch similarity property. Each vector represents a sketch and related sketches are more likely to cluster together.
The Sketching Mechanism
The sketching mechanism we propose can be applied to a pre-trained modular network. It produces a single top-level sketch summarizing the operation of this network, simultaneously satisfying all of the desired properties above. To understand how it does this, it helps to first consider a one-layer network. In this case, we ensure that all the information pertaining to a specific node is “packed” into two separate subspaces, one corresponding to the node itself and one corresponding to its associated module. Using suitable projections, the first subspace lets us recover the attributes of the node whereas the second subspace facilitates quick estimates of summary statistics. Both subspaces help enforce the aforementioned sketch-to-sketch similarity property. We demonstrate that these properties hold if all the involved subspaces are chosen independently at random.

Of course, extra care has to be taken when extending this idea to networks with more than one layer—which leads to our recursive sketching mechanism. Due to their recursive nature, these sketches can be “unrolled” to identify sub-components, capturing even complicated network structures. Finally, we utilize a dictionary learning algorithm tailored to our setup to prove that the random subspaces making up the sketching mechanism together with the network architecture can be recovered from a sufficiently large number of (input, sketch) pairs.

Future Directions
The question of succinctly summarizing the operation of a network seems to be closely related to that of model interpretability. It would be interesting to investigate whether ideas from the sketching literature can be applied to this domain. Our sketches could also be organized in a repository to implicitly form a “knowledge graph”, allowing patterns to be identified and quickly retrieved. Moreover, our sketching mechanism allows for seamlessly adding new modules to the sketch repository—it would be interesting to explore whether this feature can have applications to architecture search and evolving network topologies. Finally, our sketches can be viewed as a way of organizing previously encountered information in memory, e.g., images that share the same modules or attributes would share subcomponents of their sketches. This, on a very high level, is similar to the way humans use prior knowledge to recognize objects and generalize to unencountered situations.

Acknowledgements
This work was the joint effort of Badih Ghazi, Rina Panigrahy and Joshua R. Wang.

A "living drug" that could change the way we treat cancer | Carl June

A "living drug" that could change the way we treat cancer | Carl June

Carl June is the pioneer behind CAR T-cell therapy: a groundbreaking cancer treatment that supercharges part of a patient's own immune system to attack and kill tumors. In a talk about a breakthrough, he shares how three decades of research culminated in a therapy that's eradicated cases of leukemia once thought to be incurable -- and explains how it could be used to fight other types of cancer.

Click the above link to download the TED talk.

Europe’s antitrust chief, Margrethe Vestager, set for expanded role in next Commission


As the antitrust investigations stack up on US tech giants’ home turf there’s no sign of pressure letting up across the pond.

European Commission president-elect Ursula von der Leyen today unveiled her picks for the next team of commissioners who will take up their mandates on November 1 — giving an expanded role to competition commissioner Margrethe Vestager. The pick suggests the next Commission is preparing to dial up its scrutiny of big tech’s data monopolies.

Under the draft list of commissioners-designate, which still needs to be approved in full by the European Parliament, Vestager has been named executive VP overseeing a new portfolio called ‘Europe fit for the digital age’.

But, crucially, she will also retain the competition portfolio — which implies attention on growing Europe’s digital economy will go hand in glove with scrutiny of fairness in ecommerce and ensuring a level playing field vs US platform giants.

“Executive vice-president Margrethe Vestager will lead our work on a Europe fit for the digital age,” said von der Leyen at a press conference to announce her picks. “Digitalization has a huge impact on the way we live, we work, we communicate. In some fields Europe has to catch up — for example in the field of business to consumer but in other fields we’re excellent. Europe is the frontrunner, for example in business to business, when we talk about digital twins of products and procedures.

“We have to make more out of the field of artificial intelligence. We have to make our single market a digital single market. We have to use way more the big data that is out there but we don’t make enough out of it. What innovation and startups are concerned. It’s not only need to know but it’s need to share big data. We have to improve on cyber security. We have to work hard on our technological sovereignty just to name a few issues in these broad topics.

“Margrethe Vestager will co-ordinate the whole agenda. And be the commissioner for competition. She will work together with the commissioner for internal market, innovation and youth, transport, energy, jobs, health and justice.”

If tech giants were hoping for Europe’s next Commission to pay a little less attention to question marks hanging over the fairness of their practices they’re likely to be disappointed as Vestager is set to gain expanded powers and a broader canvas to paint on. The new role clearly positions her to act on the review of competition policy she instigated towards the end of her current mandate — which focused on the challenges posed by digital markets.

Since taking over as Europe’s competition chief back in 2014, Vestager has made a name for herself by blowing the dust off the brief and driving forward on a series of regulatory interventions targeting tech giants including Amazon, Apple and Google. In the latter case this has included opening a series of fresh probes as well as nailing the very long running Google Shopping saga inherited from her predecessor.

The activity of the department under her mandate has clearly catalyzed complainants — creating a pipeline of cases for her to tackle. And just last month Reuters reported she had been preparing an “intensive” handover of work looking into complaints against Google’s job search product to her successor — a handover that won’t now be necessary, assuming the EU parliament gives its backing to von der Leyen’s team.

While the competition commissioner has thus far generated the biggest headlines for the size of antitrust fines she’s handed down — including a record-breaking $5BN fine for Google last year for illegal restrictions attached to Android — her attention on big data holdings as a competition risk is most likely to worry tech giants going forward.

See, for example, the formal investigation of Amazon’s use of merchant data announced this summer for a sign of the direction of travel.

Vestager has also talked publicly about regulating data flows as being a more savvy route to control big tech versus swinging a break up hammer. And while — on the surface — regulating data might sound less radical a remedy than breaking giants like Google and Facebook up, placing hard limits on how data can be used has the potential to effect structural separation via a sort of regulatory keyhole surgery that’s likely to be quicker and implies a precision that may also make it more politically palatable.

That’s important given the ongoing EU-US trade friction kicked up by the Trump administration which is never shy of lashing out, especially at European interventions that seek to address some of the inequalities generated by tech giants — most recently Trump gave France’s digital tax plans a tongue-lashing.

von der Leyen was asked during the press conference whether Vestager might not been seen as a controversial choice given Trump’s views of her activity to date (Europe’s “tax lady” is one of the nicer things he’s said about Vestager). The EU president-elect dismissed the point saying the only thing that matters in assigning Commission portfolios is “quality and excellence”, adding that competition and digital is the perfect combination to make the most of Vestager’s talents.

“Vestager has done an outstanding job as a commissioner for competition,” she went on. “At competition and the issues she’s tackling there are closely linked to the digital sector too. So having her as an executive vice-president for the digital in Europe is absolutely a perfect combination.

“She’ll have this topic as a cross-cutting topic. She’ll have to work on the Digital Single Market. She will work on the fact that we want to use in a better way big data that is out there, that we collect every day — non-personalized data. That we should use way better, in the need for example to share with others for innovation, for startups, for new ideas.

“She will work on the whole topic of cyber security. Which is the more we’re digitalized, the more we’re vulnerable. So there’s a huge field in front of her. And as she’s shown excellence in the Commission portfolio she’ll keep that — the executive vice-presidents have with the DGs muscles to deal with their vast portfolios’ subject they have to deal with.”

In other choices announced today, the current commissioner for Digital Economy and Society, Mariya Gabriel, will be taking up a new portfolio called ‘Innovation and Youth’. And Sylvie Goulard was named as ‘Internal Market’ commissioner, leading on industrial policy and promoting the Digital Single Market, as well as getting responsibility for Defence Industry and Space.

Another executive VP choice, Valdis Dombrovskis, looks likely to be tackling thorny digital taxation issues — with responsibility for co-ordinating the Commission’s work on what’s been dubbed an “Economy that Works for People”, as well as also being commissioner for financial services. 

In prepared remarks on that role, von der Leyen said: We have a unique social market economy. It is the source of our prosperity and social fairness. This is all the more important when we face a twin transition: climate and digital. Valdis Dombrovskis will lead our work to bring together the social and the market in our economy.”

Frans Timmermans, who was previously in the running as a possible candidate for Commission president but lost out to von der Leyen, is another exec VP pick. He’s set to be focused on delivering a European Green Deal and managing climate action policy.

Another familiar face — current justice, consumer and gender affairs commissioner, Věra Jourová — has also been named as an exec VP, gaining responsibility for “Values and Transparency”, a portfolio title which suggests she’ll continue to be involved in EU efforts to combat online disinformation on platforms.

The rest of the Commission portfolio appointments can be found here.

There are 26 picks in all — 27 counting von der Leyen who has already been confirmed as president; one per EU country. The UK has no representation in the next Commission given it is due to leave the bloc on October 31, the day before the new Commission takes up its mandate.

von der Leyen touted the team she presented today as balanced and diverse, including on gender lines as well as geographically to take account of the full span of European Union members.

“It draws on all the strength and talents, men and women, experienced and young, east and west, south and north, a team that is well balanced, a team that brings together diversity of experience and competence,” she said. “I want a Commission that is led with determination, that is clearly focused on the issues at hand — and that provides answers.”

Commissioners elect

“There’s one fundamental that connects this team: We want to bring new impetus to Europe’s democracy,” she added. “This is our joint responsibility. And democracy is more than voting in elections in every five years; it is about having your voice heard. It’s about having been able to participate in the way our society’s built. We gave to address some of the deeper issues in our society that have led to a loss of faith in democracy.”

In a signal of her intention that the new Commission should “walk the talk” on making Europe fit for the digital age she announced that college meetings will be paperless and digital.

On lawmaking, she added that there will be a one-in, one-out policy — with any new laws and regulation supplanting an existing rule in a bid to cut red tape.

The shape of the next Commission remains in draft pending approval by the European Parliament to all the picks. The parliament must vote to accept the entire college of commissioners — a process that’s preceded by hearings of the commissioners-designate in relevant parliamentary committees.


Read Full Article

Facebook tightens policies around self-harm and suicide


Timed with World Suicide Prevention Day, Facebook is tightening its policies around some difficult topics including self-harm, suicide, and eating disorder content after consulting with a series of experts on these topics. It’s also hiring a new Safety Policy Manager to advise on these areas going forward. This person will be specifically tasked with analyzing the impacts of Facebook’s policies and its apps on people’s health and well-being, and will explore new ways to improve support for the Facebook community.

The social network, like others in the space, has to walk a fine line when it comes to self-harm content. On the one hand, allowing people to openly discuss their mental health struggles with family, friends, and other online support groups can be beneficial. But on the other, science indicates that suicide can be contagious, and that clusters and outbreaks are real phenomena. Meanwhile, graphic imagery of self-harm can unintentionally promote the behavior.

With its updated policies, Facebook aims to prevent the spread of more harmful imagery and content.

It changed its policy around self-harm images to no longer allow graphic cutting images which can unintentionally promote or trigger self-harm. These images will not be allowed even if someone is seeking support or expressing themselves to aid their recovery, Facebook says.

The same content will also now be more difficult to find on Instagram through search and Explore.

And Facebook has tightened its policy regarding eating disorder content on its apps to prevent an expanded range of content that could contribute to eating disorders. This includes content that focuses on the depiction of ribs, collar bones, thigh gaps, concave stomach, or protruding spine or scapula, when shared with terms related to eating disorders. It will also ban content that includes instructions for drastic and unhealthy weight loss, when shared with those same sorts of terms.

It will also display a sensitivity screen over healed self-harm cuts going forward to help unintentionally promote self-harm.

Even when it takes content down, Facebook says it will now continue to send resources to people who posted self-harm or eating disorder content.

Facebook will additionally include Orygen’s #chatsafe guidelines to its Safety Center and in resources on Instagram. These guidelines are meant to help those who are responding to suicide-related content posted by others or are looking to express their own thoughts and feelings on the topic.

The changes came about over the course of the year, following Facebook’s consultations with a variety of the experts in the field, across a number of countries including the U.S., Canada, U.K. Australia, Brazil, Bulgaria, India, Mexico, Philippines, and Thailand. Several of the policies were updated prior to today, but Facebook is now publicly announcing the combined lot.

The company says it’s also looking into sharing the public data from its platform on how people talk about suicide with academic researchers by way of the CrowdTangle monitoring tool. Before, this was made available primarily to newsrooms and media publishers

Suicide helplines provide help to those in need. Contact a helpline if you need support yourself or need help supporting a friend. Click here for Facebook’s list of helplines around the world. 


Read Full Article

Understand Blockchain and Start Building Apps With This Huge Learning Bundle


Cryptocurrencies such as Bitcoin are all built on blockchain technology. This system allows information to be transferred securely through a decentralized network. Although the technology has been mostly used for digital money so far, many other apps are now being built on blockchain technology. If you want to get ahead of the curve, the Mega Blockchain Mastery Bundle offers 56.5 hours of hands-on training. Right now, you can get the bundle for only $39 at MakeUseOf Deals.

Blockchain Technology

With blockchain, data is spread across thousands of devices. As a result, it’s almost impossible to crash the network or intercept data — great for financial transactions and sharing sensitive data. For obvious reasons, many tech companies are now working on blockchain-based apps. As a result, developers are in demand.

This bundle helps you understand the technology and start building your own blockchain apps. Through 13 in-depth tutorials, you discover how exactly blockchain works and how it can be used. You also learn about cryptocurrencies, smart contracts, and blockchain security.

You should come away with the skills to take advantage of the blockchain revolution — whether that is through working as a developer or investing in cryptocurrencies. When you complete the course, you can even claim a certificate to prove your knowledge.

56.5 Hours of Training for $39

Order now for just $39 to get lifetime access to this blockchain training, worth $1,587.

Read the full article: Understand Blockchain and Start Building Apps With This Huge Learning Bundle


Read Full Article

5 Freedom of Information Sites Full of Declassified Documents and Secrets


freedom-information-sites

This is a time when governments across the world are getting more secretive and difficult to trust. So the greatest tool the common man can have is his freedom to demand information. A little bit of it can come from declassified documents and un-redacted files. And some from the FOIA.

Leaks from whistle-blowers and informants often lead to damaging exposes like with WikiLeaks. But more and more hidden information is surfacing through common people and activists. The trailblazers leverage the Freedom of Information Act (FOIA) in the USA and its equivalents in other countries.

These portals show how they are forcing changes and making an impact, and even gives you the power and guidelines to dig out data yourself.

1. FBI and CIA Vaults (Web): Official Declassified Documents

The FBI Vault and CIA's Reading Room have all files declassified through freedom of information act requests

Everyone wants to know what’s happening at two of the most secretive government bodies: The Federal Bureau of Investigations (FBI) and the Central Intelligence Agency (CIA). Well, you’ll be happy to know that they have dedicated portals for declassified documents.

The FBI Vault is the official library formed due to the FOIA. It houses over 6,700 scanned documents.

Every new file that is released gets added to this vault. You can search the database with the help of optical character recognition (but that doesn’t always work well). You can also browse it alphabetically. Go through the categories on war, crime, courts, unexplained phenomena, intelligence and terrorism, civil rights, etc.

The CIA Reading Room is a similar library of all records released under the FOIA.

This repository is as well-stocked as the FBI, but I’d dare say it’s a little more entertaining. The CIA has a few “greatest hits” for you to browse. Under Frequently Requested Records, you’ll find the top 11 CIA files the public wants to read, from UFOs to spies. And it voluntarily declassifies certain records, such as its files during the 7-year Argentinian war, under the Historical Collections.

2. National Security Archive (Web): The King of FOIA Requests

National Security Archive is the largest repository of declassified documents through Freedom of Information Act

In 1985, journalists and scholars at the George Washington University formed the non-profit National Security Archive as its own watchdog for government secrecy. The world’s largest non-governmental collection of declassified U.S. documents is a result of over 30 years of FOIA requests.

The archives do an excellent job of explaining what you’re reading. The staff puts together detailed articles that compile different declassified documents. Subsections such as the Cyber Vault and the Nuclear Vault will help you narrow down resources about with a massive number of declassified documents.

Head to the virtual reading room to browse or search through all the documents published by the National Security Archive, and even a YouTube channel for some of the videos it has obtained. There are also guidelines on how to file FOIA requests, but if you’re interested in being part of that movement, check out Muck Rock below.

3. The Black Vault (Web): UFOs and Other Secret Files

John Greenwald's The Black Vault features declassified documents and government files about UFOs

No article about FOIA requests would be complete without mentioning the Black Vault. Its founder, John Greenwald, has been filing requests since he was 15 and shares all the files on his website.

The Black Vault has become the internet’s virtual storehouse for all kinds of documents released by different arms of the U.S. government, with a focus on UFOs and unexplained activity. The site is divided into sections for UFOs, government, history, military/defense, intelligence, science, the fringe, and wartime. The forums host an active community of like-minded people looking to find out what’s out there.

UFO coverage gets the most attention, with sections on spotting weird aircraft, alien encounters, space anomalies, and cryptozoology. Before you scoff at it, remember, Greenwald has hundreds of declassified government documents that he is basing this on. In fact, there is also a section on debunking hoaxes. There is a method to the madness here.

4. Mark Curtis (Web): UK Foreign Policy and World Impact

Mark Curtis's blog collects declassified documents about UK foreign policy

What John Greenwald is for UFOs in America, Mark Curtis is for foreign policy in the UK. The British government has been one of the most influential forces in shaping the 20th century. Curtis analyzes its foreign policies through original documents and transcripts from the National Archives.

The files are broken down by country or region, as well as topics like censorship, nuclear weapons, torture, and intelligence. Of course, if you only got the documents, it would be pretty useless to a lay reader. But Curtis adds reference links from news agencies that pertain to the documents, providing much-needed context to the policy at play.

Away from the Declassified page, the main site has articles by Curtis, where he marries his knowledge of foreign policy history with current news to provide insight into what’s happening in the world. If you like his analysis, check out one of the five books he wrote.

5. Muck Rock (Web): File, Share, and Analyze FOIA Requests

Muck Rock teaches you how to file freedom of information act requests and analyze them through journalists

Do you want to start filing FOIA requests for public records, track their progress or rejection, and figure out how to analyze them to expose miscreants and secrets? Stop by Muck Rock, a space for all the guidance you’ll need.

Muck Rock has everything you need to file requests, teaching you the tools of the trade for total transparency. The site also has a few experienced journalists at its helm and writers and analysts who go through the released files to find newsworthy items in them. You can even contribute and be part of the transparency movement by dedicating your knowledge, local expertise, or funds.

Muck Rock also releases a weekly round-up of news made possible by FOIA requests. It’s not going to be earth-shattering, but it’s heartening to see the impact transparency can have at the grass-roots level. It’s the first step to change.

Going Beyond FOIA

There are plenty of other websites and activists that rely on FOIA requests. But to get the secrets, you often need to rely on whistle-blowers. There are renegades like Wikileaks and there are journalistic organizations like the ICIJ, which released the Panama Papers.

In the end, what you’re looking for is transparency, and trustworthy sources that don’t hide or redact information. In a world where news is often gagged and you’re uncertain of media biases, the first step is to find world news outlets guaranteed free from censorship.

Read the full article: 5 Freedom of Information Sites Full of Declassified Documents and Secrets


Read Full Article

InMobi’s Glance raises $45M to expand outside of India


Glance, a subsidiary of Indian mobile ad business firm InMobi, said today it has raised $45 million as it prepares to scale its business outside of India and expand its product offerings.

The unnamed, maiden financing round for Glance was funded by Mithril Capital, a growth stage investment firm co-founded by Silicon Valley investors Peter Thiel and Ajay Royan.

In an interview with TechCrunch, Naveen Tewari, founder and CEO of InMobi Group, said the current round has not closed and could bag another $30 million to $55 million in the next two months.

Glance operates an eponymous service that shows media content in local languages on the lock screen of Android-powered smartphones. InMobi has partnered with a number of top smartphone vendors including Xaomi, Samsung, and Gionee to integrate Glance into their operating systems.

Glance, which was launched in September last year and supports English, Hindi, Tamil and Telugu, has amassed 50 million monthly active users in India, its primary market. Users are spending an average of 22 minutes with Glance each day, he said.

“All the new smartphone models launched by Samsung, Xiaomi, and a handful of other vendors have launched with Glance on them,” Tiwari said.

In a statement, Mithril Capital’s Rohan, said, “We share Glance’s global vision of breaking through the constraints of application architectures and linguistic markets to deliver rich, frictionless, and engaging experiences across a myriad of cultures and languages.” As part of the financing round, he is joining Glance’s board.

Glance does not show traditional ads, something it intends to never change, but shows a certain kind of content to drive engagement for brands.

In the months to come, Glance plans to expand the platform and bring short-form videos (Glance TV), and mini games (Glance Games) to the lock screen. It is also working on a feature dubbed Glance Location that will enable brands to court users in their vicinity, and Glance Shopping to explore ways to bring commerce around content.

As of today, InMobi Group is not monetizing Glance platform, but plans to explore ways to make money from it by early next year, Tiwari said.

The 12-year-old firm said it plans to expand footprints of Glance outside of India. The company plans to take Glance to some Southeast Asian markets like Malaysia, Indonesia, and Thailand. InMobi’s Tiwari said the Glance has already started to find users in these markets.

InMobi Group, which had raised $320 million prior to today’s financing round, has been profitable for several years. But the company decided to raise external funding to accelerate Glance’s growth, Tiwari said.

The firm, which has three subsidiaries including its marquee marketing cloud division, plans to go public in the next few years. But instead of taking the entire group public, Tiwari said the firm is thinking of publicly listing each division as they mature. The marketing cloud division, which brings in the vast majority of revenue for the firm, will go public first, he said.

“The IPO plans remain, and we will evaluate them as we go along. The reality, however, is that the market is so big and there is so much room that we can continue to be private for a few more years,” he said.


Read Full Article

Watch Apple unveil the new iPhone live right here


Apple is set to announce new iPhone models today. The company is holding a keynote on its campus at 10 AM PT (1 PM in New York, 6 PM in London, 7 PM in Paris). And you’ll be able to watch the event right here as the company is streaming it live.

Rumor has it that the company plans to unveil three new smartphones. The iPhone 11 should replace the iPhone XR in the lineup, while the iPhone 11 Pro and 11 Pro Max should replace the iPhone XS and XS Max respectively.

Apple could also update the Apple Watch with a new titanium version. You can also expect to get the release date of iOS 13, iPadOS 13, tvOS 13, macOS Catalina and watchOS 6. Let’s see if Apple announces the launch dates of Apple TV+ and Apple Arcade as well.

When it comes to less likely announcements that could still happen, Apple has been working on new MacBooks, a new Apple TV with a more powerful system-on-a-chip and new iPads. All eyes are on the new iPhone, but Apple could use today’s conference to announce those other products.

You can watch the live stream directly on this page. For the first time, Apple is streaming its conference on YouTube.

If you have an Apple TV, you can download the Apple Events app in the App Store. It lets you stream today’s event and rewatch old ones. The app icon was updated a few days ago for the event.

And if you don’t have an Apple TV and don’t want to use YouTube, the company also lets you live-stream the event from the Apple Events section on its website. This video feed now works in all major browsers — Safari, Microsoft Edge, Google Chrome and Mozilla Firefox.

Of course, you also can read TechCrunch’s live blog if you’re stuck at work and really need our entertaining commentary track to help you get through your day. We have a team in the room.


Read Full Article

SimShine raises $8 million for home security cameras that use edge computing


SimShine, a computer vision startup based in Shenzhen, has raised $8 million in pre-Series A funding for SimCam, its line of home security cameras that use edge computing to keep data on-device. The funding was led by Cheetah Mobile, with participation from Skychee, Skyview Fund and Oak Pacific Investment.

Earlier this year, SimShine raised $310,095 in a crowdfunding campaign on Kickstarter. It will use its pre-Series A round for product development and hiring.

SimShine’s team started off developing computer vision and edge computing software, spending five years working with enterprise clients before launching SimCam.

The company plans to release more smart home products that use edge computing with the ultimate goal of building a IoT platform to connect different devices, co-founder and chief marketing officer Joe Pham tells TechCrunch. SimCam currently integrates with Amazon Alexa and Google Assistant, with support for Apple Homekit in the works.

Pham says edge computing protects users’ privacy by keeping data, including face recognition data, on device, while also decreasing latency and false alarms, because calculations are performed continuously on the device (cameras connect to Wi-Fi so customers can watch surveillance video on their smartphones). It also means customers don’t have to sign up for the subscription plans that many cloud-based home security cameras require and reduces the price of each device since SimCam does not have to maintain cloud servers.


Read Full Article

Watch Apple unveil the new iPhone live right here


Apple is set to announce new iPhone models today. The company is holding a keynote on its campus at 10 AM PT (1 PM in New York, 6 PM in London, 7 PM in Paris). And you’ll be able to watch the event right here as the company is streaming it live.

Rumor has it that the company plans to unveil three new smartphones. The iPhone 11 should replace the iPhone XR in the lineup, while the iPhone 11 Pro and 11 Pro Max should replace the iPhone XS and XS Max respectively.

Apple could also update the Apple Watch with a new titanium version. You can also expect to get the release date of iOS 13, iPadOS 13, tvOS 13, macOS Catalina and watchOS 6. Let’s see if Apple announces the launch dates of Apple TV+ and Apple Arcade as well.

When it comes to less likely announcements that could still happen, Apple has been working on new MacBooks, a new Apple TV with a more powerful system-on-a-chip and new iPads. All eyes are on the new iPhone, but Apple could use today’s conference to announce those other products.

You can watch the live stream directly on this page. For the first time, Apple is streaming its conference on YouTube.

If you have an Apple TV, you can download the Apple Events app in the App Store. It lets you stream today’s event and rewatch old ones. The app icon was updated a few days ago for the event.

And if you don’t have an Apple TV and don’t want to use YouTube, the company also lets you live-stream the event from the Apple Events section on its website. This video feed now works in all major browsers — Safari, Microsoft Edge, Google Chrome and Mozilla Firefox.

Of course, you also can read TechCrunch’s live blog if you’re stuck at work and really need our entertaining commentary track to help you get through your day. We have a team in the room.


Read Full Article

Paytm’s annual loss doubles to $549M


Running a payments business in India is not cheap. Just ask Paytm. One of India’s largest payment companies reported a net loss of Rs 3959 crore ($549 million) for the financial year that ended in March, up 165% over 1490 crore ($206 million) in the same period last year.

During the same period, the company’s revenue rose to Rs 3232 crore ($448 million), compared to Rs 3052 crore ($423 million) in the year before. The firm’s debt also surged to Rs 695 crore ($96 million), One97 Communications, the parent firm of Paytm, told investors in its annual report.

One97 Communications also runs an e-commerce business, which recently raised money from eBay, and Paytm Money, that runs mutual funds business. On a consolidated basis, the 9-year-old firm reported an annual loss of Rs 4217.20 crore ($584 million), up from Rs 1604.34 crore ($222 million) from the year before.

Indian news outlet BloombergQuint first reported (paywalled) the financial performance of Paytm.

The loss should worry Paytm, whose CEO Vijay Shekhar Sharma said in a conference last week that the firm would begin to work on going public in the next 22 to 24 months. The level of competition that Paytm faces today is only about to increase in the coming future, and unlike earlier, the Indian firm is not facing off financially weaker local rivals.

Paytm, which has raised over $2 billion to date from a range of investors including SoftBank, Alibaba, and Berkshire Hathaway, continues to be the largest mobile wallet app provider in India, but increasingly users are moving to government-backed UPI payments infrastructure. In UPI land, Paytm competes with Flipkart’s PhonePe and Google Pay, both of which are heavily-backed.

As of July, both PhonePe and Google Pay commanded a bigger market share across UPI apps than Paytm.

Also in UPI land, you don’t make money on each transaction. So lately, every payments firm in India, including Paytm, has expanded it offering to include financial services such as a credit card, or loan, or insurance.

In many ways, this has created a level playing field for payment firms that did not dominate the wallet business.

In a statement, Paytm said it has been investing $1 billion per year for the last two years to “expand payments ecosystem in our country.” The company plans to invest a further $3 billion in the next two years.

“We believe India is at the inflection point of digital payments and Paytm’s sole focus is towards solving the merchant payments and offering them financial services. We will invest Rs 20,000 crore ($2.7 billion) in the next two years towards achieving this,” a company spokesperson said.

The biggest challenge for Paytm and other UPI payment apps has yet to emerge. Before the end of this year, WhatsApp, which has over 400 million users in India, plans to offer UPI payment option to all its years in the coming month.


Read Full Article

You Can Now Listen to Apple Music on the Web


Apple Music now has a web player. This means that if you’re an Apple Music subscriber you can listen to Apple Music on the web on any device with an internet connection. And all without having to install either iTunes or the standalone Music app.

Apple Music Catches Up With Spotify

Apple first launched Apple Music in 2015. Since then the streaming music service has gained a host of fans, with the number of subscribers increasing every year. And while it isn’t yet as popular as Spotify, it’s certainly Spotify’s biggest competition.

Apple Music and Spotify differ in a number of ways. The lack of a free Apple Music option being the most obvious. There’s also the fact that Spotify has offered a web player since 2013. But now, finally, Apple is bringing Apple Music to the web as well.

How to Listen to Apple Music on the Web

To listen to Apple Music on the web, just head to beta.music.apple.com. As the URL suggests, the Apple Music web player is currently in beta, but it works absolutely fine. All you need to do is sign in with your Apple ID and you can then start streaming music.

Apple Music’s web player looks almost exactly like the standalone Apple Music app being launched alongside macOS Catalina later this year. You’ll be able to play any song in the Apple Music catalog, access your playlists, and listen to radio stations.

Apple is suggesting that this is just the core set of features for now, but that the company will be adding more functionality over time. This is, after all, just a beta. Which is why Apple will also be soliciting feedback from users during the testing phase.

The Apple Music Tips Everyone Should Know

Bringing Apple Music to the web effectively makes Apple Music platform-agnostic. Now, people can use Apple Music on any device, regardless of which operating system it’s running, and all without the need to install iTunes or a custom-built app.

Apple is even promising to allow new users to sign up for Apple Music through the website, but that isn’t possible at the moment. Still, if that option attracts new users to the streaming music service, here are the Apple Music tips all users need to know.

Read the full article: You Can Now Listen to Apple Music on the Web


Read Full Article

Facebook Dating Is Now Available in the US


Facebook has officially launched Facebook Dating in the US. Facebook Dating has been available in various countries since 2018. But now, after a year of testing, Facebook is confident that Facebook Dating is ready for its debut in the US.

For the uninitiated, Facebook Dating shares some similarities with Hinge, the dating app for people who hate dating apps. Unlike Tinder, which is all about matching people based on looks, Facebook Dating suggests people who share similar interests with you.

How Facebook Dating Works

In a post on the Facebook Newsroom, Facebook explains how Dating works. To use Facebook Dating you’ll need to be 18 or older and have the most recent version of Facebook installed. Then, just open the Facebook Dating tab and create a profile.

You’ll be matched with people who share similar interests as you, but only those who have also opted in to using Facebook Dating. By default, Facebook Dating won’t match you with friends, but will match you with friends of friends.

If you want to get to know someone better you can either click the Like button or comment directly on their Facebook Dating profile. Facebook Dating profiles are kept separate from normal Facebook profiles to stop your family and friends spying on you.

You can (if you want to) use Facebook’s Secret Crush feature to potentially match with people you already know. Once you have opted in you can choose up to nine friends who you’re interested in. And if they’re also interested in you, you’ll both be notified.

Other Dating Apps Worth Trying

Facebook Dating is free to use and is now available in 20 countries, including Argentina, Brazil, Canada, Colombia, Malaysia, Mexico, the Philippines, Thailand, and the United States. Facebook has also promised that Dating will be launching in Europe in 2020.

There are already plenty of online dating sites and apps available, including Bumble, Tinder, Grindr, and OkCupid. And with young people generally steering clear of Facebook, here are the best free dating apps for teenagers and young adults.

Read the full article: Facebook Dating Is Now Available in the US


Read Full Article

The 3 Best Free Apps for Syncing Android Contact Photos With Facebook

The 10 Best Psychology Websites to Help Educate Yourself


Psychology is one of the most interesting things you can study. At its core, it explores human behavior and explains why we do things the way that we do.

Psychology is also a valuable tool for healing interpersonal relationships, but it’s a difficult field to study because the web is full of pseudo-science. When browsing Google, it’s hard to escape BuzzFeed personality quizzes and clickbait “studies.” How do you get around this mess?

Well, here are some of the best psychology websites. Through these, you can learn everything you’ve ever wanted to know.

1. Online University Courses

The best place to start is an introductory psychology course taught by an expert. Schools such as Yale, Oxford, and MIT sometimes offer free online psychology courses for beginners.

Because these courses are free, you won’t earn credit. However, you will receive a world-class introduction to psychology. You can also find supplementary info on the YouTube channel YaleCourses.

Additionally: you can take an online introductory statistics class. Statistics are key for understanding study results.

2. Explore Psychology With Khan Academy

Khan Academy is one of the best online tools for beginners. They don’t have a ton of resources, but the ones they do post are great.

Their lecture series begins with human development and the role of genetics in behavior. They then touch upon classic psychological areas for beginners, including personality theories and disorders.

Many of the online psychology courses offered by Khan Academy come with supplemental materials. As well, you can study for the MCAT or similar tests with their review questions.

3. Understanding Key Psychology Experiments With Simple Psychology

Simple Psychology Website

Experiments form the backbone of what we know about human psychology. Many experiments are so famous that lecturers only refer to them in passing, as they expect you to know the key points.

Online psychology courses can be difficult, so take every opportunity that you can to make things easier on yourself! Use Simple Psychology’s overview on famous experiments to help: they’re listed by their titles under the “studies” tab in the top menu.

4. Crash Course Psychology

If you’re looking for a gentle introduction to psychology, Crash Course (created by Hank and John Green) is the way to go.

The Crash Course YouTube channel offers bite-sized lectures on academic subjects at a high-school level. There are a ton of videos, but each video is under twelve minutes long.

This series covers many of psychology’s biggest hits. It’s a great way to review difficult concepts by yourself.

5. Testing Yourself With Psych Central

Psych Central Website

Assessments and analysis are key to studying psychology. While Psych Central offers several great resources, one of its biggest strengths is a list of psychological tests and links.

Now, we need to caution that these tests should not be used for diagnostic purposes. Some of the tests on this list are more legitimate than others, too!

True psychological assessments must be performed by trained professionals. However, online psychology tests are important for understanding what assessments look like and how they work.

With that in mind, spend some time exploring these psychological tests and quizzes. Take note of how the questions are phrased, research the theories linked to each assessment, and have some fun learning about yourself!

You can also learn more about your “life’s purpose” with these additional online tests.

6. Get to the Source With APA.org

APA Website

The American Psychology Association (APA) is a major authority when it comes to an academic study on how the mind works. Its website is a never-ending treasure trove of information for professionals, researchers, and students alike.

APA’s educational psychology website also offers detailed overviews on most psychological conditions. Additionally, they cover:

  • New research involved in the discipline.
  • New ways to get involved in psychology research, politics, and education.
  • Opportunities to learn from the professionals involved.

Another great resource on their website is the “Psychology Help Center,” which offers advice to those considering therapy or treatment.

7. Pursue Big Ideas With TED Talks

Once you grasp the psychological basics, you can appreciate its impact on our world. TED Talks are short, powerful speeches by global experts working in technology, education, and design. The site is divided by subject, and there are over a hundred psychology TED Talks available!

Right now, the psychology videos range from:

  • Mental illness.
  • The reliability of memory.
  • The power of perseverance.

Often a single video will grab your interest and send you down a whole new path of research and learning.

If you like the engaging style of TED talks, but want more in-depth discussions, consider listening to podcasts instead!

8. Practical Psychology

Practical Psychology is a YouTube channel I personally enjoy. It’s also a great place to go if you’re looking for easy info on psychology and how it affects our everyday lives.

Practical Psychology is perfectly designed for the average YouTube viewer, and it’s similar to Crash Course Psychology at #4 on our list. Video topics range from the dangers of limited thinking to the habits of highly effective people.

9. Mind Hacks

Mind Hacks Psychology Blog

Mind Hacks is a “news and views” psychology blog that publishes insights into neuroscience. Despite the simplicity of the blog’s design, the coverage is broad and the analysis is complex.

This broad coverage can make Mind Hacks difficult to navigate, but a fan-made psychology Wiki helps readers locate topics of interest. In each article, authors Dr. Tom Stafford and Dr. Vaughn Bell summarize and explore recent studies.

If you want to read more scholarly research, this psychology blog is a great place to start! Their statistics and methods are well explained, and you can follow up with their list of referenced articles.

10. Follow Academic Research With Google Scholar

Google Scholar Home

The psychology websites listed above are amazing for learning the basics. However—if you want to follow psychology in its truest form—your best bet is to go to the primary source.

Google Scholar makes psychology as simple as possible. A quick search directs you to journal articles, books, and PDFs of interest. You can also set up alerts to track authors, topics, or journals of note. Additionally, you can save important papers to your library.

If you want to read more, try something from this list of Google Scholar’s most popular psychology journals.

Where Can You Go From Here?

One of the best things about psychology is the number of directions you can take when researching the different subcategories online! There are so many areas of psychology, and it’s impossible to run out of things to learn.

If you want to find some great trivia to use at your next dinner party, here are some psychological reasons why video games are so addicting.

Read the full article: The 10 Best Psychology Websites to Help Educate Yourself


Read Full Article