American History X
At once as far as angel’s ken he views
The dismal situation waste and wild,
A dungeon horrible, on all sides round
As one great furnace flamed, yet from those flames
No light, but rather darkness visible
Paradise Lost – John Milton (and how a fair few in the US feel right now)
Readers of this blog will know that the spectre of a Trump presidency has loomed large in many 2016 roundup posts. Perhaps overly so. For good reason because that possibility became reality last night in a turn of events that unfolded with the same eerie symmetry as Brexit v1.0 five months ago with exit pollsters being progressively confounded on the night. The chronicle of a triumph/disaster that felt inevitable. It is impossible to assess what impact a President Trump will have both within the US and in relation to US standing in the rest of the world though it is clear that many inside and outside the US are filled with foreboding and concern based on the reactionary, prejudiced and isolationist stance he assumed during the campaign and the breadth of mandate his party has. However, the election revealed a bitterly divided nation facing huge challenges several of which were laid out here last week. Trump doesn’t appear to have a detailed strategy on any of them and many others besides. He will need to build a team fast that does. And presumably shelve several of the entirely undeliverable and preposterous promises he made in his bruising desperation to secure electoral victory. From here in at least he won’t be able to claim the system is “rigged” or blame shift his responsibility for addressing America’s manifold issues. Even so, he will doubtless claim for a while claim that he is shackled by the ‘legacy’ bequeathed him by Obama. Much of that will probably be undone within months of his ascendancy whereupon he will have a largely blank slate to draw upon. That’s when he’ll need to start convincing the US public he has a real plan or face the possibility of public discontent at home. He has indubitably been granted a profound and powerful political platform for change albeit one powered by an unstable base of social conservatives, evangelicals, white supremacists and laissez faire libertarians liable to fall apart without any unifying electoral enemy to galvanise against. Whether he has the ability to exploit his mandate for the benefit of the US and the world is something we all get to find out from January. The job of US President requires a lot of stamina, attention to detail, negotiation, diplomacy and frankly hard work. Trump has a huge mountain to climb to avoid disappointing the hopes and expectations of the millions who voted for him – many of the problems he inherits such as irreversible erosion of the US manufacturing base are not easily fixable and prospects for blue collar manufacturing workers remain bleak – the jobs they hope for will either be done overseas or by US based robots in the future. Tim Urban among others suggests reasons why it is going to be okay but then he (like Trump) is a privileged white male. The next few years look could be very difficult for many not in that circumstance across what looks today a disunited States of America.
For anyone thinking this Brexit 2.0 puts some kind of halo on the original one, it’s worth reflecting on the implications of a Trump presidency for the UK outside of the EU. Trump may portray himself as “Mr Brexit” but you sense it is to channel the same underdog energy rather than any genuine meeting of transatlantic minds. In fact there are massive policy differences likely to emerge between the UK and US over the next four years. In particular, Trump’s “US first” trade protectionism, stated intent to unravel major international accords on climate control and Iran, and his skepticism of NATO. In many ways, his ascent could hardly have come at a worse time for the UK which finds itself at risk of being cast adrift of both EU and US spheres of influence on the critical issues of trade and security:
If, as seems likely, Britain leaves the EU’s customs union on quitting the organisation, it may well find itself trying to negotiate new trade terms at a time when economies around the world are pulling up the drawbridge. … Then there is security. A staple of the pro-Brexit campaign was that the existence of NATO made European defence cooperation unnecessary and that quitting the EU would thus not knock Britain’s influence as a military power. That did not reckon with America’s next president being as equivocal about NATO as is Mr Trump, who has pledged an “America first” doctrine requiring countries under its security umbrella to make their own arrangements. Britain could thus find itself falling into the gap between a less effective, more divided NATO on the one side and rapid moves towards EU defence integration on the other.
The main connection is that Trump’s victory has ratcheted up the general level of post-Brexit global uncertainty and unpredictability. Or as Vlad Savov in the Verge put it:
Trump’s election is, as he himself had promised it would be, Brexit 2.0 — only exponentially bigger, more consequential, and lacking the Brits’ dark sense of self-deprecating humor.
One thing’s for sure; it is safe to assume Brexit will be the word of the year in 2016.
A key feature of the election has been the utter failure of the Fourth Estate to forsee a Trump triumph. With a few notable exceptions including Michael Moore, Scott Adams and Allan J. Lichtmann, few pundits or journalists gave him any chance. Week after week, the majority of mainstream press piled into Trump. Key tech influencers like Mark Suster continued to ratchet the stakes to existential level in the final days. Newsweek reported extensively on Vladimir Putin’s keen support for Trump, something that just a few decades ago during the Cold War era would have been utterly inconceivable. None of the opprobrium stuck. In fact it very likely didn’t even register with most of those that voted Republican. The post-truth landscape has shifted that much. Politicians around the globe need to face the reality that reactionary forces are in the ascendent not just in the UK and US but more widely across Europe and facts won’t stop them though failure to deliver may well do. They are reaping the reward for years of careful groundwork to present a more persuasive case to the electorate.
Tech was central in enabling the election of Donald Trump. Social media fuelled and fed his network of supporters on the ground helping to neutralise the campaign funding disparity. Trump’s extensive use of Twitter essentially allowed him to inflate the first ever lean startup style “B2C” populist campaign on what was by previous standards a very small budget and use it to cut across negative mainstream publicity:
The public discussion about technology’s impact on the election has mostly centered on the direct-to-consumer populism of Donald Trump on social media, but this goes way beyond Twitter and Facebook. The real story is about what it means to operate a democracy in a world of perfect memory and free communication. … We have to come to terms with the fact that if everything is remembered, there will be tons of unflattering stuff available on everyone. Of course, not all embarrassing stuff is equal, and we have to learn to forgive some things without excusing the unforgivable.
The collapsing cost of communication and a huge video-led inflation of information available to communicate does not seem to have led to a more greatly informed public, however. It seems to have served to reinforce existing confirmation bias in the opposing camps. The huge gap in understanding between different interest groups suggests that Americans (and Britons for that matter) are already living in a “build your own reality” era insulated in separate tech-supported echo chambers which are no longer possible to puncture using traditional political approaches only through theatrics. Of course, it took a reality TV star to demonstrate that. Further advances in technology promise to make it a whole lot worse yet. Future American History may be chronicled Black Mirror style:
If you think this polarization of reality is bad now, you’d better strap in. Technological innovation is going to make it much worse by facilitating the manufacture of alternate realities. … Just imagine how much worse the echo-chamber effect can get when people can easily manufacture photos, videos, and audio recordings to support their position. Disproving these false recordings could be very difficult, but in the build-your-own-reality era, I’m not sure that proof will matter anyway.
Ironically the very companies that enabled Trump supporters to build online communities also seemed to end up lumped in with the elite as Trump became a beacon for disaffected blue collar workers at risk of technological unemployment. The Information reflect the thoughts of in tech in suggesting that a rise in “anti-technology” sentiment is a plausible outcome from the result. Perhaps the Silicon Valley robots will be paused in their relentless march to the Singularity by the truck drivers standing in their way:
“My sense is that the electorate that swept Trump into power is an anti-intellectual, anti-technology electorate. I would not be surprised if he includes the tech industry as part of an overall group of enemies that would include foreign countries, immigrants, intellectuals.”
Jeff Bezos of Amazon would seem to be right in the cross-hairs given his ownership of the Washington Post. Many will watch carefully to see how much effort Trump expends on supporting the continued global dominance of US tech companies versus trying to get even with the likes of Bezos for perceived personal slights. The stakes are high not least for foreign tech ventures and non-US based operations which stand to profit from any fallout. Also for Trump who will surely be held to account on the day job he so craved. Either way, the deed is now done and the pollsters can go back to the drawing board while everyone else can look to the bright side – at least the awful US electoral circus is over for another four years.
Steven Levy on why Google Assistant needs you to talk to it to help build its knowledge base of conversational text. That in turn will provide the raw material to further improve its performance dramatically over the next year or during a profound paradigm shift called “The Transition”. According to Fernando Pereira who heads up natural language processing (NLP) research at Google:
This 2016-to-2017 Transition is going to move us from systems that are explicitly taught to ones that implicitly learn.” Think of it as a mini-Singularity.
More “AIAIO” than AI, an Amazon Alexa enabled fish had its fifteen minutes of fame this week:
who did thispic.twitter.com/Wriy531ALr
— 🦃 Wolf ウルフ (@Ouren) November 4, 2016
Les Schmidt, a software industry veteran takes on machine learning in the first of a 3 part series. He highlights the huge potential for disruption in the healthcare sector – much of the work of detecting abnormalities requires human experts to interpret visual information during inspection. Many such cases could be offloaded to machine learning systems which are already capable of superb image anomaly detection given existing datasets to train models from:
Increasingly, real-time customer decisions and manufacturing processes are creating new data sets that continuously feed machine learning algorithms. As conditions change, the machine may or may not be equipped to handle the new conditions. Today’s tools and technologies enable the ability to rapidly update or prototype changes to the machine’s knowledge and rule sets.
The machines are coming for US jobs right across the spectrum whoever you voted for.
Bruce Schneier’s recent ominous assertion pre-Mirai that “someone is learning how to take down the internet” was picked up in last week’s blog. This week the Economist follows suit. Here’s their colorful take on how Mirai operates highlighting the enabling role played by poorly designed IoT kit:
It mainly scours the internet for devices such as webcams, digital video recorders and home routers in which easy-to-guess factory-set passwords (“12345” or even “password”) have not been changed. The program then turns those it can gain access to into a huge army of digital slaves that can be directed to inundate targets with requests.
As if things aren’t sufficiently grim in the real world, Black Mirror’s Shut Up and Dance episode offered a blisteringly unforgettable glimpse of darkness visible. Central to the viewer’s extreme discomfort is the mundane proximate nature of the dystopia it outlines – a world where smartphone addiction meets Mirai and paranoid nightmare ensues:
The Black Mirror episode “Shut Up and Dance” is hacking torture-porn, and it doesn’t need to be set in the future for the story to work. Hacked webcams, GPS tracking, catfishing — all the tools are here already.
It’s not often you see something that so profoundly shakes unspoken core beliefs in the progressive possibility of technology. Then again, it’s not often you watch something that makes you unable to sleep, a sentiment shared by several reviewers:
It’s worth clarifying that Shut Up and Dance is, by some distance, the most nihilistic episode of Black Mirror so far, with no space for comfort and an acrid aftertaste that doesn’t easily shift; that its denouement unfolds to Radiohead’s Exit Music (For a Film) should give you some idea of the concluding mood. Its vision of humanity – as creatures who’ll stake their lasting happiness on a squalid, lizard-brain-twitching thrill – is clear-eyed and uncompromisingly negative. It’s also closer to the truth than any of us would probably care to admit.
The Atlantic too felt a sense of urgency about disabling that baleful invisible panopticon on your laptop staring into your soul invisible. You probably will too if you haven’t seen it yet. Exit Music (For a Film) has rarely sounded better:
I didn’t take anything away from this episode other than a sense of doom, and an urge to cover up every camera I own.
Apps and Services
The Information profile the “small programs” feature that Tencent are planning to launch for WeChat. It would allow developers to build ‘instant apps’ that reside within the WeChat app framework in a mobile analogue of the way Facebook host third-party apps. It’s an interesting development and one that could further cement WeChat omnipotence in China:
WeChat’s pitch to software developers is that instead of having to build one version of their app for Android phones and another for the 20% of Chinese who use iPhones, they can just build on WeChat to serve both sets of customers. And the use case will get strengthened as more users find it natural to stay within WeChat to open the easier-to-build mini-apps. That’s an especially attractive proposition as Chinese users are loading fewer and fewer apps.
Stratechery on why Apple should buy Netflix to gain access to their superior user experience and unparalleled broadcast content library. The only other contenders in the global streaming video on demand (SVOD) space are Amazon Prime Video (which is not up for sale) and Hulu (which might be):
The problem Apple has in premium video — and given that the company has been trying and failing to secure video content on its terms for years now, it definitely has a problem — is that its executives seem to have forgotten just how important the piracy leverage was to the iTunes Music Store’s success.
Meanwhile Amazon Prime has launched in China albeit in a neutered form which doesn’t allow access to Amazon streaming content including Prime Video. Right now, Amazon have less than 1% eCommerce market share in China. It’ll be interesting to see if this grows.
Cloud and Data
Interesting rather contrarian insights into the reality of using the Docker containerisation framework in production. The gist of the piece seems to be “don’t bother” and wait for hosted Kubernetes courtesy of Google Cloud Platform (GCP) which gives way to this punchline:
One more reason why Google Cloud is the future and AWS is the past (on top of 33% cheaper instances with 3 times the network speed and IOPS, in average).
The instability and shifting nature of Docker alluded to in the previous link is a recurring theme in this article which presents the case for Terraform as your go-to IAC (Infrastructure as Code) tool choice. Essentially it is feature-equivalent to AWS CloudFormation but not tied to a specific platform:
Ten ways your data project is going to fail is entertaining reading and all too familiar. The point about executive support and being careful to set realistic expectations ring particularly true.
Handy resource page containing information on Linux performance. Here’s a sample:
Ten emerging coding trends “you need to know” which include React, Bootstrap, Grunt, Scala and er … Docker.
This is brilliant in multifarious ways. If you’ve ever worked on commercial software development you will laugh too:
— ashley maar in nl (@rabcyr) November 2, 2016
The bewildering and exhausting array of obstacles software developers face in China is excellently portrayed in this TechInAsia post:
If you’ve ever been to mainland China, chances are you’re familiar with the Great Firewall, the country’s all-encompassing internet censorship apparatus. You know the despair of not being able to open Facebook, the pain of going mute on Twitter. But with a good VPN, you can magic many of these inconveniences away – at least temporarily. … For software developers based in China, however, it’s not that simple. You’re not just censored from certain websites. Basic building blocks that you use for product development are suddenly beyond your reach. With software services and libraries spread across the globe, China’s internet sovereignty can be a real pain in the ass.
The crazy creative destruction that is the mobile sector in China is underscored by the news that Oppo and Vivo have now both leapfrogged Xiaomi to become the top two native Chinese smartphone OEMs. It’s a dramatic turnaround in fortunes for the poster child of Chinese tech in less than a year. TechInAsia tries and get to the root of what happened to Xiaomi here. A combination of bad luck, missteps and growth in the low to mid tier space.
Ewan Spence on how Apple’s continued smartphone dominance provides the revenue that allows the company to overwhelm it’s would-be Android rivals:
Its financial power allows Apple to invest heavily in new technology and services. Most Android manufacturers do not have equivalent power and are left to use off-the-shelf solutions, reference designs, and a race to the bottom in terms of pricing. Where will the next smartphone advances come from?
A look inside Apple’s custom GPU developed for iPhone6 and beyond – some of the fruits of that profit:
Just as Apple began by licensing standard ARM CPU cores but now designs its own, we believe the company has similarly shifted from licensing PowerVR to designing a custom GPU. This new GPU first shipped in the A8 processor that is in the iPhone 6, and its descendants are also in the A9 and A10 Fusion processors in the iPhone 6S and 7.
Many are speculating that this GPU will be leveraged by future iPhone-based AI features. It’s a good example of how AI is beginning to change the game for the chip manufacturers who face serious disruption unless they move fast:
Today, Internet giants like Google, Facebook, Microsoft, Amazon, and China’s Baidu are exploring a wide range of chip technologies that can drive AI forward, and the choices they make will shift the fortunes of chipmakers like Intel and nVidia.
Economist piece on niche phones calls out UK based OEMs Vertu and Bullitt.
Why reminding everyone “You’re all going to die” remains hard to beat as a way of commanding attention and remains a ‘scientifically proven pep talk for winning‘.
InfoQ presentation on OKRs (Objectives and Key Results) and how they work within a Lean Startup context.
Bloomberg on why conventional job interviews “are useless”. They merely reinforce both Confirmation Bias and Unconscious Bias. Expending effort on setting up a strictly objective scoring offers better outcomes. It helps explain why Behavioural Interviewing approaches are increasingly in vogue.
There is something unutterably sad about this article explaining why the light you see in the distant night sky using telescopes in many cases comes from galaxies we will never be able to reach because they’ve already moved out of our light cone.
Courtesy of Quora, the world’s shortest intelligence test. You may have seen some of these before. They’re easy enough provided you think before going with your intuition and gut instinct. That’s why it’s called an intelligence test.
Relatedly, NPR on the difficulty of learning in the era of constant digital distraction in which mechanisms evolved to serve us in the world of the hunter gatherer have been adapted and co-opted for “information gathering”. Tell me about it:
Monkey see. Click. Swipe. Reward.
Last word this week to the pollsters and how much people trust judges vs politicians. Wonder where pollsters would be on this index.
— Ben Page, Ipsos MORI (@benatipsosmori) November 4, 2016