Tag Archives: machinelearning

Start up: Theranos’s last days?, Samsung’s water-unproof S7 Active, the Pokemon Go craze, and more

Posted on by
Reply


Planning a crewed lunar mission? There’s some code for you on Github! Photo from Nasa Goddard Space Research Centre on Flickr.

You can now sign up to receive each day’s Start Up post by email. You’ll need to click a confirmation link, so no spam.

A selection of 12 links for you. Apply topically. I’m charlesarthur on Twitter. Observations and links welcome.

Theranos dealt sharp blow as Elizabeth Holmes is banned from operating labs • WSJ

John Carreyrou, Michael Siconolfi and Christopher Weaver:

»Silicon Valley startup Theranos Inc. is fighting for its life after regulators decided to revoke its license to operate a lab in California because of unsafe practices and to ban founder Elizabeth Holmes from the blood-testing business for at least two years.

The sanctions were laid out in a letter to Theranos released Friday by the agency that oversees US labs, the Centers for Medicare and Medicaid Services. Theranos said it is still seeking to resolve its issues with the regulator.

One sanction, a monetary fine of $10,000 a day until all deficiencies have been corrected, goes into effect July 12. The most serious sanctions, such as the ban of Ms. Holmes, won’t go into effect for 60 days.

If it fails to reach a settlement with the government, Theranos’s options are limited. Almost any course it takes will dramatically reshape the company that Ms. Holmes founded in 2003 as a Stanford University dropout and grew to a valuation of more than $9 billion in a 2014 fundraising round.

«

The first version of this that I saw at 0643 BST (0143 EST) Friday had a single byline (Siconolfi’s) and began more tamely: “US federal health regulators dealt a major blow to Theranos by banning founder Elizabeth Holmes from operating a blood-testing laboratory for at least two years and pulling regulatory approval for the company’s California lab.”

Clearly, the addition of two reporters and 18 hours sharpened up the intro (“lede” in the US; first paragraph to everyone else) quite a bit. And gave them time to put a very spooky picture of Holmes at the top.

And Theranos indeed looks cooked.
link to this extract

DNA sequencing costs plotted over time • National Human Genome Research Institute (NHGRI)

»

To illustrate the nature of the reductions in DNA sequencing costs, each graph also shows hypothetical data reflecting Moore’s Law, which describes a long-term trend in the computer hardware industry that involves the doubling of ‘compute power’ every two years (See: Moore’s Law [wikipedia.org]). Technology improvements that ‘keep up’ with Moore’s Law are widely regarded to be doing exceedingly well, making it useful for comparison.

In both graphs, note: (1) the use a logarithmic scale on the Y axis; and (2) the sudden and profound outpacing of Moore’s Law beginning in January 2008. The latter represents the time when the sequencing centers transitioned from Sanger-based (dideoxy chain termination sequencing) to ‘second generation’ (or ‘next-generation’) DNA sequencing technologies. Additional details about these graphs are provided below.

These data, however, do not capture all of the costs associated with the NHGRI Large-Scale Genome Sequencing Program. The sequencing centers perform a number of additional activities whose costs are not appropriate to include when calculating costs for production-oriented DNA sequencing. In other words, NHGRI makes a distinction between ‘production’ activities and ‘non-production’ activities. Production activities are essential to the routine generation of large amounts of quality DNA sequence data that are made available in public databases; the costs associated with production DNA sequencing are summarized here and depicted on the two graphs.

«

We’re good at sequencing, but less good at understanding what genomes tell us. That hasn’t improved as quickly.
link to this extract

Samsung Galaxy S7 Active fails Consumer Reports water-resistance test • Consumer Reports

Jerry Bellinson put not one but two successive Galaxy S7 Actives into the equivalent of five feet of water for 30 minutes. They didn’t make it:

»For a couple of days following the test, the screens of both phones would light up when the phones were plugged in, though the displays could not be read. The phones never returned to functionality.

Samsung says it has received “very few complaints” about this issue, and that in all cases, the phones were covered under warranty.

“The Samsung Galaxy S7 active device is one of the most rugged phones to date and is highly resistant to scratches and IP68 certified,” the company said in a written statement. “There may be an off-chance that a defective device is not as watertight as it should be.” The company says it is investigating the issue.

The Active is one of three versions of the Samsung Galaxy S7, and it was the only one to fail our water-immersion test.

«

Could be two lemons, but that doesn’t speak well to the quality control. Waterproofing seems to be a popular feature with testers, at least, because you can.. test it.
link to this extract

Teen playing new Pokémon game on phone discovers body in Wind River • County 10

»Shayla [Wiggins] tells County 10 that she woke up this morning and began playing a game on her cell phone called Pokémon Go, an augmented reality game that encourages the user to capture as many Pokémon as possible. “The Pokémon are all over Riverton,” she said. Shayla showed County 10 the game on her cellphone which displayed a map of Riverton where these Pokémon are located.

“I was trying to get a Pokémon from a natural water resource,” she explained. She said that she jumped over the fence to go towards the river in search of a Pokémon.

“I was walking towards the bridge along the shore when I saw something in the water,” Shayla said. “I had to take a second look and I realized it was a body.” She said the figure was floating about three feet from the shore and it looked like an average size male body. She reports that she thinks the man was native, but she can’t be certain. She saw a black shirt and black pants. All of the body was reportedly submerged except for part of his back and butt.

«

This game is taking people into bizarre situations. There are even reports of people setting up armed robberies (unproven) and using it while on patrol against Isis with Kurdish militias (verified). I’m amazed; Pokemon seems to me so transparently stupid – a set of Top Trump cards – that I’m amazed anyone over the age of 12 indulges in it. And yet…
link to this extract

A malicious ‘Pokémon Go’ app is installing backdoors on Android devices • Motherboard

Joshua Kopstein:

»wannabe Pokémon masters should take heed: amid high demand for the game as it slowly rolls out across the globe, security researchers have discovered a malicious version of the Pokémon GO app floating around that installs a backdoor on Android phones, allowing hackers to exploit Poké-hype to completely compromise a user’s device.

The security firm Proofpoint discovered the malicious application, or APK, which was infected with DroidJack, a remote access tool (RAT) that compromises Android devices by silently opening a backdoor for hackers. The malicious app was uploaded to an online malware detection repository on July 7, less than 72 hours after Nintendo released the game in Australia and New Zealand.

To install it, a user needs to “side-load” the malicious app by disabling an Android security setting that normally prevents the installation of unverified third-party apps from “unknown sources.”

This is potentially a huge deal, since the game’s slow roll-out to different regions has led some impatient players to download the app from third-party websites instead of waiting for the official release on Android’s Play store, which requires side-loading to install. Proofpoint notes that several major news outlets have even provided instructions on how to find and install the app from a third party.

«

link to this extract

Original Apollo 11 Guidance Computer (AGC) source code • Github

Lots of people are cloning it and improving it – just in case they, you know, need to pilot a lunar lander mission.
link to this extract

We need to talk about AI and access to publicly funded data-sets • TechCrunch

Natasha Lomas with a hugely important analysis:

»DeepMind says it will be publishing “results” of the Moorfields research [on eye disease] in academic literature. But it does not say it will be open sourcing any AI models it is able to train off of the publicly funded data.

Which means that data might well end up fueling the future profits of one of the world’s wealthiest technology companies. Instead of that value remaining in the hands of the public, whose data it is.

And not just that — early access to large amounts of valuable taxpayer-funded data could potentially lock in massive commercial advantage for Google in healthcare. Which is perhaps the single most important sector there is, given it affects everyone on the planet. If you don’t think Google has designed on becoming the world’s medic, why do you think it’s doing things like this?

Google will argue that the potential social benefits of algorithmically improved healthcare outcomes are worth this trade off of giving it advantageous access to the locked medicine cabinet where the really powerful data is kept.

But that detracts from the wider point: if valuable public data-sets can create really powerful benefits, shouldn’t that value remain in public hands?

«

Yes. Exactly. This is a key point which is being ignored: data is the necessity for Google and the British government is not seeking sufficiently clear repayment for it.
link to this extract

AI, Apple and Google • Benedict Evans

Quite a long musing on where we are with AI – which typically never quite arrives, because every time it does something smart (understands speech, identifies faces) we say “oh, that’s just computing“:

»A common thread for both Apple and Google, and the apps on their platforms, is that eventually many ‘AI’ techniques will be APIs and development tools across everything, rather like, say, location. 15 years ago geolocating a mobile phone was witchcraft and mobile operators had revenue forecasts for ‘location-based services’. GPS and wifi-lookup made LBS just another API call: ‘where are you?’ became another question that a computer never has to ask you. But though location became just an API – just a database lookup – just another IF statement – the services created with it sit on a spectrum. At one end are things like Foursquare – products that are only possible with real-time location and use it to do magic. Slightly behind are Uber or Lyft – it’s useful for Lyft to know where you are when you call a car, but not essential (it is essential for the drivers’ app, or course). But then there’s something like Instagram, where location is a free nice-to-have – it’s useful to be able to geotag a photo automatically, but not essential and you might not want to anyway. (Conversely, image recognition is going to transform Instagram, though they’ll need a careful taxonomy of different types of coffee in the training data). And finally, there is, say, an airline app, that can ask you what city you’re in when you do a flight search, but really needn’t bother.

In the same way, there will be products that are only possible because of machine learning, whether applied to images or speech or something else entirely (no-one at all looked at location and thought ‘this could change taxis”). There will be services that are enriched by it but could do without, and there will be things where it may not be that relevant at all (that anyone has realised yet). So, Apple offers photo recognition, but also a smarter keyboard and venue suggestions in the calendar app – it’s sprinkled ‘AI’ all over the place, much like location. And, like any computer science tool, there will be techniques that are commodities and techniques that aren’t, yet.

«

link to this extract

Exclusive: why Microsoft is betting its future on AI • The Verge

Casey Newton got to meet lots of people at Microsoft who are working on bots and AI:

»I meet with Kirk Koenigsbauer, corporate vice president of marketing for Office. He shows me a range of ways where intelligence is making Office easier to use. In September 2014 Microsoft introduced Delve, a kind of Fitbit for productivity that is included with Office 365. The app analyzes how much time you spend in email and in meetings, and highlights times on your calendar where you have extended periods of time to do more complicated, meaningful work. It tells you what percentage of people you sent an email to actually read it, and how quickly. It will suggest reaching out to colleagues that you haven’t emailed in a while. It even shows you response times for your colleagues, and for yourself.

If your organization lives in Google Apps, as do many big Silicon Valley companies, browsing Delve felt like a revelation. You don’t have to be a numbers nerd to find this kind of information useful. If you’re a manager, Delve can tell you at a glance how much time you’ve spent with each of your employees over the past week. This kind of intelligence isn’t as sexy as a general AI that anticipates your every need — but it’s here today, it works, and it makes Google Apps look like a neglected backwater by comparison.

«

1) Google Apps pretty much is a neglected backwater
2) would love to know if the statistics gathered by Delve actually have any meaning in the real world, or are just numbers collected because they can be.
link to this extract

Security Flaw in OS X displays all keychain passwords in plain text • Medium

Brenton Henry:

»This afternoon, a friend learned the hard way that you don’t let an unofficial company take control of your computer to provide “support”. However, it was what I learned that shocked me the most.

There is a method in OS X that will allow any user to export your keychain, without sudo privileges or any system dialogs, to a text file, with the username and passwords displayed in plain text. As of this writing, this method works in at least 10.10 and 10.11.5, and presumably at the least all iterations in between.

«

I tried his method; I had to click an “Allow” dialog for every single item in my keychain, which wasn’t a trivial number. So this exploit isn’t one to think deeply about. More to the point: what happened to his friend? Was it keychain-related?
link to this extract

How the Feds use Photoshop to track down paedophiles • Motherboard

Joseph Cox:

»The most innocent clues can crack a case. In 2012, a holiday photo of a woman and child holding freshly caught fish ended up being a key lead in a child pornography investigation.

Found within a cache of illegal, explicit material, the photo would eventually point detectives to a outdoor camping site in Richville, Minnesota, and result in the victims’ rescue, and suspect’s conviction in December 2012.

But first, detectives had to determine where the photo was taken. To do that, they cropped out the fish, sanitized the image, and sent it to Cornell University for identification, Jim Cole, the National Program Manager for Victim Identification at US Immigration and Customs Enforcement (ICE), an agency within the Department of Homeland Security (DHS), recalled to Motherboard in a phone call.

The university determined the species of fish, which was found in a particular region. Investigators then edited the suspect and victim out of the photo, Cole said, and distributed it to advertisers for camping grounds in the area, one of which recognized the location.

When detectives arrived, the same photo was on the wall of the camping office, Cole added.

“It’s all about making the haystack smaller, so we can find the needle,” he said.

«

A logo on a sweatshirt? A bottle of pills in the background? It can all contribute to cracking the case
link to this extract

Exclusive: Google is building two Android Wear smartwatches with Google Assistant integration • Android Police

David Ruddock has a strong and detailed rumour:

»The inevitable question with these Google smartwatches is “why?” I’m afraid I don’t have a concrete answer for you. But I can speculate. As Android Wear has evolved, manufacturer interest in it has not skyrocketed as Google likely hoped it would. At best, it appears to be holding steady. Once considered Wear’s strongest partner, LG has announced no new mainstream Wear device since the old Urbane last spring (the LTE is unashamedly niche with limited availability, and was heavily delayed). The number of new Wear OEMs announced lately has been modest, aside from a few niche fashion products that are unlikely to have a major impact on Wear’s distribution.

By building its own smartwatches, Google can implement exactly the hardware and features it believes will best demonstrate Android Wear’s capabilities.

«

Good luck with that. The OEMs aren’t doing it because they aren’t selling. (Unless they’re selling in China, in which case Google will have trouble too.)
link to this extract

Errata, corrigenda and ai no corrida: none notified

Start up: Twitter gets AI too, Apple’s photo AI, would Brexit raise roaming prices?, Spotify’s 100m, and more

Posted on by
Reply


Presenting! It can be so easy, but can be so bad. Photo by Alice Bartlett on Flickr.

You might sign up to receive each day’s Start Up post by email. Unless you already did.

A selection of 12 links for you. Now that the summer solstice has passed in the northern hemisphere. I’m charlesarthur on Twitter. Observations and links welcome.

Alicia Keys is done playing nice. Your phone is getting locked up at her shows now. • The Washington Post

Geoff Edgers:

»On a cool Manhattan night, DJ Walton, who helps manage Alicia Keys, steps outside the Highline Ballroom to tell the guy at the door who, exactly, he may allow to bring a cellphone into the singer’s sold-out gig. The list is very short.

“Like, Queen Latifah,” says Walton.

Benji Spanier nods and spreads the news to everybody else. This is a “phone-free event,” he tells fans waiting in line. And that doesn’t mean airplane mode. Spanier holds a gray, rubbery pouch in his hand. Your phone goes in here, he says, and then we lock it.

“What?” one fan grumbles.

Quickly, Spanier adds an important addendum.

You keep that locked pouch with you. Spanier also explains that if you need to use your phone, you can just come outside and he can quickly unlock it by tapping it on a metal disk slightly larger than a bagel. The tension breaks.

“If you had told me you were going to put it in a locker, I’d have been pissed off,” Kevin Schmidt, 37, tells him. “This is okay.”

«

Special pouch, called Yondr. Good business for them.
link to this extract

 

Did Jeep’s recalled gear shifter contribute to the death of Star Trek Actor Anton Yelchin? (Updating) • Jalopnik

David Tracy:

»Earlier today, we learned about the tragic death of Anton Yelchin after the Star Trek actor was found pinned between his car and a mailbox. Now news from TMZ indicates that Yelchin’s car was a Jeep Grand Cherokee, leading us to wonder whether it was among the 1.1 million cars recalled in April due to a confusing shifter that owners often inadvertently left in neutral instead of park. [Update: it was.]

In mid April, Jeep recalled 1.1 million vehicles equipped with ZF’s eight-speed automatic transmission because, according to the automaker’s recall notice, “Some drivers… exited their vehicles without first selecting ‘PARK,’” ultimately causing the cars to roll away uncontrollably.

«

Lots of maybes, but this looks like a case where poor design led to death.
link to this extract

 

Twitter buys machine learning start-up Magic Pony • FT.com

Hannah Kuchler:

»Twitter has acquired Magic Pony, a London-based machine learning start-up, as the messaging platform tries to bolster its video and live streaming capabilities.

Magic Pony specialises in creating algorithms that can understand pictures, which could be helpful to Twitter as it pushes further into live streaming and moves away from a chronological timeline to a more curated Facebook-style news feed.

The startup was set up in 2014 and has 14 engineers, including 11 with PhDs and expertise across computer vision, the ability to understand pictures, machine learning and computational neurosciences…

Twitter acquired Magic Pony for an undisclosed sum. Its investors include Octopus Ventures, a UK-based venture capital firm that has invested in other artificial intelligence companies. These include Evi, which was acquired by Amazon, and SwiftKey, which was bought by Microsoft.

«

Airplanes to San Francisco await the chief executive Rob Bishop. Congratulations to that team.
link to this extract

 

Speaker style bingo: 10 presentation anti-patterns • Troy Hunt

Hunt nails the ten awful habits of speakers you wish hadn’t got the slot you’re attending, and offers this advice:

»seriously, you’ve got to rehearse these things like crazy and also recognise that your pace changes between private rehearsals and public presentations. On that point, I always have a timing sheet in large letters next to my iPad with a timer in an easily glanceable location:

This is invaluable. I refine the timing on subsequent rehearsal and ensure it’s accurate to the minute with two or three minutes to spare at the end just in case I start late or have an issue. I glance at it very few minutes and either slow down the pace (usually by embellishing on a topic) or speed it up to get back on track. But here’s the key – this has to be something you can check with a glance.

«

Then again, I think all of us who have spoken in public could do a similar one about the audience – the ones checking their phones, checking their laptops, etc.
link to this extract

 

Behind Apple’s advanced computer vision for Photos app • Medium

Kay Yin:

»[iOS 10’s] Photos app recognises and distinguishes the following 7 facial expressions. Expressions are distinguished after forming a “faceprint”. These distinction are used for searching. They are also rated and indexed for generation Memories and montages.
•Greedy, Disgust, Neutral, Scream, Smiling, Surprise, Suspicious

Photos app will generate Moments that falls within the following 33 categories. Default name of the moment will be automatically generated using metadata from the photos and tags from analysis of photos.

• Memories from areas of interest, Best of past memories, Memories that break out of routine, Celebration in history, Contextual memories, Crowd, Day in history, Holiday in history, Location of interest, Nearby, New contextual memories, New memories, Person’s Birthdays, Person’s memories, Recent events (calendar, crowd, holiday, people, person, social, trip, weekend), Region of interest, Social group memories, Sometime memories, Special memories, Favourited, Trips, Week in history, Weekend, Year summary, Last week, Last Weekend

Photos app supports detecting 4,432 different scenes and objects. These scenes or objects can be searched for in all languages.

Additionally, you can search for various landmarks.

«

He doesn’t specify how he knows this – possibly from using the macOS beta and digging into the accompanying files. It seems like a limited number compared to what Google must have; Google’s scenes/objects list is probably growing by 4,432 every day.
link to this extract

 

Could Brexit result in higher roaming charges? • CCS Insigh

Kester Mann:

»Should the UK vote for Brexit, mobile operators would no longer be accountable to Brussels’ regulation on roaming. Under pressure from declining revenue in traditional areas such as voice and messaging, they would be foolish not to at least consider seizing an opportunity to reapply charges.

In reality however, this would be much easier said than done in a hugely competitive market that includes a number of strong virtual providers. Indeed, some operators have already gone a long way toward abolishing roaming ahead of the ruling next June. Backtracking would be extremely unpopular and probably only work if operators moved in unison. Even then, Ofcom may still be within its rights to clamp down if it deemed the move unnecessary.

Already more than 3 million customers of Three have taken advantage of inclusive roaming since the operator launched its Feel at Home offer in 2013. Significantly, it includes popular tourist and business destinations beyond Brussels’ jurisdiction, such as Australia, Hong Kong, New Zealand and the US. In my view, this demonstrates a long-term strategy to offer low-cost roaming charges, whatever the outcome of the referendum.

Other UK providers have followed suit. Carphone Warehouse currently offers inclusive roaming in 29 countries, including Australia and the US, through its virtual service, iD. Meanwhile, Vodafone last month moved to largely abolish roaming across Europe. Tesco Mobile has a similar offer, although it is only available during the summer, a possible indication that it will review its options after the UK goes to the polls.

Perhaps the most interesting aspect of this debate is Swisscom’s recent decision to virtually abolish roaming within the EU for its Natel Infinity Plus subscribers. Given that Switzerland is not a member state and has a hugely dominant market position, this was a surprising move that suggests the value of roaming may be overestimated by some commentators.

«

(My family loves Three’s “Feel At Home” international roaming for no extra cost.)
link to this extract

 

One million machines, including routers, used to attack banks • Boing Boing

Cory Doctorow:

»Akamai’s Ryan Barnett reports on two attacks against the service’s financial customers last year: attackers used nearly 1m compromised systems to attempt to log in to users’ accounts using logins and passwords from earlier breaches.

Many of the attacks originated from proxies, but the response team found a high number of Xyxel and Arris home routers – provided by ISPs in an insecure state and not patched after deployment.

While distributed attacks are common, this story is a kind of trifecta of infosec badness: hacked, headless IoT devices rented to customers who aren’t allowed to reconfigure them; email/password breaches leaked from insecure services being leveraged on the assumption of password re-use; and attacks originating from a million IPs – all directed to financial accounts in a way that could clean out its victims of their life’s savings.

«

There must come a point where the sheer firepower is going to overwhelm any protection, surely? And what happens after that? Here’s the full Akamai report.
link to this extract

 

Get more out of your battery with Microsoft Edge • Windows Experience Blog

Jason Weber, director of Web platform team, Microsoft Edge:

»We connected a Surface Book to specialized power monitoring equipment and measured the actual power usage during typical browsing activities in Microsoft Edge, Chrome, Firefox, and Opera. We then automated each browser to perform the same series of activities: opening websites, scrolling through articles, and watching videos, opening new tabs for each task. We used the same websites you spend your time on – Facebook, Google, YouTube, Amazon, Wikipedia and more.

Average power consumption in milliwatts for identical workloads in Microsoft Edge, Google Chrome, Mozilla Firefox, and Opera (with battery saver mode enabled). Unless specified, all browser settings were left at their defaults.

For these browsing activities, our tests show Microsoft Edge is a more energy efficient browser on Windows 10, with up to 36%-53% more battery life to get what you need done —whether you’re studying at the library, researching dream vacation destinations, or checking in with your friends on social networks.

«

Bet Apple would get the same for Safari on an Apple machine. Chrome is a battery hog – no two ways about it.
link to this extract

 

Google vs. Apple: contrasting approaches to app store evolution • Tech-Thoughts

Sameer Singh:

»Instant Apps and Google Now On Tap are mildly interesting products when looked at individually. But when combined, they have the potential to reshape the app interaction model as we know it. That said, this is a risk because a change on this scale will take quite a bit of time to diffuse through to developers and consumers at scale. But if executed correctly, app downloads could be a thing of the past within five years.

Now let’s take a look at Apple’s approach to the app store. Apple appears to be doubling down on the existing app distribution / discovery paradigm. The only change on this front was the introduction of app store search ads (which have been available on Google Play for a year, with no major impact). Instead, Apple’s major announcements focused on subscription-based revenue models to help developers better monetize digital content. Of course, it also helps that app revenue is the lone bright spot for the company as iPhone sales continue to decline.

Apple’s moves will certainly improve monetization in certain app store categories, notably Productivity, but it could hardly be called a drastic change to the app store model. This serves some developer needs, but it does not solve the app discovery challenge faced by consumers and the conversion rate issue that plagues developers. Time will tell if this was the right approach.

«

Is app monetisation more important for developers, or being able to get their apps onto peoples’ devices? The latter is comparatively easy, though neither is a cakewalk. Apple seems to be focussing first on the former, and fixing that quickly. Though Google could follow it quickly too.
link to this extract

 

Spotify monthly active user base reaches 100 million • Reuters

Mia Shanley:

»Swedish music streaming service Spotify said on Monday its user base had grown to 100 million, up from 75 million previously, as it pushed into new markets and despite competition from the likes of Apple Music.

Spotify has the music streaming industry’s biggest paid subscriber base, with 30 million users paying to listen, but the vast majority still tune in for free with commercial breaks.

Competition is fierce with Apple Music launched just last year and already claiming 13 million paid users while Alphabet’s Google competes with Google Music and Youtube.

«

Apple claimed last week to have 15 million subscribers – time to update the database, Reuters. Quite how it counts them (is each member of a family membership a “subscriber”, or only the main paying member?) isn’t yet clear.

What is clear is that Spotify can’t let a single Apple Music statistic go past without upping the ante. Notably, that 30m paid subscriber number hasn’t shifted since it released it in March. Possibly it is being conservative with its numbers, and only releasing bigger subscriber numbers when it needs to.

A related problem: those 70m non-paying listeners have to be monetised through advertising, and that growing inventory (= ad spots to fill) inevitably means falling ad prices, which means worse losses.
link to this extract

 

#THEDAO: Failing fast vs Failing unnecessarily • Preston J. Byrne

Byrne is extremely unimpressed with the setup which allowed millions of dollars worth of Ethereum cryptocurrency to be drained away, unlike some VCs who are saying “ooh, it’ll get fixed next release!”:

»Having lawyers – or legal-coders- involved in this is absolutely critical. The future doesn’t belong to the guy who just slings code or the guy who does the front-office function, but someone who can bridge the gap and do both – bringing the best of Silicon Valley’s approach to life to the professional services which run the rest of the world, and doing so in a way which gybes with local rule-frameworks. (Note, I run into this all the time when speaking with the banks – architects and front-office guys aren’t accustomed to talking to each other, or even considering themselves as part of the same team. I suspect this is a large contributor to most banks’ heaps of technical debt.)

Bridging the gap becomes especially important if you want to take your idea and turn it into an investable business, as many Solidity programmers do.

With respect to the DAO, there was a similar breakdown in communication – only this time between the wider community and the developers doing the codeslinging. Serious professional objections, from persons extremely well-versed on every layer of this conceptual stack, were made known very early. And not “this is a silly idea which will never work” kinds of objections, but “this is technically bankrupt and flies in the face of all best practice for what you are attempting to do” kinds of objections.

«

I still find the story around this impenetrable, but Byrne’s angry headshaking sounds like what ought to be the reaction.
link to this extract

 

Top ten reasons to doubt Trump is even a billionaire • Talking Points Memo

Josh Marshall, with the aforesaid ten, of which this is striking:

»During the research for his book Timothy o’Brien received estimates of Trump’s wealth ranging from $250m to $788m. Trump himself originally told O’Brien he was worth between $4b and $5b before dramatically revising down his estimate to $1.7 billion the same day. If we take $250m, $788m and $1.7b together and rough average them out we can get around $1 billion circa 2004/05. Today Trump claims he is worth $10 billion. This would require a tenfold run up in Trump’s wealth over roughly a decade. Even if we take Trump’s own estimate of $1.7 billion it would require a five fold run up over a decade. The problem is that Trump hasn’t done anything over that period that would account for that kind of wealth accumulation. Trump does very few major building projects these days and the few he does he does mainly with other people’s money. After the bankruptcy crises of 25 years ago, Trump shifted his business model from high profile real estate development to licensing and television. He licenses his name for hotels, buildings and golf courses on the high end and steaks, water, ties and more on the low end. This probably generates a massive amount of income for us mortals. But not many billions of dollars over a decade.

«

There must be a moment of truth, rather than truthiness, coming. Also: Trump fired his campaign manager on Monday. Things aren’t looking too clever.
link to this extract

 

Errata, corrigenda and ai no corrida: none notified.

Start up: Apple v Trump, Ev Williams v text, Google’s learning bet, Snapchat’s magazine plan, and more

Posted on by
Reply


iOS 10’s notifications are different – but there are plenty of other changes forthcoming in September (or so). Photo by tualamac on Flickr.

You can now sign up to receive each day’s Start Up post by email. You’ll need to click a confirmation link, so no spam.

A selection of 14 links for you. Save some for later – don’t bloat. I’m charlesarthur on Twitter. Observations and links welcome.

Ev Williams is the Forrest Gump of the internet • The Atlantic

Robinson Meyer:

»“The worst world, the scary version, is if the tricks to get attention are a skill developed and owned primarily by profit-driven companies,” [Ev Williams] told me. “I’d go back to the food analogy. What are people going to be consuming most of the time? They’re optimizing for clicks and dollars. Can a person who has a unique perspective play that game? Are they just going to get trounced?”

This is Medium’s reason for existing: to protect individual writers in the fierce and nasty content jungles. Resistance to the centralization generally is futile, he believes, citing Wu. “That’s the way the Internet works, and that’s the way humans work,” he says. “Efficiency and ROI and economies of scale and user experience—they’re all going to drive more things to consolidate. I kind of look at that as a force of nature. But if things consolidate, does that mean that everything is shit?”

That is the Medium appeal, in a nutshell. Keeping everything from being shit. It wants to do so by adopting many of the tics and habits of the original blogosphere—the intertextuality, the back-and-forth, the sense of amateurism—without being the open web. It will use its own custom metrics, like time-spent-reading, to decide who sees what stories; and it will tend to show your friends something if you “recommend” it. Medium, yes, will just be another platform, but it will run the open web in an emulator.

«

link to this extract

 

Google’s bold move to reinvent every device on the planet • Forbes

Miguel Helft:

»the techniques used to recognize images in Google Photos are able to power StreetView’s ability to “read” signs and Project Sunroof’s ability to identify rooftops that are suitable for solar panels based on aerial images. It’s also enabling a small experimental team at Google to effectively detect diabetic retinopathy, an eye disease that can lead to blindness, by looking at iris scans. “It’s a pretty significant shift,” Dean says. “Word is spreading throughout the company that there is this new capability to solve problems in this way,” he says, in reference to the new AI techniques.

What started as a research project with a handful of people has grown to perhaps hundreds–Dean refuses to say how many–who have developed algorithms, computer systems and, more recently, Google’s own chips, all customized for these AI approaches. (Google Brain’s software tools are known as TensorFlow and the chips as Tensor Processing Units.) As a result there are now more than 2,000 projects inside the company applying Google Brain’s capabilities to scores of products. Dean’s group has held machine-learning office hours, and thousands of Google engineers have gone through internal courses that can last weeks. “It went from being a research project to a mainstream engineering activity,” says John Giannandrea, an AI expert appointed by Pichai to lead the company’s search efforts.

«

You have to wade through a certain amount if you’re familiar with Google, but there are useful insights here too.
link to this extract

 

Facebook is wrong, text is deathless • Kottke

Tim Carmody on the suggestion from Facebook that “video is going to take over from text”:

»Maybe this is coming from deep within the literacy bubble, but:

Text is surprisingly resilient. It’s cheap, it’s flexible, it’s discreet. Human brains process it absurdly well considering there’s nothing really built-in for it. Plenty of people can deal with text better than they can spoken language, whether as a matter of preference or necessity. And it’s endlessly computable – you can search it, code it. You can use text to make it do other things.

In short, all of the same technological advances that enable more and more video, audio, and immersive VR entertainment also enable more and more text. We will see more of all of them as the technological bottlenecks open up.

And text itself will get weirder, its properties less distinct, as it reflects new assumptions and possibilities borrowed from other tech and media. It already has! Text can be real-time, text can be ephemeral – text has taken on almost all of the attributes we always used to distinguish speech, but it’s still remained text. It’s still visual characters registered by the eye standing in for (and shaping its own) language.

«

link to this extract

 

And another thing about Theranos… • LinkedIn

Sten Westgard lists the ten stories about Theranos you may have missed last week, which range from negative to more negative to neutral:

»There’s so much that’s happened that it’s hard to know where to start. Indeed, most of the stories have been covered by other news outlets already, and by real journalists. About the only additional insight we can add here is a closer reading of the lightly redacted inspection report. Because buried in that are some performance details that no one else seems to have noticed.

Let’s start with the QC [quality control] failure rates. The inspection report details that there were significant out-of-control results for many tests, sometimes up to 87% of QC results were out more than 2 standard deviations!!

«

link to this extract

 

No Signal (and other cellular drama) • YouTube

After last week’s wonderment about whether people in Star Wars are post- (or pre-) literate, here’s how screenplay writers deal with those damn mobile phones which could scupper plots in which people are meant to be out of contact and able to call help. Texas Chainsaw Massacre never had to deal with this (though probably would have in a scene like this).
link to this extract

 

‘Could he actually win?’ Dave Eggers at a Donald Trump rally • The Guardian

Dave Eggers went to a Trump rally in Sacramento, California, incognito, and found it more mixed (racially, sexually) than you’d expect, and more relaxed, but found this:

»He has reversed himself on nearly every major issue, often in the same week, and has offered scant specifics on anything in particular – though in Sacramento, about infrastructure, he did say, “We’re gonna have new roads, bridges, all that stuff”.

His supporters do not care. Nothing in Trump’s platform matters. There is no policy that matters. There is no promise that matters. There is no villain, no scapegoat, that matters. If, tomorrow, he said that Canadians, not Mexicans, were rapists and drug dealers, and the wall should be built on that border, no one would blink. His poll numbers would not waver. Because there are no positions and no statements that matter to them. There is only the man, the name, the brand, the personality they have seen on television.

Believing that Trump’s supporters are all fascists or racists is a grave mistake. This day in Sacramento presented a different picture, of a thousand or so regular people who thought it was pretty cool how Trump showed up in a plane with his name on it. How naughty it was when he called the president “stupid”. How funny it was when he said the word “huge” the peculiar way he does, without the “h” (the audience yelled back “uuuuge!”, laughing half with him, half at him). In the same way we rooted for Clay a few years ago when he showed up as an actual actor in a Woody Allen movie, the audience at a Trump rally is thinking, How funny would it be if this guy were across the table from Angela Merkel? That would be classic.

«

It’s long, but eminently worth reading. My next question is: will Eggers go to a Hillary Clinton rally, and what would he think of what he found there? I’d like to know.
link to this extract

 

Apple won’t aid GOP convention over Trump • POLITICO

Tony Romm:

»Apple has told Republican leaders it will not provide funding or other support for the party’s 2016 presidential convention, as it’s done in the past, citing Donald Trump’s controversial comments about women, immigrants and minorities.

Unlike Facebook, Google and Microsoft, which have all said they will provide some support to the GOP event in Cleveland next month, Apple decided against donating technology or cash to the effort, according to two sources familiar with the iPhone maker’s plans.

Apple’s political stand against Trump, communicated privately to Republicans, is a sign of the widening schism between Silicon Valley and the GOP’s bombastic presumptive nominee. Trump has trained his rhetorical fire on the entire tech industry, but he’s singled out Apple for particular criticism – calling for a boycott of the company’s products, and slamming CEO Tim Cook, over Apple’s stance on encryption.

«

link to this extract

 

Understanding the DAO hack for journalists • Medium

David Siegel, with a long long long explanation of how someone hacked a cryptocurrency (another event that’s becoming everyday) and siphoned off a ton:

»I will call the attacker a lone male, even though I have no idea if he is one. What happened next was interesting. In an open letter to The DAO and Ethereum Community, the attacker supposedly claimed that his “reward” was legal and threatened to take legal action against anyone who tried to invalidate his work. Several people pointed out that the cryptographic signature in this message wasn’t valid — it could be fake. But it’s well written and, from a certain point of view, well reasoned: the premise of smart contracts is that they are their own arbiters and that nothing outside the code can “change the rules” of the transaction.

Later, through an intermediary, the attacker claimed that he would put a stop to the organized “theft” of his property by rewarding miners (nodes) who don’t go along with the proposed soft fork, saying:

»

[S]oon we will have a smart contract to reward miners who oppose the soft fork and mines the transaction. 1 million ether + 100 btc will be shared with miners.

«

This is clearly a complex dynamic system. These messages from “The Attacker” cannot be verified, so we’ll have to wait and see what happens. Next, I will try to categorize the responses from the community.

«

I’m really glad I’m not the person writing the story about this if this is the “understanding” bit. First explain to a newsdesk what DAO is; then what Ethereum is; then smart contracts; then…
link to this extract

 

Ericsson Mobility Report Q1 2016 • Ericsson

Lots and lots of things in this, such as this:

»although teens reported the lowest cellular data consumption for video streaming apps among all age groups in both July 2014 and October 2015, the higher reliance on smartphones for video viewing at any time of the day means that teen video data consumption over cellular networks is growing rapidly.

Only 30–35 year olds have a higher growth rate than teens for cellular video streaming data usage. However, the overall mobile video data consumption (including both cellular and Wi-Fi) among this group is around 2.5 GB/month. That is only a fth of the teens’ data consumption and the potential for further growth is limited due to the fact that 30–35 year olds are still rooted in traditional TV viewing behavior.

Overall, teens are the heaviest users of data for smartphone video streaming apps and have the second highest rate of cellular video data consumption growth. Since we are witnessing a generational change, current teens are likely to increase their appetite for cellular data as they grow older – making them the most important group to watch for cellular operators.

«

But plenty more, such as the internet of things outnumbering smartphone subscriptions by 2018.
link to this extract

 

Snapchat is starting Real Life, an online magazine about technology • VentureBeat

Jordan Novet:

»In a blog post today describing the new initiative, Snapchat employee and social media critic Nathan Jurgenson writes that “Snapchat is now funding Real Life.” In an email to VentureBeat, he declined to elaborate on the nature of the funding, but he did confirm that Real Life is “owned” by Snapchat.

“Real Life will publish essays, arguments, and narratives about living with technology,” Jurgenson writes. “It won’t be a news site with gadget reviews or industry gossip. It will be about how we live today and how our lives are mediated by devices.” (This sounds a little like the turf of New York Magazine‘s recently launched Select All.) The publication will cover beauty, power, privacy, and relationships, among other things, and “we aim to address the political uses of technology, including some of the worst practices both inside and outside the tech industry itself,” writes Jurgenson.

So now Snapchat will technically have web content that is visible on desktop computers. No longer will Snapchat be constrained to mobile devices. And, at least initially, the medium will be primarily text, unlike the video stories and snaps the Southern California company has become known for.

«

Unfathomable. How does this do anything for Snapchat?
link to this extract

 

The Talk Show ✪: Live From WWDC 2016, With Phil Schiller and Craig Federighi • Daring Fireball

»Recorded in front of a live audience in San Francisco, John Gruber is joined by Phil Schiller and Craig Federighi to discuss the news from WWDC: WatchOS 3, MacOS 10.12 Sierra, iOS 10, and more.

«

There’s also a transcript. Last year it was just Schiller. (“Just” Schiller.) I guess they can pick from Schiller, Federighi and Eddy Cue for a few years before it has to aim for the top with Cook. After whom, what?
link to this extract

 

All the hidden, awesome stuff in iOS 10 Apple didn’t announce • Lifehacker

Thorin Klosowski:

»iOS updates aren’t as exciting as they used to be, so the best stuff is often the little features that slip through the keynote cracks but make your iPhone or iPad work much better. Case in point, some of the hidden stuff in early iOS 10 betas is way more exciting than what Apple actually announced this week.

«

It isn’t all but it’s a few of the more fun things – alarm redesign, Maps remembering where you parked if you used it to navigate in a car, no more “slide to unlock”, a few more. I think the death of “slide to unlock” (and its companion, where Music controls in Control Centre are now to the right) is going to be the one that causes the most perplexity.
link to this extract

 

The elements of stickers • Andreessen Horowitz

Connie Chan, a partner at venture capital company Andreessen Horowitz:

»What is surprising — especially when compared to the more mature messaging ecosystem in Asia — is that many people still tend to treat stickers (i.e., the ability to easily incorporate pre-set images into texts) as just-for-fun frivolity, when they’re an important visual digital language fully capable of communicating a nuanced range of thoughts. For example, a single sticker could convey very different messages: “I’m so hungry I could collapse” or “I miss you” or “I’m sound asleep snoring”. Complex feelings, actions, punch lines, and memes are all possible with stickers.

They are an acceptable response to “end” a real-time back and forth conversation (great for punchlines). They are a low-risk way of saying hi and initiating a chat with an acquaintance. And they reduce the social friction of saying something emotional in text form; this is especially helpful in a culture that is known to be less outwardly expressive even to one’s own family members and friends (where it is far less awkward to send a virtual-fistbump sticker than it is to tell someone directly that they’re a wonderful friend).

And sometimes stickers can convey what words cannot! This form of visual communication has become so popular in Asia — especially in China’s WeChat and Japan’s LINE — that it is not uncommon to see a deep thread of multiple messages without a single word. They’re not just for those crazy young kids. More notably, stickers are commonly used in professional, not just personal, chats as well. Not so frivolous after all. In fact, stickers are so core to the success of Line, that its CEO actually credited them as the “turning point” for that app. He shared that it took Line Messenger almost four months to find its first two million users … but after stickers were launched, it took only two days to find the next million. The company now makes over $270m a year just from selling stickers.

«

This is essential to understanding why Apple has gone so big on stickers for iOS 10’s iMessage. Chan is highly worth reading on all these topics.
link to this extract

 

How hired hackers got “complete control” of Palantir • BuzzFeed News

William Alden on how Veris Systems was hired to hack into Palantir:

»Even Palantir’s defense efforts were visible to the red team. The intruders found an “InfoSec Onboarding” page on the wiki that detailed Palantir’s security infrastructure. They monitored security devices and “ensured that their actions were not being logged.”

This was when, according to the report, the red team intruders had “complete control” of the Palantir domain. Their final task was to break into the Mac laptops of information security employees — the fortress guards. This they did, using a system that typically sent out software updates, and soon were able to get passwords and screenshots, review saved files, and “observe all user activity,” the report says.

They were finally caught while attempting to upload a screenshot to one of their own servers, according to the report. A piece of security software called Little Snitch — which regulates data sent out from a computer to the internet — was installed on one of the information security employees’ laptops, and it flagged the suspicious upload attempt, the report says. Little Snitch, while popular in the cybersecurity world, was not standard software for these employees, according to one person familiar with the matter.

Soon, Palantir security employees identified the red team’s attack tools and set up firewalls to block communications to the red team servers. These defenders “successfully demonstrated the ability to trace malicious activity across the domain and take the appropriate steps to neutralize an insider threat,” the report says.

But the red team still had an edge.

«

Veris was let through the firewall on purpose, to see what would happen if someone was spearphished. Turns out: a lot.
link to this extract

 

Errata, corrigenda and ai no corrida: none notifed.

Start up: Facebook’s video hope, Amy on Outlook, Apple’s neural nets, a Trump rally in Greensboro, and more

Posted on by
Reply

06

Deleting the default apps on iOS 10 will get rid of them, right? Wrong. Photo by tuaulamac on Flickr.

You can now sign up to receive each day’s Start Up post by email. You’ll need to click a confirmation link, so no spam.

A selection of 12 links for you. Use them wisely. I’m charlesarthur on Twitter. Observations and links welcome.

Facebook is predicting the end of the written word • Quartz

Cassie Werber:

»Facebook has arguably made us all writers, since it has become the medium of choice for millions to share their views and life experiences. But in five years that creativity may look very different. Facebook is predicting the end of the written word on its platform.

In five years time Facebook “will be definitely mobile, it will be probably all video,” said Nicola Mendelsohn, who heads up Facebook’s operations in Europe, the Middle East and Africa, at a conference in London this morning. Mark Zuckerberg, Facebook’s CEO, has already noted that video will be more and more important for the platform. But Mendelsohn went further, suggesting that stats showed the written word becoming all but obsolete, replaced by moving images and speech.

“The best way to tell stories in this world, where so much information is coming at us, actually is video,” Mendelsohn said. “It conveys so much more information in a much quicker period. So actually the trend helps us to digest much more information.”

«

Not buying this.
link to this extract

 

How ‘deleting’ built-in Apple apps works in iOS 10 • iMore

Rene Ritchie:

»when you delete a built-in app, you don’t really delete it. You do remove the icon from the Home screen, the user data is flushed, and the hooks into the system for things like default links and Siri handling are removed. But, it doesn’t delete the actual app binary.

There are two reasons for this:

• Apple’s built-in apps are very, very small, taking up only 150 MB of storage. That’s because they wrap a lot of core functionality and so don’t introduce a lot of extraneous code or assets.

• When a version of iOS is released, Apple signs it so your iPhone or iPad can verify it’s legitimate and hasn’t been tampered with by a third party. That code signing covers the entirety of iOS, including built-in apps. If everyone had different apps, some present, some not, the current form of signing security wouldn’t work.

«

Deleting the user data might save a fair amount of storage, though.
link to this extract

 

X.ai works with Microsoft Outlook.com • Business Insider

Matt Weinberger:

»For the last two years, the popular x.ai virtual personal assistant has been helping Google Calendar users manage their meetings.

Today, x.ai is finally coming to Microsoft calendars, with support for Office 365 and Outlook.com, as the company moves closer to the release of its paid business edition later this year.

«

Amy is a really terrific system – I don’t know why Google or Microsoft hasn’t snapped up x.ai.
link to this extract

 

BNNS • Apple Developer Documentation

»Basic neural network subroutines (BNNS) is a collection of functions that you use to implement and run neural networks, using previously obtained training data.

«

Embedded in all four platforms (iOS, tvOS, watchOS, OSX/MacOS):

»BNNS supports implementation and operation of neural networks for inference, using input data previously derived from training. BNNS does not do training, however. Its purpose is to provide very high performance inference on already trained neural networks.

«

Does Android have anything comparable?
link to this extract

 

The perils and promises of gene-drive technology • The New Yorker

Michael Specter:

»Normally, the progeny of any sexually reproductive organism receives half its genome from each parent. For decades, however, biologists have been aware that some genetic elements are “selfish”: evolution has bestowed on them a better-than-fifty-per-cent chance of being inherited. But, until scientists began to work with Crispr, which permits DNA to be edited with uncanny ease and accuracy, they lacked the tools to make those changes.

Then the evolutionary biologist Kevin Esvelt, who runs the Sculpting Evolution Group at M.I.T.’s Media Lab, realized that, by attaching a gene drive to a desired DNA sequence with crispr, you could permanently alter the genetic destiny of a species. That’s because, with crispr, a change made on one chromosome would copy itself in every successive generation, so that nearly all descendants would inherit the change. A mutation that blocked the parasite responsible for malaria, for instance, could be engineered into a mosquito and passed down every time it reproduced. Within a year or two, none of the original mosquito’s offspring would be able to transmit the infection. And if gene drives work for malaria they ought to work for other mosquito-borne diseases, such as dengue, yellow fever, and Zika.

This is tremendously promising news, but nothing so powerful comes without risk—and there has never been a more powerful biological tool…

…Pretty soon, we are going to have to make some of the most pressing decisions we have ever made about how, whether, and when to deploy a new technology.

«

link to this extract

 

The end of reflection • The New York Times

Teddy Wayne:

»By 2012, Google engineers had discovered that when results take longer than two-fifths of a second to appear, people search less, and lagging just one quarter of a second behind a rival site can drive users away.

“That hints at the way that, as our technologies increase the intensity of stimulation and the flow of new things, we adapt to that pace,” [author of The Shallows, Nichola] Carr said. “We become less patient. When moments without stimulation arise, we start to feel panicked and don’t know what to do with them, because we’ve trained ourselves to expect this stimulation — new notifications and alerts and so on.”

What this often translates to in the discourse of the internet is demand for immediate and perfunctory “hot takes” rather than carefully weighed judgments, whether they’re about serious or superficial matters.

Mr. Carr also noted counterarguments: Formulating relatively simple thoughts on the internet can yield more complex ones through real-time exchanges with others, and people whose reflex is to post a notion hastily rather than let it sit may not have been the most deliberative thinkers in a pre-smartphone time, either.

Nevertheless, he sees our current direction as indicative of “the loss of the contemplative mind,” he said.

«

link to this extract

 

What is Differential Privacy’? • A Few Thoughts On Cryptographic Engineering

Matthew Green of Johns Hopkins University, explaining the system Apple says it’s using for its machine learning system:

»A much more promising approach is not to collect the raw data at all. This approach was recently pioneered by Google to collect usage statistics in their Chrome browser. The system, called RAPPOR, is based on an implementation of the 50-year old randomized response technique. Randomized response works as follows:

• When a user wants to report a piece of potentially embarrassing information (made up example: “Do you use Bing?”), they first flip a coin, and if the coin comes up “heads”, they return a random answer — calculated by flipping a second coin. Otherwise they answer honestly.

• The server then collects answers from the entire population, and (knowing the probability that the coins will come up “heads”), adjusts for the included “noise” to compute an approximate answer for the true response rate.

Intuitively, randomized response protects the privacy of individual user responses, because a “yes” result could mean that you use Bing, or it could just be the effect of the first mechanism (the random coin flip). More formally, randomized response has been shown to achieve Differential Privacy, with specific guarantees that can adjusted by fiddling with the coin bias.

RAPPOR takes this relatively old technique and turns it into something much more powerful. Instead of simply responding to a single question, it can report on complex vectors of questions, and may even return complicated answers, such as strings – e.g., which default homepage you use. The latter is accomplished by first encoding the string into a Bloom filter – a bitstring constructed using hash functions in a very specific way. The resulting bits are then injected with noise, and summed, and the answers recovered using a (fairly complex) decoding process.

«

I think “it’s complicated” will probably do as a first pass.
link to this extract

 

Android share growth slows after historic gains last period • Kantar Worldpanel

»“In Great Britain, both Android and iOS had higher market share in the three months ending April 2016. Android represented 58.5% of the market in that period, a gain of 4.1% year-on-year,” said Dominic Sunnebo, Business Unit Director for Kantar Worldpanel ComTech Europe. “And for iOS, this term showed the first increase since October 2015, though modest at just 0.4%, from 34.7% to 35.1%. Android gains came from Windows phone owners switching, a trend that produced nearly 10% of new Android customers, while 21.8% of new iOS buyers switched from Android.”

«

In other words: Windows Phone, the platform, is burning, and not in a good way. This will sound familiar to students of history, and not in a good way either.

»

“In Urban China, Android share rose 4.8% year-over-year, and 1.1% period-over-period, to capture 78.8% of smartphone sales in the three months ending April 2016,” noted Tamsin Timpson, Strategic Insight Director at Kantar Worldpanel ComTech Asia. “While movement from featurephones to smartphones has slowed significantly in developed markets like the US and EU5, this still makes up a significant proportion of smartphone sales in Urban China. Nearly a third of Android users during this time were purchasing their first smartphone, in contrast to iOS buyers, of whom only 14% were first-time smartphone customers.”

«

That doesn’t tell us whether Chinese iOS buyers were moving from Android in any measurable quantity. But clearly Android is still effective at gaining from featurephones.
link to this extract

 

A Trump rally in Greensboro • · Storify

Jared Yates Sexton went along and tweeted what he saw and heard, with this as the tagline:

»”Anger in here is palpable”: in which a sane man live tweets insanity.

«

It really is scary. (Over 170,000 views at the time of tagging.) A question one might like to consider is whether Trump would let his wife walk unaccompanied through the car park following one of his rallies. (There’s more of Sexton’s work on this blog.)
link to this extract

 

OnePlus X series is no more, says CEO • Engadget

Richard Lai:

»While it’s common practice for smartphone makers to offer two or three product lines to cover all the bases, OnePlus has recently decided to go from two to one. At the OnePlus 3 launch event in Shenzhen today, CEO Pete Lau confirmed that his company’s more affordable offering, the OnePlus X, will not have a followup model. That’s not to say it was a bad phone (even we liked it) nor was it unpopular, but Lau reasoned that OnePlus will instead focus on just one “true flagship” line from now on, in order to strengthen its foundation – something that Lau admitted his team neglected last year – rather than fighting the low-end price war.

«

OnePlus is on thin margins and (comparatively) low volumes, so it has to shift towards premium pricing to survive.
link to this extract

 

Now Peter Thiel’s lawyer wants to silence reporting on Trump’s hair [Updated] • Gawker

J K Trotter:

»But if you were under the impression that praise-worthy journalism [investigating whether Donald Trump’s hair is a $60,000 wig/weave – which I would think is very likely indeed] is somehow inoculated against campaigns like Thiel’s, you’d be mistaken. Last week, Thiel’s lawyer-for-hire, Charles J. Harder, sent Gawker a letter on behalf of Ivari International’s owner and namesake, Edward Ivari, in which Harder claims that Feinberg’s story was “false and defamatory,” invaded Ivari’s privacy, intentionally inflicted emotional distress, and committed “tortious interference” with Ivari’s business relations. Harder enumerates 19 different purportedly defamatory statements—almost all of which were drawn from several publicly available lawsuits filed against Ivari.

Harder’s demands included the immediate removal of the story from Gawker, a public apology, the preservation of “all physical and electronic documents, materials and data in your possession” related to the story, and, notably, that we reveal our sources.

«

Thiel’s lawyer’s filing is nonsense; and Gawker now does not give a flying one how much it offends either of them. When you’re on Death Row, death threats hardly scare you.
link to this extract

 

Shutterbugs, rejoice: Apple’s iOS 10 will shoot raw photos • CNET

Stephen Shankland:

»Apple’s next-gen iOS 10 software adds a new programming interface that will let camera apps retrieve unprocessed raw photo data from the camera hardware, according to Apple developer documentation. Google’s Android has supported raw photos since the release of the Lollipop version in 2014.

There’s a good reason Apple didn’t include raw photo support in its top-10 list of new iOS 10 features unveiled at its annual Worldwide Developer Conference (WWDC) Monday. Raw photography is complex and too much of a hassle for most people to bother with. But with photography now so central to mobile phones, and with photo enthusiasts being such an active and visible type of customer, raw photo support is a major improvement. Raw photos should help Apple’s iPhones keep their place atop the list of most popular cameras on Flickr, the photo-sharing site.

«

link to this extract

 

Errata, corrigenda and ai no corrida: none notified.

Start up: Bitcoin’s nightmare, the cheating economy, how Snapchat took off, Oculus spurns Macs, and more

Posted on by
2

SIM swaps are leading to bank fraud. Photo by mroach on Flickr.

You can now sign up to receive each day’s Start Up post by email. You’ll need to click a confirmation link, so no spam.

A selection of 10 links for you. Use them wisely. I’m charlesarthur on Twitter. Observations and links welcome.

How Snapchat built a business by confusing olds » Bloomberg BusinessWeek

Max Chafkin and Sarah Frier:

»Compared with Twitter or Facebook, Snapchat can seem almost aggressively user-unfriendly. If you’re new to the app and looking for posts by your kid, your boyfriend, or DJ Khaled, good luck. It’s hard to find somebody without knowing his or her screen name. This is by design. “We’ve made it very hard for parents to embarrass their children,” [Snapchat founder Evan] Spiegel said at a conference in January. “It’s much more for sharing personal moments than it is about this public display.”

Spiegel, who declined to be interviewed, has been cagey about Snapchat’s business prospects. Its annual revenue is small—perhaps $200m, according to several press reports—but it has already drawn many big-name advertisers. Earlier this year, PepsiCo, Amazon.com, Marriott International, and Budweiser paid more than $1m to have their ads appear within the company’s Super Bowl coverage, according to a person familiar with the deals. And because Snapchat has yet to really try to sell ads to the small and midsize businesses that make up most of Google’s and Facebook’s customer base, there’s a lot of potential.

As Facebook has transformed from a slightly wild place to a communications tool for parents, teachers, and heads of state, Snapchat’s more playful ethos, and the fact that anything posted on it disappears in 24 hours, has made it the looser, goofier social network. “You’re sending this ephemera back and forth to your friends,” says Charlie McKittrick, the head of strategy at Mother New York, an ad agency. “It’s the detritus of life. But it’s really funny.” Last September, while Mark Zuckerberg hosted Indian Prime Minister Narendra Modi on Facebook’s campus, the big news at Snapchat’s offices in Venice was a feature called Lenses, which makes your selfies look like you’re vomiting a rainbow.

«

link to this extract

 

We’re moving away from torrents, so whats next? » Strike

“Andrew”:

»As you can see if just a teeny bit taxing on my server, so as of today I wanted to officially annouce that Strike will no longer focus on torrents, in fact I’ve decided to phase Strike into creating open source utilities that help every day life. Our first project is already under development and called Ulterius, an open source C# based framework that allows you to remotely manage windows based systems, all from any HTML5 enabled browser…

…Q: Will you ever do torrent related things again?

A: Most likely not. It’s easier to create completely original content than to attempt to ride the tails of existing content. While I found P2P technology fun, and I’ll continue to follow it and maybe develop stuff around it. I don’t foresee myself ever hosting Anything as a service in the future.

«

Combination of lawsuits against others, and the gigantic bandwidth demand on his site. Mostly the bandwidth, it seems.
link to this extract

 

Will we compile? » ROUGH TYPE

Nick Carr:

»Getting machines to understand, and speak, the language used by people — natural language processing — has long been a central goal of artificial intelligence research. In a provocative new interview at Edge, Stephen Wolfram turns that goal on its head. The real challenge, he suggests, is getting people to understand, and speak, the language used by machines. In a future world in which we rely on computers to fulfill our desires, we’re going to need to be able to express those desires in a way that computers can understand…

…Computers can’t choose our goals for us, Wolfram correctly observes. “Goals are a human construct.” Determining our purposes will remain a human activity, beyond the reach of automation. But will it really matter? If we are required to formulate our goals in a language a machine can understand, is not the machine determining, or at least circumscribing, our purposes? Can you assume another’s language without also assuming its system of meaning and its system of being?

«

Very deep questions underlying this. And speaking of controlling machines through spoken language..
link to this extract

 

Amazon adds the $130 Amazon Tap and the $90 Echo Dot to the Echo family » Techcrunch

Sarah Buhr:

»The Echo has received more than 33,000 Amazon reviews at a nearly five-star rating since launching in late 2014 and was one of the best-selling items going for more than $100 over the holidays. Amazon has not released sales figures for Echo, but its rise in popularity and the ability to build upon and integrate with the companion Alexa API have moved the Echo front and center as a must-have device for the smart home.

Amazon is now introducing two new members to the Echo family with slightly different uses in hopes of achieving a similar reaction: Amazon Tap is a portable version of the original Echo, and Echo Dot is a tiny, hockey-puck-sized version that includes a built-in line-out connector to hook into your choice of speaker.

«

link to this extract

 

Online break-in forces bank to tighten security » BBC News

Shari Vahl:

»Two major high street banks will change security procedures after journalists from BBC Radio 4’s You and Yours programme broke into an account online and removed money.
Recently bank customers accounts have been successfully attacked by criminals who divert mobile phone accounts.

Criminals persuade phone providers to divert mobile phone numbers in what is sometimes called “SIM swap fraud”.

Some banks text security details when customers forget their details.

The activation codes sent by text to mobile phones also allow payments to be made from an account.

The scam works by blocking the genuine phone. The owner is unaware of why the phone has been blocked and allows the criminal – who now has control of their phone – to syphon money from their bank account.

You and Yours has been contacted by dozens of people affected by the scam. All say they have never revealed their security details to anyone, and the that first they knew something was wrong was their mobile phone going dead.

«

Wow.
link to this extract

 

Chinese ISPs caught injecting ads and malware into web pages » The Hacker News

Rakesh Krishnan:

»Chinese Internet Service Providers (ISPs) have been caught red-handed injecting advertisements as well as malware through their network traffic.

Three Israeli researchers uncovered that the major Chinese-based ISPs named China Telecom and China Unicom, two of Asia’s largest network operators, have been engaged in an illegal practice of content injection in network traffic.

Chinese ISPs had set up many proxy servers to pollute the client’s network traffic not only with insignificant advertisements but also malware links, in some cases, inside the websites they visit.
If an Internet user tries to access a domain that resides under these Chinese ISPs, the forged packet redirects the user’s browser to parse the rogue network routes. As a result, the client’s legitimate traffic will be redirected to malicious sites/ads, benefiting the ISPs.

«

link to this extract

 

TensorFlow for Poets » Pete Warden’s blog

»I want to show how anyone with a Mac laptop and the ability to use the Terminal can create their own image classifier using TensorFlow, without having to do any coding.

I feel very lucky to be a part of building TensorFlow, because it’s a great opportunity to bring the power of deep learning to a mass audience. I look around and see so many applications that could benefit from the technology by understanding the images, speech, or text their users enter. The frustrating part is that deep learning is still seen as a very hard topic for product engineers to grasp. That’s true at the cutting edge of research, but otherwise it’s mostly a holdover from the early days. There’s already a lot of great documentation on the TensorFlow site, but to demonstrate how easy it can be for general software engineers to pick up I’m going to present a walk-through that takes you from a clean OS X laptop all the way to classifying your own categories of images. You’ll find written instructions in this post, along with a screencast showing exactly what I’m doing.

«

Warden was at Jetpac, which was bought by Google because of its expertise at machine learning and image classification. This is the one to follow to dive into deep learning (aka machine learning, aka AI).
link to this extract

 

Oculus’ Palmer Luckey will consider Mac support if Apple ‘ever releases a good computer’ » Shacknews

Daniel Perez:

»We spoke to Oculus VR founder Palmer Luckey recently during an Xbox press event where we took the opportunity to ask him some questions regarding the future of his company, and his product, the Oculus Rift.

One question we were dying to ask is he sees a future for the Oculus Rift with Apple computers. When asked if there would ever be Mac support for the Rift, Palmer responds by saying “That is up to Apple. If they ever release a good computer, we will do it.”

Palmer continues to clarify what he meant by that blunt statement by saying “It just boils down to the fact that Apple doesn’t prioritize high-end GPUs. You can buy a $6,000 Mac Pro with the top of the line AMD FirePro D700, and it still doesn’t match our recommended specs. So if they prioritize higher-end GPUs like they used to for a while back in the day, we’d love to support Mac. But right now, there’s just not a single machine out there that supports it.”

«

There aren’t that many Windows PCs that support it, either. Wonder if this is a high priority for Apple just now.
link to this extract

 

The cheating economy » Medium

Doug Bierend on Studypool, which lets students “hire” tutors for “help understanding their homework” – which the students of course translate into “doing their homework”, and give bad grades to those tutors who don’t comply:

»Rarely is the sharing model of enterprise, epitomized by the likes of Uber and Airbnb, sensitive to the costs incurred by its host system — those two companies are hardly compelled to preserve the integrity of the “legacy” cab companies and hoteliers they are undercutting. Likewise, success for this platform isn’t determined by whether it actually helps people learn. After all, optimizing and reducing the latency in busing information from one place to another makes sense — a lot of sense — for servers and data, but where brains and ideas are concerned, learning isn’t always efficient. And any approach that offers a backdoor — knowingly or not—where intellectual honesty is concerned is bound to reap the patronage of the many people willing to buy an answer or grade rather than earn it.

«

A passing thought: Bierend is a professional journalist (it shines through in this piece – read it all), and this appeared in “Bright” – which is funded by the Gates Foundation, and subsumed into Medium. The brave new world where a non-profit created from the money out of a brief technology monopoly pays for journalism published on a site created from the money paid to the creator of free publishing platforms (Blogger and Twitter) that were funded by advertising. Who says there aren’t new business models for journalism?
link to this extract

 

Bitcoin’s nightmare scenario has come to pass » The Verge

Ben Popper:

»Over the last year and a half a number of prominent voices in the Bitcoin community have been warning that the system needed to make fundamental changes to its core software code to avoid being overwhelmed by the continued growth of Bitcoin transactions. There was strong disagreement within the community, however, about how to solve this problem, or if the problem would ever materialize.

This week the dire predictions came to pass, as the network reached its capacity, causing transactions around the world to be massively delayed, and in some cases to fail completely. The average time to confirm a transaction has ballooned from 10 minutes to 43 minutes. Users are left confused and shops that once accepted Bitcoin are dropping out.

«

Remember how Mike Hearn, who saw this problem coming and proposed an increase in block size which would have headed it off, was criticised to hell and back for being “misleading”? I bet he’s feeling vindicated now. Wonder how his then-critics feel. (Update: not great, apparently, since the Pond Politics page I referenced has been deleted in the meantime.)
link to this extract

 

Errata, corrigenda and ai no corrida: none notified.

Start up: hedge funds like AI, Facebook’s close separation, what if Twitter died?, BlackBerry cuts, and more

Posted on by
3

A break like this, affecting the home button, is probably going to lead in time to an #error53 fault if you don’t get it repaired by Apple. But what causes it, exactly? Photo by wZa HK on Flickr.

You can now sign up to receive each day’s Start Up post by email. You’ll need to click a confirmation link, so no spam.

A selection of 12 links for you. Use them wisely. I’m charlesarthur on Twitter. Observations and links welcome.

Will AI-powered hedge funds outsmart the market? » MIT Tech Review

Will Knight:

Anthony Ledford, chief scientist of MAN AHL, explains that the company is exploring whether techniques like deep learning might lend themselves to finance. “It’s at an early stage,” Ledford says. “We have set aside a pot of money for test trading. With deep learning, if all goes well, it will go into test trading, as other machine-learning approaches have.”

Trading might seem like an obvious place to apply deep learning, but actually it isn’t clear how comparable the challenge of finding subtle patterns in real-time trading data is to, say, spotting faces in digital photographs. “It’s a very different problem,” Ledford admits.

Academic experts also sound a note of caution. Stephen Roberts, a professor of machine learning at Oxford University, says deep learning could be good “for extracting hidden trends, information, and relationships,” but adds that it “is still too brittle with regard to handling of high uncertainty and noise, which are prevalent in finance.”

You just know that this isn’t really going to work, but also that it’s going to be used by a ton of funds to try to get ahead of the market – a market composed of other funds also trying to use the same processes.
link to this extract

 

iOS security – iOS 9 or later » Apple

Let’s try to get on top of this #error53 stuff:

During an iOS upgrade, iTunes (or the device itself, in the case of OTA software updates) connects to the Apple installation authorization server and sends it a list of cryptographic measurements for each part of the installation bundle to be installed (for example, LLB, iBoot, the kernel, and OS image), a random anti-replay value (nonce), and the device’s unique ID (ECID).

The authorization server checks the presented list of measurements against versions for which installation is permitted and, if it finds a match, adds the ECID to the measurement and signs the result. The server passes a complete set of signed data to the device as part of the upgrade process. Adding the ECID “personalizes” the authorization for the requesting device. By authorizing and signing only for known measurements, the server ensures that the update takes place exactly as provided by Apple.

The boot-time chain-of-trust evaluation verifies that the signature comes from Apple and that the measurement of the item loaded from disk, combined with the device’s ECID, matches what was covered by the signature.

These steps ensure that the authorization is for a specific device and that an old iOS version from one device can’t be copied to another. The nonce prevents an attacker from saving the server’s response and using it to tamper with a device or otherwise alter the system software.

To recap, with #error53, people who have had third-party replacements of screens and/or home buttons on the iPhone 6/Plus and 6S/Plus (but not the 5S) find that it works fine – though they can’t use TouchID (it’s greyed out as an option). But when they do an OS update, the phone bricks: can’t get data, can’t restore.

So my understanding of this is: the reason why devices which have had third-party replacement parts only brick after an OS update, yet work fine before it, is this: on trying to install the update they connect to the auth server. The server decides that the cryptographic measurements no longer match what it has on record. So it decides the chain of trust is broken, and effectively shuts down the device.

But it’s poor decision-making by Apple, and equally poor communication. Why doesn’t it happen on the 5S? Update: because the 5S doesn’t have NFC for Apple Pay. (Thanks, Andy.) What’s the process that Apple uses when it does the repair to revalidate the TouchID system (which fails even with valid parts)? Why can’t the system tell that it’s just TouchID that’s affected? The safety process has overshot its requirements. Every part of what happens makes sense from a security perspective  – but not if considering that many people will get third-party repairs.
link to this extract

 

Three and a half degrees of separation » Research at Facebook

How connected is the world? Playwrights, poets, and scientists have proposed that everyone on the planet is connected to everyone else by six other people. In honor of Friends Day, we’ve crunched the Facebook friend graph and determined that the number is 3.57. Each person in the world (at least among the 1.59 billion people active on Facebook) is connected to every other person by an average of three and a half other people. The average distance we observe is 4.57, corresponding to 3.57 intermediaries or “degrees of separation.” Within the US, people are connected to each other by an average of 3.46 degrees.

Our collective “degrees of separation” have shrunk over the past five years. In 2011, researchers at Cornell, the Università degli Studi di Milano, and Facebook computed the average across the 721 million people using the site then, and found that it was 3.74 [4,5]. Now, with twice as many people using the site, we’ve grown more interconnected, thus shortening the distance between any two people in the world.

Apparently my average is 3.26 so ya boo. Zuckerberg is 3.17. Sheryl Sandberg is 2.92 – blimey.
link to this extract

 

On your cute release notes » The Brooks Review

Ben Brooks:

We’ve all seen them. Notes about a fictional engineer who was hired and then fired. A cute story about something completely irrelevant to the matter at hand. Recipe for ‘squash bug soup’ or something along those lines.

With disturbingly increasing frequency, companies are deciding to let their marketing departments handle their release notes instead of the engineering team or product manager.

And we are all worse off for it.

As a user I mostly look at release notes to find out about one (or more) of three things:

• Have you added something new to the app which will make it better for me? That is: what are the new features, what do those features do, and perhaps how do I get to them.
• Have you fixed that bug which was making the app hard for me to use, perhaps even impossible for me to use? Aka: What bugs did you fix?
• How active is development on this app? Before I invest or move to most apps I look at recent release notes to get a sense of whether they are in maintenance mode (just major bug fixes), or under some kind of active development (minor bug fixes and feature releases, optimized for current version of iOS, etc).

link to this extract

 

BlackBerry cuts 200 jobs in Ontario and Florida to trim costs » Reuters

Alastair Sharp:

The layoffs will affect 75 manufacturing jobs in Sunrise, Florida, a state government website showed.

The company also confirmed that Gary Klassen is one of the people who has departed in the latest round of cuts. Klassen was one of its longest-tenured employees and the inventor of its BBM messaging service.

One source familiar with the matter, who declined to be identified due to the sensitivity of the issue, said many of the Canadian cuts were people working on its BB10 handset software at its Waterloo, Ontario, headquarters.

A spokeswoman for BlackBerry declined to comment on which divisions will be affected by the cuts, but said the company stood by its commitment to release further updates on its BB10 software.

BB10 is so, so dead.
link to this extract

 

“Dangerous ramifications” » Medium

Rohin Dharmakumar, with some examples of things that didn’t happen:

In January 2015, users of Microsoft’s Office in India were suddenly greeted with a pop-up asking them to “Support Microsoft Office”. The Indian government under PM Narendra Modi was said to be formulating an “Open Source Policy” under which all government offices were to either mandate or prefer open-source software for official work.

Clicking the “Support Office” button caused Microsoft to send the PMO and the Ministry of IT a letter from the user’s name with a pre-determined format. It said the user’s loved Microsoft’s products and wanted their government interactions to be based on the same. “I urge you not to ban Microsoft Office,” it ended.

The same message popped up on users of various Microsoft products in India – Windows, XBox, Windows Phone, Skype etc.

Within a few weeks, over 7 million emails had been sent in support to Microsoft.

“Support Monsanto”

In January 2014, farmers in the southern Indian state of Karnataka were surprised to see a notice attached to every bag of seed they bought from Mahyco, the market leader.

“Tell the Karnataka Govt. not to ban MMB”, said the notice. MMB was Monsanto-Mahyco Biotech, the joint-venture that licensed Monsanto’s crop technologies in India.

He has some more examples of things that didn’t happen – and then one which did.
link to this extract

 

Why most A/B tests give you bullshit results » Mixpanel

We’ve all seen the articles. Company X increases conversions 38% with this simple trick. Hell, I’ve written some of them.

But those success stories have hidden the grey underbelly of testing and experimentation.

AppSumo revealed that only 1 out of 8 tests produce results. Kaiser Fung estimates that 80 to 90 percent of the A/B tests he’s run yield statistically insignificant results.

Yet many new testers walk into A/B testing thinking it’ll be quick and easy to get results. After running a handful of simple tests, they think they’ll find the right color for this button or the right tweak to that subject line, and conversions will, poof, increase by 38% like magic.

Then they start running tests on their apps or sites, and reality suddenly sets in. Tests are inconclusive. They yield “statistically insignificant” results and no valuable insights about the product or users. What’s happening? Where’s that 38% bump and subsequent pat on the back?

Don’t get frustrated. If you’re going to be running A/B tests, you’re going to have some tests that fail to produce meaningful results you can learn from. But if you run good tests, you’ll have fewer failures and more successes.

link to this extract

 

Advice for companies with less than one year of runway » The Macro

Dalton Caldwell:

Let’s imagine that you are the founder of a company that has successfully raised an angel or institutional round and are currently in a situation where you have 12 months or less of runway.

The hardest part of dealing with a low runway situation is managing your own psychology. You have to simultaneously manage your own anxiety to not be overly negative about your prospects, but also not be irrationally positive. It’s a delicate balance.

Watch companies do the various things in this post over the next year or so.
link to this extract

 

Technology: the rift with reality » FT.com

Tim Bradshaw:

With so many [virtual reality] headsets hitting the market this year, the challenge may be figuring out what people will do with them. Video games are seen as the first popular application, and some are experimenting with VR versions of films including The Martian. Futuresource Consulting believes the VR content market could be worth $8.3bn within four years.

Beyond entertainment, advocates say these headsets could transform education, travel, real estate and architecture, not to mention videoconferencing and social networking. Some inside Uber are worried that Oculus could one day prove disruptive to their business by removing the need for people to travel. Why hail a taxi when you can teleport?

“Whenever a market is this early, you have to have strong convictions loosely held,” says Nabeel Hyatt, a venture partner at Spark Capital, which also backed Oculus. “We don’t know what’s going to happen.”
That uncertainty provides fertile ground for entrepreneurs. “There will be billion-dollar companies started by college students because someone gave them a Rift as a present and they solved a very specific problem,” says Anjney Midha, a partner at KPCB Edge.

However, as any sci-fi reader knows, new technologies have inherent risks, too. The futures depicted in Ready Player One and Snow Crash are dystopian and chaotic.

In December, academics led by Christian Sandor of the Nara Institute, Japan, wrote that “true augmented reality”, where the digital is indistinguishable from the physical, “will be the most powerful medium that humanity ever had at its disposal”.

link to this extract

 

What if Twitter Died? » Tech.pinions

Bob O’Donnell:

this seems to be one of the fundamental problems of Twitter. It’s appealing to Hollywood, TV, music and sports celebrities as a means to interact more intimately with their fans and share the kinds of details they’d never provide to traditional celebrity media. It’s appealing to the tech industry as a mouthpiece for those who want to determine the course of what is or isn’t important. The digital taste-setters, so to speak.

But for mainstream business and consumer users? Not so much. Arguably, this is the biggest problem with Twitter—it can’t seem to stretch beyond its celebrity, celebrity follower, and tech roots. If you aren’t into celebrities or the tech industry, Twitter just isn’t that appealing, especially given all the other options for online social interactions.

Despite these points, I think the navel gazing value of Twitter to the tech industry is so high, I seriously doubt they’ll let Twitter actually die. Someone with enough money and enough self-interest will likely make sure that, no matter what, Twitter will continue in some shape or form. Eventually, it’s value may start to fade, as some have already started to argue, but at least the Twittersphere will have a few years to adapt and find new alternatives.

The fundamental challenge is a publishing service that’s essentially based on self-promotion, self-aggrandizement, and self-importance at some point is going to run into the wall of indifference. Not everyone cares to read about what the self-elected are all doing all the time.

link to this extract

 

Yahoo loses mobile entrepreneur Arjun Sethi to venture firm » WSJ

Douglas MacMillan:

Mr. Sethi helped lead Yahoo’s effort to compete with Facebook Inc.FB -2.29% and Snapchat Inc. in the emerging area of mobile chat apps. Last July, his team released Livetext, a mobile app that lets users send live video and text without any sound.

Livetext failed to take off with users. In its first month, the program dropped out of the ranking of the 1,000 most popular apps in Apple Inc.’s app store and never returned, according to data from App Annie.

Yahoo’s struggles to produce a hit mobile app has hurt Ms. Mayer’s chances at turning around the 20-year-old Internet icon. This week, Yahoo said its board is weighing “strategic alternatives” to the turnaround which likely include a sale of its core Web business…

…Mr. Sethi is one of dozens of startup founders Ms. Mayer brought into Yahoo through a series of small acquisitions. In her three-and-a-half years as CEO, Yahoo has spent more than $2.3bn on at least 53 acquisitions, largely for small mobile-software developers whose apps were shuttered and whose founders were enticed to work on new projects at the company. At least 26, or over one-third, of the more than 70 startup founders and CEOs who joined Yahoo through an acquisition during Ms. Mayer’s tenure have left the company, according to their profiles on LinkedIn Corp.

As has also been pointed out, Yahoo last week wrote down the value of those acquisitions by $1.2bn. The idea of a video app without sound appears dumb, but then again lots are like that; but Instagram, Facebook and Vine were all there ages earlier. Yahoo’s problem is that it’s late and has no traction in mobile, not that the ideas are of themselves bad.
link to this extract

 

Sacked in Dublin by a boss in… London » Private Eye

Private Eye is always anonymous:

Google’s claim that all its real business is handled through its European HQ in Dublin while its multiple UK offices exist merely to count the paperclips, organise staff leaving collections and do the morning coffee run is further undermined by evidence it gave to an employment appeals tribunal in the Irish capital in 2013.
Rachel Berthold had been sacked in May 2011 from a position as a “level six” manager, which the tribunal heard put her in the top 7% of employees in Google’s Dublin office.

Anne-Catrin Sallaba, her former boss as Google Europe’s Head of Publisher Services, gave evidence to the tribunal that Berthold had failed to meet performance targets – but Sallaba had to cross the Irish sea to do so, given that as Berthold’s line manager she was employed in, er, London.

Berthold was eventually awarded €100,000 for unfair dismissal. Sallaba has in the meantime been promoted twice, and now rejoices in the job title “Senior People Development Manager, Head of Global Onboarding” – still in London!

As it happens, Matt Brittin of Google UK will be testifying before the UK Parliament this week.
link to this extract

 

Errata, corrigenda and ai no corrida:

Start up: Google’s search rejig, adblockers can’t Play, Sony to exit tablets?, Magic Leap’s big step, and more

Posted on by
2

No longer can you seek him here or there. Photo by abrinsky on Flickr.

You can now sign up to receive each day’s Start Up post by email. You’ll need to click a confirmation link, so no spam.

A selection of 10 links for you. Use them wisely. I’m charlesarthur on Twitter. Observations and links welcome.

Google search chief [Amit] Singhal to retire, replaced by AI exec » Bloomberg Business

Jack Clark:

“When I started [at Google in 2000], who would have imagined that in a short period of fifteen years, we would tap a button, ask Google anything and get the answer,” Singhal wrote in a Google+ post announcing his retirement. “My dream Star Trek computer is becoming a reality, and it is far better than what I ever imagined.”

With Giannandrea’s appointment, the technology may get smarter. The executive has overseen recent artificial intelligence efforts, including RankBrain, which saw Google plug an AI technology called a neural network into its search engine to boost the accuracy of results and an e-mail service called Smart Reply that automatically writes responses. Other work he has managed include efforts in image recognition and technologies that fetch information based on what users are doing with their devices, rather than what they’re explicitly searching for.

[John] Giannandrea joined Google in 2010 when it acquired a company he co-founded called Metaweb Technologies. Those assets became the basis for Google’s knowledge graph, a vast store of information on hundreds of millions of entities that helps the search engine present factual data in response to certain queries. Singhal’s last day is scheduled to be Feb. 26.
The elevation of Giannandrea represents a further emphasis on the importance of artificial intelligence to Google, a unit of Alphabet Inc. Chief executive officer Sundar Pichai said the technology has been key to recent efforts in search on mobile devices and personal assistant technologies.

Speaking of search..
link to this extract

 

Garth Gibbs: ‘The archetypal diary writer’ » Press Gazette

An obituary from August 2011:

Garth also managed to spend much of his time chasing various ‘sightings’ of ‘Lucky’ Lord Lucan, who was thought to have fled abroad after apparently mistaking his nanny for his wife and bludgeoning the ‘wrong’ woman to death. Of this colourful period in an almost continually helter-skelter career, Garth himself wrote: ‘As that brilliantly bigoted and crusty old columnist John Junor once cannily observed: ‘Laddie, you don’t ever want to shoot the fox. Once the fox is dead there is nothing left to chase.'”

With a wonderfully fertile imagination – a prerequisite of any good tabloid journalist – plus a good deal of chutzpah, Garth relished the challenge of keeping Lord Lucan alive, but never finding him.

‘I regard not finding Lord Lucan as my most spectacular success in journalism,’he said. ‘Of course, many of my colleagues have also been fairly successful in not finding Lord Lucan. But I have successfully not found him in more exotic spots than anybody else.

‘I spent three glorious weeks not finding him in Cape Town, magical days and nights not finding him in the Black Mountains of Wales, and wonderful and successful short breaks not finding him in Macau either, or in Hong Kong or even in Green Turtle Cay in the Bahamas where you can find anyone.”

Lucan was finally declared dead – though never found – on Wednesday. Not finding him was indeed a splendid task allotted to many journalists down the years. Speaking of search…
link to this extract

 

#SEO for sale?! Exposing Google loopholes in light of FTC native guidelines » aimClear® Blog

Marty Weintraub:

Mashable, a respected global media company focused on informing and entertaining “the digital generation,” was our inspiration.  Mashable has joined the swelling ranks of websites selling native content articles to advertisers.  Initially we were interested in participating in the program and reached out to Mashable regarding their native post advertising, which is called BrandSpeak or BrandLab.

As the conversation progressed, we were curious as to how Mashable native posts show up in Google search results and disclosure verbiage in light of new FTC native advertising guidelines. After we corresponded with a Mashable sales associate and researched BrandSpeak/BrandLab in detail, we were motivated to share our findings with the community as a point of learning about native content.

Those findings surprised (and astonished) us. Aimclear analysts identified a Google SEO loophole, which is perhaps the greatest ranking algorithm gap in years, allowing marketers to literally buy their way into Google search results with paid content…

…At best, allowing paid SEO tilts the playing field, making it even harder for smaller, perhaps more relevant players to compete for free Web Search results.

Google’s Webmaster Guidelines governing native content and Web Search are firmly rooted in 2013.

Tricky; this stuff is low-quality, but sites are desperate to generate revenue somehow. Speaking of revenue…
link to this extract

 

Google boots ad blockers from Google Play » TechCrunch

Sarah Perez:

According to Rockship Apps founder and CEO Brian Kennish, maker of Adblock Fast, Google’s app reviews team informed him the app was being removed for violating “Section 4.4” of the Android Developer Distribution Agreement.

This is the section that informs developers they can’t release apps that interfere with “the devices, servers, networks, or other properties or services of any third-party including, but not limited to, Android users, Google or any mobile network operator.”

If that text sounds a little broad-reaching and vague, that’s because it is. It’s also what allows Google to react to changes in the industry, like this one, on the fly.

Kennish says that Google’s app reviews team informed him that he could resubmit after modifying his app so it didn’t “interfere with another app, service or product in an unauthorized manner.”

“We’ve been trying to contact Google through their public channels since Monday, and I tried through private ones all day yesterday…but we haven’t gotten any official response from a human – just autoresponders,” notes Kennish.

He suspects that Adblock Fast was the first to be pulled from Google’s app store because it had climbed the charts so quickly and had achieved a 4.25 rating. Kennish says that the app had around 50,000 installs at the time of its removal.

In addition, the company could have gotten on Google’s radar by pushing out an update that offered a better user experience. (Some people didn’t realize it only worked on Samsung’s 4.0 browser and left 1-star reviews. The update was meant to better highlight the app’s requirements.)

Meanwhile, as of the time of writing, other ad blockers are still live, including Crystal and Adblock Plus (Samsung Browser). However, that may not be the case for long.

Crystal’s developer Dean Murphy also just submitted an update that’s just been declined by Google’s app review team for the same reason cited above. Again, Google references section 4.4 of the Developer Agreement as the reason for stopping the update from going live.

“I have appealed the update rejection, as I assume that I am rejected for ‘interfering’ with Samsung Internet Browser, citing the developer documentation that Samsung have for the content blocking feature,” explains Murphy. “I’m still awaiting their reply.”

Wow, that was fast. Crystal was still there on Wednesday. This is going to ratchet up tensions between Google and Samsung (again); in the comments on the Verge article on this topic (which has less detail) there are people who switched to iOS because of adblocking, or are considering moving back because they can’t get it on Android. A small but possibly significant group.

Google has clearly set its face against adblocking on mobile, but the pressure is starting to build up behind the dam.
link to this extract

 

About » DeepDetect

DeepDetect (http://www.deepdetect.com/) is a machine learning API and server written in C++11. It makes state of the art machine learning (such as deep learning) easy to work with and integrate into existing applications. Its goal is to simplify and secure both the development and production phases by using possibly different servers and passing models from one to the other.

It originates from the need for industries, businesses and researchers to quickly fit a machine learning pipeline into existing applications, starting with well-known models, and moving toward more targeted ones while measuring accuracy.

DeepDetect allows this by coupling a generic API and a server with high performance machine learning libraries. At the moment it has support for the deep learning library Caffe. More libraries are to be supported in order to span over a larger set of common use cases.

There are free (even for commercial use) models that are downloadable from the site. This lies just over my event horizon for understanding – but reading the details about “getting started” puts me in mind of people feeding a giant brain, or disembodied intelligence, and that gives me pause.

But this stuff is going to be everywhere in two years.
link to this extract

 

Would you be sad to see Sony withdraw from the tablet market? » Xperia Blog

The mysterious “XB”:

given the challenging smartphone market, as evidenced by last week’s results there is no guarantee that Sony will continue to cater for the tablet market. A recent Japanese blog post by a Sony store manager speculated that the company may withdraw from the tablet market after receiving marketing material suggesting so.

The news would not surprise us, after all, we know that tablets made just 5% of Sony Mobile’s revenues back in 2013 and that was expected to shrink even further. Given the R&D costs of developing and supporting new devices, Sony may feel that producing another tablet for 2016 might not be commercially viable.

I didn’t know that about the tablet revenues; apparently they’re meant to be down to 3-4% now. The question is whether they generate more than 0% in profit – because they must be eating up R+D time and money, which is opportunity cost that Sony probably can’t afford.
link to this extract

 

The joy of shortcuts » Allen Pike

Next January, Parse is shutting down. The successful Parse apps will get moved to a custom backend like ours was, perhaps using Parse’s excellent open-source server and migration tool. The unsuccessful Parse apps will die. Hundreds of thousands of unsuccessful Parse apps will perish. Like links to long-dead Geocities pages, dead mobile apps that relied on Parse will linger in the App Stores for years, slowly accumulating one-star reviews.

As much as Parse will try to get the word out that they’re shutting down, many apps’ owners don’t even know that they’re reliant on Parse. Parse’s overly generous free plan made them popular with freelancers and consultants building quick app backends for their clients. Many of those clients don’t know what Parse is, let alone that the little app they commissioned a couple years ago is a ticking time bomb.

How many iOS apps, how many Android apps relied on Parse? There needs to be an enumeration.
link to this extract

 

How the iPhone 6 ruined Apple » All this

Dr Drang:

While it’s certainly possible that the great days of iPhone sales growth are over, I wouldn’t make that prediction just yet. In fact, I was surprised to learn that iPhone sales were merely flat. I was expecting a decline—not because the iPhone is losing popularity, but because the iPhone 6’s first quarter of sales was such a gigantic leap upward. The pent-up demand for a larger iPhone caused sales to increase nearly 50% year over year, to 74.47m from 51.03m the year before. This was the biggest percentage jump in year-over-year sales since the introduction of the 4S (which was goosed a bit because the 4S was delayed). I just didn’t think the 6S could keep up with that. And maybe it won’t.

But look at how things were going before the iPhone 6. Had the trend of 2012–2014 continued through 2015, iPhone sales last quarter would have been 65–70 million. Instead they were just under 75m. It’s only in comparison to the huge holiday quarter of 2014 that last quarter looks dull.

I’m reminded of the devotion climate change deniers had to the year 1998. Because of an intense El Niño that year, global temperatures rose well above the trend line, and it remained the hottest year on record for several years. Deniers hit upon this fact, and claimed that global warming had stopped, even though the overall warming trend had continued. The iPhone 6 was Apple’s El Niño.

link to this extract

 

Magic Leap Just Landed an Astounding Amount of VC Money » WIRED

Jessi Hempel on the company which has just raised $793.5m in a C round from Alibaba and others (Google and Qualcomm were already on board):

Many believe Magic Leap’s technology—along with a handful of competing virtual and augmented reality products—will usher in a sea change in how we use computers. By placing sensors everywhere and processing the volumes of data they produce, it’s possible to create better immersive environments and believable layers of digital images on top of the physical world. Facebook, Samsung, and Microsoft are creating competing technology and have chosen to make their headsets available even as they’re engineering the products. Google is also beefing up its virtual reality team, and Apple is also reportedly getting into the action. Magic Leap claims to be using a different technology to achieve its effect, and it’s keeping its efforts mostly secret.

The company has made converts out of many of those who have seen demos. New Zealand design studio Weta Workshops has teamed up with Magic Leap to build games. Science fiction writer Neal Stephenson joined the company as its chief futurist. Last fall, Google led a $542m investment, bringing its previous funding total to $592m. But so far, it hasn’t been clear when Magic Leap’s tech will be available for consumers.

Expectations around this are going to be huge, which usually leads to disappointments of the same size. Shipping product matters; having that much of a cash pillow can’t be good, because it won’t help the financial discipline needed to make things (of whatever sort) to a price, for a user, to a specification. Don’t forget the lesson of Leap Motion – big hype, big letdown.
link to this extract

 

​Startup lands $100m to challenge smartphone superpowers Apple and Google » CNET

Stephen Shankland:

Acadine, which CNET previously reported was initially known by the placeholder name Gone Fishing, plans to build an operating system for smartphones, tablets, wearable devicess and the Internet of Things.

That’ll be a tough challenge. But [fouder Li] Gong believes Acadine’s generous funding [from a Chinese state-controlled company], fast development and international reach will mean consumers finally will see the alternative to Apple and Google that so many other companies have failed to build.

And of course that means his startup and its investors will reap the rewards. “Owning an OS is extremely important if you can do it,” Gong said. “It’s very profitable if you can do it.”

Potential sources of money, Gong said, include being paid to promote services like search, storage, music streaming and e-commerce; revenue sharing from those services when customers pay to use them; and fees generated by advertising and game sales. All of those, though, depend on Acadine succeeding in finding and exploiting gaps where existing OSes are weak then expanding from there to a large user base.

The list of mobile operating systems that have struggled to compete against Android and iOS and gain that large population of users is long: Microsoft’s Windows Phone, Samsung’s Tizen, Jolla’s Sailfish OS, Canonical’s Ubuntu, Hewlett-Packard’s WebOS, BlackBerry’s BlackBerry OS and Mozilla’s Firefox OS. This last project is the one Gong led at Mozilla until he left in April, and it’s the starting point for H5OS.

One hates to say “a fool and his $100m are soon parted”, but it’ll do.
link to this extract

 

Errata, corrigenda and ai no corrida: I was going to include a link to a video of a male cyclist who was suspected of having a motor in his bicycle (and man, it looked fishy) but realised it is a rabbit hole one would never emerge from.

Start up: Apple’s conundrum, watch birds migrate!, the 5am startup, and more

Posted on by
Reply

A Go endgame: where would you play next? DeepMind’s systems would know. Photo by chadmiller on Flickr.

Only humans can sign up to receive each day’s Start Up post by email. They need to click a confirmation link, so no spam.

A selection of 9 links for you. Use them wisely. I’m charlesarthur on Twitter. Observations and links welcome.

Apple wants to be a services company » Business Insider

Jay Yarow with a neat summary of Apple’s conundrum:

So, Apple is in this weird cycle: It wants to grow services revenue, but services revenue depends on iPhone sales. Currencies are falling because the global economy is weak relative to the US economy, which is leading Apple to raise prices on the iPhone, which is hurting iPhone sales, which will limit services revenues.

Apple could lower prices of the iPhone to sell more units and then grow services, but it doesn’t seem to want to do that.

Cook said Apple already had a variety of price points, from the low-tier iPhone 5s to the high-end iPhone 6s Plus.

“I don’t see us deviating from that approach,” Cook said.

This makes sense since Apple is a hardware company. If it were a services company, it would lower prices, go for smartphone unit volume, then get more money from that. But it is not a services company.

It is not principally a services company. The money it does make from services is actually pretty substantial, and threatens to overtake iPad revenue. Poor iPad.
link to this extract

 

iPhone Headwinds » Tech-Thoughts

Sameer Singh:

Apple also increased iPhone channel inventory by 3.3m over the quarter compared to a decline of 0.2m in the same quarter last year. Keeping in mind that Apple reports sell-in (shipments), not sales to end users, this implies that iPhone sell-through actually declined by 4.3% YoY, from 74.7m to 71.5m units. This is the first such decline in the history of the product.

In addition to this sell-through decline, Apple’s revenue guidance for the next quarter implies a steeper YoY decline in unit sales. And finally, this was accompanied with a YoY increase in average selling price (ASP) which suggests that the iPhone mix is shifting towards higher end models…

…As Clay Christensen is fond of saying, “Disruption is a process, not an event”. The mechanics of low-end disruption have been working ever since the first Android smartphone was unveiled in 2008. The key test for the iPhone will be the iPhone 7 cycle, starting in fiscal Q1 [Oct-Dec] 2017. In a “redesign year”, it will be easier to gauge whether meaningful product improvements can continue to generate increased demand or if today’s smartphones have already reached “good enough” territory.

The sell-through point is valid (only BlackBerry also gives the same data). The odd thing is that rising ASP: it points to people having the disposable income to buy the pricier models, which argues against the “low-end disruption” thesis.

But the gap between the average selling price of an Android handset and the average selling price of an iPhone has never been larger. That must have knock-on effects. But what?
link to this extract

 

Google AI algorithm masters ancient game of Go » Nature

Elizabeth Gibney:

DeepMind’s program AlphaGo beat Fan Hui, the European Go champion, five times out of five in tournament conditions, the firm reveals in research published in Nature on 27 January. It also defeated its silicon-based rivals, winning 99.8% of games against the current best programs. The program has yet to play the Go equivalent of a world champion, but a match against South Korean professional Lee Sedol, considered by many to be the world’s strongest player, is scheduled for March. “We’re pretty confident,” says DeepMind co-founder Demis Hassabis.

“This is a really big result, it’s huge,” says Rémi Coulom, a programmer in Lille, France, who designed a commercial Go program called Crazy Stone. He had thought computer mastery of the game was a decade away.

The IBM chess computer Deep Blue, which famously beat grandmaster Garry Kasparov in 1997, was explicitly programmed to win at the game. But AlphaGo was not preprogrammed to play Go: rather, it learned using a general-purpose algorithm that allowed it to interpret the game’s patterns, in a similar way to how a DeepMind program learned to play 49 different arcade games.

This means that similar techniques could be applied to other AI domains that require recognition of complex patterns, long-term planning and decision-making, says Hassabis. “A lot of the things we’re trying to do in the world come under that rubric.” Examples are using medical images to make diagnoses or treatment plans, and improving climate-change models.

This is a gigantic result; being good at Go requires a subtle intuition and feel for space. (At least, for a human.) It’s far, far more complex than chess. And that this was done by a non-specific program has colossal implications.

As Nature’s leader on the topic comments, as these systems spread into our lives, “The machine becomes an oracle; its pronouncements have to be believed… Intuitive machines will need more than trust: they will demand faith.”

As a side note, DeepMind looks like one of Google’s smartest purchases – perhaps after YouTube.
link to this extract

 

Nest thermostat goes from ‘Internet Of Things’ darling to cautionary tale » Techdirt

Karl Bode:

[Tech writer Stacey Higginbotham’s] Nest device began trying to cook her family in the middle of the night, something Nest first tried to blame on her smart garage door opener, then tried to blame on her Jawbone fitness tracker (Nest never did seem to pinpoint the cause). Her report suggests that an overall culture of “arrogance” at Nest shockingly isn’t helping pinpoint and resolve bugs:

“One Nest partner, who declined to be named to preserve his business relationship with the company, said that Nest being quick with the blame didn’t surprise him, citing a culture of arrogance at the company. When something went wrong during integration testing between his device and Nest’s, problems were first blamed on his servers and team.”

And fast-forward to last week, when researchers putting various internet of thing devices through tests found that the Nest thermostat was one of many IOT devices happily leaking subscriber location data in cleartext (with Nest, it’s only the zip code, something the company quickly fixed in a patch). Granted Nest’s not alone in being an inadvertent advertisement for a product’s “dumb” alternatives. In 2016, smart tea kettles, refrigerators, televisions and automobiles are all busy leaking your private information and exposing you to malicious intrusion (or worse).

It’s a fascinating, in-progress lesson about how our lust for the sexy ideal of the connected home appears to be taking a brief pit stop in reality.

link to this extract

 

Vigilant solutions and the spread of police surveillance » The Atlantic

Conor Friedersdorf:

Throughout the United States—outside private houses, apartment complexes, shopping centers, and businesses with large employee parking lots—a private corporation, Vigilant Solutions, is taking photos of cars and trucks with its vast network of unobtrusive cameras. It retains location data on each of those pictures, and sells it.

It’s happening right now in nearly every major American city.

The company has taken roughly 2.2bn license-plate photos to date. Each month, it captures and permanently stores about 80m additional geotagged images. They may well have photographed your license plate. As a result, your whereabouts at given moments in the past are permanently stored. Vigilant Solutions profits by selling access to this data (and tries to safeguard it against hackers). Your diminished privacy is their product. And the police are their customers.

Sounds a bit similar to the UK police’s Automatic Number Plate Reader network, which extends around the UK, except this is historic too. (Then again, the UK’s ANPR system probably is too.)
link to this extract

 

How to wake up at 5am and build your startup » Medium

Patrick Park (once he’s got past the stuff about how to, you know, get up at 5am, and gets on to the stuff about building your startup):

It’s hard to admit, but no one really knows what they’re doing at first. Even the largest startup unicorns in the world, took a long time to find their footing. AirBnb survived by selling political themed cereal Obama O’s and Cap’n McCain’s. #Slack came out of a failed game “Glitch.” Even YouTube began as a video dating service “Tune in Hook Up.”

Micro-startups work well with the Lean Startup methodology. Not only are micro-startups easier to implement, but provide a shorter feedback loop that keeps you more in touch with your audience. With 75% of all startups failing, would you rather spend 100 days in a basement building a product your customers “might” like or connect with your audience everyday “while” you adapt your startup to address your customer’s pain-points? Don’t build another Blockbuster.

Start with the assumption that you’re wrong. Constantly validate your theories through micro-experiments. When I first started “Krown.io”. I explained the service as an “Annotation Blogging Platform.” That was, until I found out the majority of people have no idea what “Annotations” are. We tried a variation of “Smart Blogging,” “Highlight Blogging,” “Feedback Blogs,” and “Contextual Blogging Platform.” Which surprise, surprise. People still had no idea what we were talking about. So we added a bare-to-the-bones explanation, “Highlight a text and add comments directly on the highlighted text.” Validate your hypothesis.

The startup advice is fine. Waking up at 5am isn’t that hard, but he makes it sound like one of the 12 tasks of Hercules.
link to this extract

 

Lenovo used 12345678 as hard-coded password in SHAREit for Windows » Graham Cluley

Cluley is amazed – as you will be:

The first vulnerability [of four, all of which could be exploited remotely] (CVE-2016-1491) is perhaps the most infuriating. CoreLabs discovered that whenever SHAREit for Windows is configured to receive files, this process creates a Wi-Fi hotspot that is ‘protected’ by the password “12345678”.

Not surprisingly, this password just recently earned a top spot on the latest list of worst passwords you could possibly choose.

What is surprising is the fact that Lenovo would incorporate such an insecure password into its application — and one that does not change, no less!

Then again, I suppose the issue could be worse. In the second vulnerability (CVE-2016-1492), which applied only to SHAREit for Android, there is no password set up to protect the Wi-Fi hotspot when the app is configured to receive files.

To be sure, it doesn’t say much when Lenovo could have mitigated two separate vulnerabilities by adhering to the most basic principles of password security.

But moving right along. The third vulnerability (CVE-2016-1490) discovered by CoreLabs builds upon the insecure Windows password issue discussed above

When elephants do software while dancing. Yes, it was Lenovo which preinstalled Superfish.
link to this extract

 

Mesmerizing migration: watch 118 bird species migrate across a map of the western hemisphere » All About Birds

Pat Leonard:

For the first time, scientists at the Cornell Lab of Ornithology have documented migratory movements of bird populations spanning the entire year for 118 species throughout the Western Hemisphere. The study finds broad similarity in the routes used by specific groups of species—vividly demonstrated by animated maps showing patterns of movement across the annual cycle.

There’s also a version showing which species is which.
link to this extract

 

Apple India enterprise head Sharad Mehrotra quits, starts up with Hyve Mobility » timesofindia-economictimes

Muntazir Abbas:

Sharad Mehrotra, iPhone maker Apple’s enterprise mobility head in India, has quit and set up a handset company to compete in the world’s fastest-growing smartphone market. Mehrotra, who was with Apple India since October 2007, has cofounded Hyve Mobility , which will launch smartphones using Android, the operating system that rivals Apple’s iOS, by March.

“While consumers today have a lot of options to choose a smartphone, the element of customer delight is seriously missing,” he told ET, adding that it was high time to get into the smartphone market…

…”In the current Android scenario, no brand enjoys customer loyalty and we want to bring the change with our path-breaking products portfolio and service offerings,” said [Aditya] Agarwal, MD of Hyve Mobility.

link to this extract

 

Errata, corrigenda and ai no corrida: none noted.

Start up: AI for your app, quantum computing works?, Yahoo’s future, Watch watch, and more

Posted on by
1


Firefox OS: heading rapidly for the exit. Photo by Wojciech Szczęsny on Flickr.

You can now sign up to receive each day’s Start Up post by email. You’ll need to click a confirmation link, so no spam.

A selection of 9 links for you. Use them wisely. I’m charlesarthur on Twitter. Observations and links welcome.

How predictive APIs simplify machine learning » ProgrammableWeb

Louis Dorard:

App developers are always looking for ways to make the lives of their users easier and for ways to introduce innovative features that help users save time. For this reason, Machine Learning (ML) has been increasingly popular in app development. Classical examples include spam filtering, priority filtering, smart tagging, and product recommendations. Some people estimate that Machine Learning is now being used in more than half of a typical smartphone’s apps. Because of the new functionality gained by these apps, we can talk of “predictive apps,” a term coined by Forrester Research which refers to “apps that provide the right functionality and content at the right time, for the right person, by continuously learning about them and predicting what they’ll need.” 

If you’re writing an app that would fit that description, this is a great primer.
link to this extract

Mozilla will stop developing and selling Firefox OS smartphones » TechCrunch

Ingrid Lunden:

Firefox OS was first unveiled in 2013, with the aim of targeting the developing world and late adopters with low-cost handsets.

To differentiate from Android and iOS, Mozilla and its carrier partners focused on a web-first platform, with no native and only web apps. Sales, however, were always poor and the devices themselves failed to ignite a lot of consumer interest, and a number of OEMs cornered the market with a flood of cheap handsets. In a business that depends on economies of scale, it was a failure.

Mozilla has been on a streamlining track lately. Last week it announced that it would be looking for alternative homes for its Thunderbird email and chat client. The aim is for the company to focus more on its strongest and core products and reputation.

Came really late to the game, and never made table stakes – an app ecosystem – because it didn’t think that that table was right. Apps trump the mobile web.
link to this extract

Drones save over two hundred people in Chennai floods » DRONELIFE

A senior officer of the Chennai police said that the force has deployed drones in several of the most unreachable neighborhoods, and have been able to locate as many as 200 people, rescuing all of them.  The search and rescue operation sends drones up from a control vehicle.  The aerial images obtained are then sent to a control room, where staff reviews footage and pinpoints affected homes and people.  When a rescue site is identified, the control room communicates with teams of volunteers nearest to the location through wireless walkie-talkie, sending rescue workers to retrieve victims stranded in their homes.

link to this extract

Controversial quantum machine bought by NASA and Google shows promise » MIT Technology Review

Tom Simonite:

Hartmut Neven, leader of Google’s Quantum AI Lab in Los Angeles, said today that his researchers have delivered some firm proof of that. They set up a series of races between the D-Wave computer installed at NASA against a conventional computer with a single processor. “For a specific, carefully crafted proof-of-concept problem we achieve a 100-million-fold speed-up,” said Neven.

Google posted a research paper describing its results online last night, but it has not been formally peer-reviewed. Neven said that journal publications would be forthcoming.

Google’s results are striking—but even if verified, they would only represent partial vindication for D-Wave. The computer that lost in the contest with the quantum machine was running code that had it solve the problem at hand using an algorithm similar to the one baked into the D-Wave chip. An alternative algorithm is known that could have let the conventional computer be more competitive, or even win, by exploiting what Neven called a “bug” in D-Wave’s design. Neven said the test his group staged is still important because that shortcut won’t be available to regular computers when they compete with future quantum annealers capable of working on larger amounts of data.

Been a long time coming, but this is just starting to look promising. Hell, even if it’s off by a few orders of magnitude, it’s amazing.
link to this extract

What’s going on at Yahoo? Here are seven things worth knowing » BuzzFeed News

Mathew Zeitlin draws up the list, in which No.1 and No.5 are the important ones:

Here’s the deal. Yahoo’s current market value is about $32.9bn.

This is much less than the value of the things it owns. Yahoo’s stake in Alibaba is worth about $32.4bn, and its stake in Yahoo Japan is worth about $8.7bn. It also has $1.3bn in cash and about $5.5bn in other securities, and $1.2bn in debt. All that adds up to around $46bn.

So if the market values Yahoo at $33 billion, does that imply the actual Yahoo business — the websites, the apps, the digital advertising tech — is worth less than zero?

Not quite — and here is where those tax issues come into play. Yahoo’s investments in Japan and China have all gained value massively over the years, and all that is subject to taxes if it’s sold. Hedge fund Starboard Value estimates the tax bill on Alibaba shares put their true value to shareholders at around $19.6bn; the Yahoo Japan stake would be worth around $5.3bn.

Once you take those taxes into account, it looks more like Yahoo investors are valuing its actual business at a little over $2bn. That’s a figure that has been promoted by activist investor Starboard Value, as well as analysts at Nomura and Pivotal Research.

And now No.5:

There may be cooler kids on the block these days, but Yahoo still has a massive presence on the web.

According to ComScore, Yahoo has a global audience of 618 million — the fourth largest of any company, behind only Google, Microsoft, and Facebook. In the U.S., Yahoo’s 211 million desktop and mobile unique visitors make it the third biggest destination, behind Google and Facebook.

“Our overall network including Tumblr continued to serve a global user base of more than 1 billion monthly active users,” Yahoo CEO Marissa Mayer said in a recent earnings call. Facebook, in comparison, has over 1 billion daily active users. In terms of headcount the two are comparable: Yahoo has 10,700 full-time employees, while Facebook has about 12,000.

link to this extract

Android returns to growth in Europe’s big five Markets » Kantar Worldpanel

Carolina Milanesi:

“As the holiday season approaches, it appears smartphone upgrades are on Santa’s list, with 14% of EU5 smartphone owners planning to replace their current device with a new one in the next three months,” Milanesi said. “Among those consumers, 25% said they prefer Apple, while 38% said they prefer Samsung. Among Apple owners in the EU5 planning to upgrade over the next three months, 79% said they prefer Apple, while 62% of Samsung owners planning to upgrade say they prefer Samsung.”

High retention rate for Apple; less so for Samsung. But Samsung has more users overall, because it sells more phones. (Leaky buckets.)

What’s not visible is the general trend; iPhone sales, on this data, are trending faintly upwards in the mature markets such as the EU5 and US and China.
link to this extract

Time ticks on chances of the Apple Watch catching on » FT.com

Tim Bradshaw:

The pollsters quizzed 1,017 Britons over the age of 15. They found 66% were aware of smartwatches. Awareness was down to 60% among respondents aged 35 and older, and to 57% among the lowest three social and economic groups.

Only 2% said they owned a smartwatch, down to 1% among those over 35. The poll showed 43% believed people did not need a smartwatch; but that doesn’t mean 57% of people believe you do need one.

Similarly, 24% saw a smartwatch as a gimmick, but that’s not an indication that 76 per cent regard it as a life necessity.

Possibly the glummest news for enthusiasts was that only 6% of the smartwatch-aware were likely to buy one in the next year.

So, unless I’m reading the figures wrongly, enthusiasm for this kind of wearable technology is several degrees below lukewarm.

Wearable technology, in general, hasn’t proven its worth to the general population. Then again, smartphones didn’t prove their worth to the general population for quite some time either – about three years from the launch of the iPhone. I’d love to see a comparative study from that time. (Links welcome.)
link to this extract

Apple’s secrets about the iPhone were revealed during Samsung lawsuit » BGR

Yoni Heisler looks back to what came out in the 2012 trial during the discovery phase, particularly in the documents revealed to either side. How about the kickstand idea for the original iPad?

Yeah, perhaps you can guess how long Steve Jobs would let that one live.
link to this extract

June 2015: Which phone has the best battery life? 5 top smartphones tested and compared » Trusted Reviews

Andrew Williams, in June 2015:

For every phone we review, we perform battery tests. There are benchmarks, and just using the phone to see how long it really lasts in daily use. This combo gives you a good idea of how long any phone will stay awake between charges.

But it’s fallible.

All sorts of things can affect battery life, especially when you’re out and about using the thing. So we decided to get all the big phones of 2015 together and give them a thorough going-over with some real-life-related tests to see which phone really is the longest-lasting.

Which phones? We’ll be checking out the iPhone 6, iPhone 6 Plus, Samsung Galaxy S6, LG G4 and HTC One M9. After all, they’re the most desirable phones of the year.

Remarkable results (on video loops, web browsing, film over Wi-Fi, music in the background). Enjoyable comments too saying “but the battery is reporting it wrong!” Which might, actually, be correct. But probably isn’t. (Via Ian Betteridge.)
link to this extract

Errata, corrigenda and ai no corrida:

TEDx Hilversum: “How to spot the next big thing” – slides and commentary from my talk

Posted on by
8
I spoke at the first TEDx in Hilversum, Holland

There was a “selfie booth”, appropriately enough.

I was invited to talk at the inaugural TEDx Hilversum – the Dutch city which is the country’s medialand, and whence the TV format ideas both for “Big Brother” and “The Voice” came.

The topic: “How to spot the next big thing”, building on a column I wrote for The Guardian’s Tech Monthly supplement back in October, about how the selfie was pretty much accidental.

What I wanted to explore and expand on in the talk was how these “next big things” in social interaction happen, and where you’d look to find the next one. (This isn’t a transcript – it’s the ideas I spoke on. The talk is about 15 minutes. I’ll put up the link when it’s available.)

Spotting the next big thing

How to spot the next big thing
Photo by c@rljones

This isn’t, therefore, about which startup you should put your money into, though it might give you hints about what sort of things could generate money – if you’re ahead of the game.

Three characteristics

Three characteristics of a 'next big thing'
Photo by Japanexperterna.se

Three characteristics of “next big things”: they’re about kids and teens experimenting; adults find them a bit silly (or impossible or embarrassing); and they don’t require anything extra, because they’re immanent to the device.

Only mobile matters

Mobile is the only platform that matters
Photo by Kris Krug

When we’re looking for the “next big thing”, the only place to bother looking is mobile. It’s the only platform that matters. People might say “what about the PC? There are 1.5 billion of them installed around the world.” Nope.

Think about this: what was the last important app that launched first on the desktop (not in the browser, because browsers work on mobile too)? There were two – Spotify and Dropbox, which both launched around the autumn of 2009. Everything big since then – Instagram, Snapchat, Whatsapp, Uber – has launched first and pretty much only on mobile, with essentially no functionality on the desktop.

By contrast, everyone has a mobile phone – there are more than 2 billion smartphones in use today. Pretty soon everyone will have a smartphone. Look at the people in the picture above: they’re holding up their mobile phones, not their laptops.

The first big thing

The first big thing on mobile: texting
Photo by larskflem

The first big thing was text messaging, aka SMS. Invented in 1986 and implemented in 1992, it didn’t take off at first – until the advent of pay-as-you-go (PAYG or pre-pay) phones, which meant that they were cheaper and adults didn’t have to commit to a contract for their kids; they’d just use what they needed. And those kids discovered SMS was cheap, and fast, and personal, and they loved it. The explosion in PAYG phones, in the UK at least, happened in 1999.

Watch SMS grow

How SMS use grew globally
Here’s how SMS use grew globally. There’s almost exponential growth right up to 2011; and then it peaks in 2011-2012 and has fallen off since.

Why the falloff? Because the people who had been kids in 1999 were 12-13 years older. They’re grown up, having their own kids. Meanwhile the other kids growing up in the intervening years were moving on to their own things – OTT services using data, such as BBM, iMessage and Whatsapp. SMS was a golden goose for the carriers; now it’s being killed off. Whatsapp has only been available for five years – founded in 2009 and first released in 2010 – yet it’s now bigger than SMS in volume.

Me, my selfie, I

The source of 'selfie'
Sure, we know that there have been “selfies” for ages – painters in the Renaissance doing self-portraits, even Buzz Aldrin doing one in space during an Apollo mission in 1966. But we didn’t call it that, and “selfie” has a particular meaning in our context: pictures taken with our mobile phones, generally using the front-facing camera.

Watch ‘selfie’ grow

Photos tagged 'selfie' on Flickr, 2000-2005
I thought this 2002 origin was interesting, so I dug out data from Flickr, looking for photos tagged “selfie” by year (1 January-31 December for the respective years). There’s that first 2002 use highlighted for reference, and clearly lots of millennium-dated photos that were backloaded; Flickr didn’t even exist until 2004. Yahoo bought it in 2005, and things started growing.

..and grow

Photos tagged 'selfie' on Flickr, 2000-2012
Fast forward to 2012, and the number is exploding.

..and peak?

Photos tagged 'selfie' on Flickr, 2000-2015
Fast forward again, and this really looks like exponential growth. Though the 2015 figure – with my estimate for the final total – looks like growth has slowed substantially. Why’s that? We’ll come to that in a moment.

Trending searches

Google search trends for 'selfie'
Just for contrast, here are the Google search trends for “selfie”. Pretty much nothing until 2013, when it takes off. (Think of the selfie at the Mandela funeral by Obama and the Danish prime minister in December 2013, which is the first peak there, and the Oscars selfie by Ellen DeGeneres – actually taken by Bradley Cooper in March 2014 – which marks the high point.) But it looks as though interest in the selfie is dying, doesn’t it?

Peak selfie?

Google trends and Flickr trends for 'selfie'
If you superimpose the Google Trends data and the Flickr data, their growth looks pretty similar. So is the selfie dying off?

What’s more probably happening here is that a new generation of kids isn’t using Flickr – they’re on Instagram, where millions of selfies are posted every day. I couldn’t extract the data from Instagram, but you can be sure it’s huge. The selfie has probably got a few years left in it yet. But that generational shift is interesting, because it’s just what we saw with SMS before.

Next to arrive

Two big things; what might be a third?
So there are two examples so far of “next big things”. What might be next?

Well, the smartphone is the most personal device ever. It knows who we know, when they call, when we ignore them, what we like and what we watch, what we read, how we communicate, where we go.

Why wouldn’t it be used for… sex?

Call me maybe

Tinder's three-screen explanation
Turns out, it already is. Apps like Tinder fit all of our three criteria: it’s used by the young, it puzzles the old (“why would you judge someone based on so little information?”), and it uses qualities inherent to the device – selfies for profiles, real-time data updates, touch interaction, geolocation.

Tinder alone sees billions of “swipes” on profiles every day, and millions of meetups, and there are surely going to be Tinder babies – people created on the basis of algorithms on smartphones. Is that weird, or just natural? Your view might depend on your age.

A new search

It's in our phone, but what is it?
After SMS, selfies, sex, what will the next big thing look like? There are some clues. The biggest one is that it’s almost surely already in the phone, just as SMS was a capability in 1992, before PAYG phones, and selfies were possible from the first phones with a front-facing camera in 2003.

So here are three elements that I think might feed into the next big thing.

Three potential elements

The capability is there, and growing
Deep dream: photo by kevin dooley
Google Cardboard: photo by juan tan kwon

Artificial intelligence: The first, and biggest, is artificial intelligence or “machine learning”. This is a picture of someone as visualised through Google’s “Deep Dream” neural network. We don’t understand it because the machine refracts its view. But to the machine, it makes a sort of sense. There is a growing amount of AI/ML/NN technology in all our smartphones: Apple’s Siri, Google Now, Microsoft’s Cortana. You can also get “assistants” such as Amy (which will set up meetings) and “Charlie” (which will give you a social media profile of people you’re going to meet).

This AI tells us if we should leave early for meetings, who’s calling us (perhaps based only on phone numbers found in emails, not our contacts), what apps we look at, what news we look at, how we interact with the world through our phone. In the next few years, we should expect that it will become far more powerful, even without a connection to the cloud; if you think that sounds fanciful, just go back five years, to 2010, and none of the phones we have had those sorts of capabilities. Siri hadn’t been introduced. Think five years ahead, and that’s the sort of gap between now and then that we’re going to see.

Health/fitness: if my AI knows what I’ve been doing, shouldn’t it also know how well I’m doing? Connected to devices such as a watch or fitness band, there’s far more data about ourselves becoming available. Does that feed into the Next Big Thing by showing that you’re *really* fit in your online dating profile? Does your AI tell you before you’re going to be ill?

Virtual reality: This is Google Cardboard, a super-low-cost implementation of virtual reality: you cut and paste it together from a kit, and then slot your smartphone into the gap, and bingo. It’s on the tipping point, I think; ready to take off. And when that happens, everything becomes possible. What if the Tinder profile of the future lets you walk around the person you’re interested in dating, in 3D? What if you meet without physically being in the same place?

Conclusion

I don’t know what the Next Big Thing actually is. But consider a couple of points. SMS was invented over a decade before it actually took off. The word “selfie” was coined for the activity back in 2002 – yet it only exploded into public consciousness a decade later.

Perhaps the word for the next big thing has already been coined; it’s been tossed around carelessly on an online forum where someone is describing something they did or something that happened. And in years to come we’ll look back and say ah, it was obvious.

That’s because spotting the next big thing is a puzzle, not a mystery. There’s a difference between the two. A mystery is – well, think of a murder mystery. Only one person knows who did it, and they’re not saying. Mysteries are meant to remain unsolved.

But to understand a puzzle, think of a jigsaw puzzle. The pieces are all there, in plain view; the only problem is putting them together correctly.

For entrepreneurs, there is good news: if you’re alert, you can cash in. SMS made huge profits for carriers. Selfie sticks meanwhile have been nice business for some factories in Shenzhen. (They’ve even led to museums changing rules; that’s success, when you change society, even a little.)

Just as puzzles just need the pieces put together correctly, it’s very likely that everything necessary for the next big thing is right there, just waiting for someone to put it together. The people who do that probably won’t be the adults; it’ll be the kids and teens messing around. And adults will probably think it’s stupid. But that’s how it goes.

Still, we won’t have to look far to find it. It’ll be right there in the palm of our hands – in our phones.