Why I don’t post on Facebook any more

According to the ticket stub, it was 15th December 2016 when we went to a Red Hot Chili Peppers gig in Manchester. We met up in town after work, went for some food in the Corn Exchange and then went into the Arena. Up first was the supporting act – the fun and frenetic Baby Metal. Almost as entertaining as the band themselves was looking round the audience to see who looked the most freaked out by the J-Pop outfit.

Then the main act came on. It was actually my wife who wanted to see the Chili’s, but it was a treat to see such a good gig. Anyone who can manage to sound good in the echo-filled Manchester Arena is a class act, and the light show was beautiful. So good in fact, that like everyone else we took lots of photos and videos on our phones.

Back at home later on, I opened Facebook for the usual mindless scroll through my feed. But sitting at the top of the screen was something I’d never seen before. It was a banner that said “Dorian, we’ve made a video of your evening out to share with your friends”. The link opened up a video edited together from the photos and video clips I’d taken over the evening. What?!?!?

I’m not sure if this feature rolled out to everyone or if it was something they were testing on a selection of people, but it didn’t go down well with me. They’d helped themselves to my photo library, and processed it without my consent.

Technically, I had granted Facebook access to my photos, but that permission had been abused. When anyone wants to post a photograph on Facebook, they have to grant the app access to their Photo library just to be able to select a photo. Once the app has this permission, it can access photos in the background without user interaction, which is quite a different thing. Apple should definitely consider splitting this permission into two – one for selecting a photo and another for full access to the library, even for apps running in the background. Even so, it was still Facebook that took advantage – and it shouldn’t always be on Apple to have to firewall their users from apps made by tech companies that should know better.

Then there was the question of how Facebook knew I was on a night out – how exactly did their algorithm work this out? Did it notice my location was a well-known entertainment venue? Or was it looking at the quantity of photos taken that evening? Even worse, was Facebook running my images through AI and recognising it as a concert? However they did this, it’s creepy.

It would also be interesting to know where the processing to build the video took place. If they uploaded my media to Facebook’s servers and processed it there, that’s naughty. If they made the video locally on my phone, that’s more acceptable from a privacy point of view, but then there’s the hit on battery life that I took. And if they did process it locally, how did they manage to keep the app running in the background long enough to do it, when iOS is so good at preventing this? There’s been evidence of Facebook turning on the microphone (mimicking audio recording) as a way of tricking iOS to keep their app open for longer in the background.

I have never posted on Facebook again since that evening. My first defensive step was to revoke access to my photographs and carry on using the app, but it didn’t take long to realise that I didn’t trust the app at all so it was uninstalled. Occasionally, I’ll open Facebook in a browser and check on how the friends I don’t see regularly are getting on. I clicked the like button on all the birthday message I got recently, and feel guilty that I won’t be sending any back. So far, I have resisted deleting my account.

Since I stopped posting, the justifications for not using it have only grown. It’s been abused by fringe political groups and Russian intelligence operations, and directly affected UK and US political outcomes. They’ve caused dangerous situations in countries like the Philippines, where the government uses it as a platform to incite violence against news organisations trying to report on human rights violations.

The only thing I can do to show my dissatisfaction with the way Facebook run their network is to delete my account completely, which I pledge to do in 2019.

iPhone design decisions and poor battery life

At the end of last year, I replaced an iPhone 7 Plus with a OnePlus 5T, after deciding against the iPhone 8 Plus (too similar to justify the upgrade) and the X (not worth the money). While researching this purchase, I couldn’t help notice the significant battery capacity difference between an iPhone 8 Plus (2,691mah) and the OnePlus 5T (3330mah). The much bigger battery on the OnePlus is surprising given that it is an almost-identical form factor, has a much larger full-front screen, and has a headphone jack.

Apple has been making phones for much longer than OnePlus, and is a significantly bigger and better-funded company. So why are they selling phones that struggle to provide sufficient internal storage for their batteries?

To understand this, we need to look at the history of the iPhone 8. It’s the third revision of the iPhone 6 design that was introduced in 2014. But the original design actually had a larger battery – the 6 Plus has 2915mah. Each revision has made changes that affected the capacity:

1. In 2015, the 6S introduced 3D Touch, whose extra hardware reduced the battery capacity of the Plus sized model by around 5%, to 2750mah.
2. In 2016, the iPhone 7 introduced a larger Taptic Engine, so something had to give to avoid fitting an even smaller battery than the 6S. This (not courage) is the real reason why the headphone socket was removed. It put battery capacity back in line with the original iPhone 6 Plus design, at 2900mah.
3. In 2017, the iPhone 8 introduced wireless charging, requiring an inductive coil and a glass back. This undid the gains made by the 7, and then some. Battery capacity fell to just 2,691mah on the Plus.

Whenever battery capacity was reduced in a new model, Apple always said that it wouldn’t make any difference because they offset the loss by efficiency improvements in software and hardware. This isn’t always the case. With the iPhone 6s, benchmarks by Ars Technica  showed the 6S Plus battery life down by around 7% in the web browsing tests. Although the 7 improved matters, the 8 Plus came in around 4% worse than the 7.

3D Touch

Apple must have felt optimistic about the potential for 3D Touch, because they decided it was worth the trade off of a 5% reduction in battery. As with the MacBook Pro’s Touch Bar, it’s clever technology but of debatable user benefit. Android doesn’t have an equivalent of 3D Touch, but it’s home screen icons offer very similar context menus from a long press, and it works perfectly well. 3D Touch doesn’t live up to it’s promise because iOS still has to support older devices without it. Look at how deleting an app still requires a long press on a 3D Touch phone. It could be much slicker as an option on the 3D Touch menu.

Wireless charging

With wireless charging, the trade off is even bigger than 3D Touch – around 10% of battery capacity was sacrificed to make way for the glass back and inductive loop. Wireless charging should prove convenient and popular with users, so perhaps the trade off is worth it, but 10% is a lot.

Screen technology

Almost all top-tier Android phones moved to OLED screen technology years ago, which is thinner than LCD because it requires no backlight. Until the iPhone X, Apple insisted that OLED screens weren’t good enough, and stayed with LCD. The combination of an LCD screen in a very thin phone does limit available space for other components.


After looking at the history of the 6/7/8 series of phones, a story emerges of repeatedly adding significant new technologies into a design that was made to be as thin as possible. Instead of redesigning the case to provide more internal space, compromises were made with battery capacity and the headphone jack. Given the premium price of the iPhone, and the highly competitive smartphone sector, this is a questionable strategy. It’s as if Apple itself was unable to compromise and make the phone larger, lacking the confidence to deal with criticism.

With the iPhone X, Apple has switched to OLED screen technology and optimised their internal layout to accommodate a much larger battery relative to the size of the phone. Unfortunately, despite charging premium prices for the iPhone 8 with it’s four-year-old design, they think customers should pay a premium on top of that for a modern design.

Why I fell out with my iPhone in 2017

It has been a fantastic run of 5 years of iPhone ownership, but now I own a OnePlus 5T costing less than half the price of an iPhone X. What happened?

In September 2016 I bought a new iPhone 7, and returned it a week later because the battery life was terrible. It looked great in it’s new all-black finish, but I’d lost the headphone jack and the battery was worse than the two-year old iPhone 6 I foolishly traded in on the assumption that it’s replacement was a better phone.

After returning the 7 I bought a 7 Plus, figuring that the benefits of a bigger battery and better camera would outweigh the more inconvenient size. It turned out to be true — the Plus is a phone so big I’d have laughed at anyone for owning one not that long ago, but ownership changed my perspective. The size soon felt normal, and the camera is really, really good. On battery life, the whole “power user” badge is a bit cheesy, but I use Google Maps for 2 hours a day while driving, listen to streaming music via AirPods for hours, and have an Apple Watch. There’s also the usual gaming and surfing in the evening. Power user or not, the 7 Plus took it all in it’s stride. It was a boring phone, but a solid and trustworthy one.

When it was nearly a year old, I could see that the battery wasn’t quite as strong as it was, but there was still plenty headroom and it easily got me through the day. Then iOS 11 arrived. Straight away the phone was struggling with battery life, with the low power warning at 20% appearing every evening. It now felt like owning a Plus-sized phone but with standard iPhone battery life. A 12-month old phone that cost £819 was no longer performing as it should.

Annual iOS updates often have bugs or efficiency problems, and that’s fine to start with as long as they are eventually fixed. When Apple ships new iPhones and a new iOS, it’s perfectly reasonable for the new models to get top priority for tuning and fixes. What usually happens is that older devices get some love in a point release. But 11.1 and 11.2 came and went, and still no improvement. I tried various wipes and resets, even a clean install. This is not an unusual experience either, forums and social media are full of user complaints since iOS 11 shipped. Meanwhile, I spoke to friends on a different upgrade cycle who bought the 8 and 8 Plus, and they have no problems at all.

After 11.2 was released, I came to the conclusion that Apple wasn’t going to fix this, and started to look at alternatives. Buying an iPhone 8 Plus wasn’t attractive because it would be the third phone I’d bought with the same design, and it was too minor an upgrade to justify the cost. The iPhone X is a nice phone, but I don’t think it’s worth the money. Apple is clearly experimenting with how much people will pay for a phone, and as a disappointed Apple owner, it was out of the question to hand them so much money. Also, I don’t like the notch. Not because it actually bothers me, but it appears to exist purely to give the phone a unique appearance for branding and marketing. In the end, I decided that the current range of iPhones doesn’t interest me, and that isn’t going to change until September 2018. It was time to have a look at Android.

Too much choice, and none of them are perfect

Over in the Android world, choosing a phone is a lot more complicated because nobody makes a phone as as well-rounded as an iPhone. The Pixel range is probably the closest, with great software and camera, but the Pixel 2 XL is very expensive and the screen isn’t up to scratch. Eventually I settled on the OnePlus 5T. Although not a cheap phone, it’s price is more appropriate for someone looking to try Android but perhaps not stay there — it represents amazing value at £449/499 and I knew I could sell it for close to the new price if things didn’t work out.

But the interesting part of this is that it has worked out. Apart from one major drawback, the phone hardware is excellent. Performance, screen quality, battery life are all superb. It even has a headphone jack. And Android? It’s a real surprise. In the Gingerbread and Ice Cream Sandwich era, Android had some advantages but couldn’t complete on polish and usability. By the time Lollipop arrived, app quality was on a par with iOS, and Android wasn’t far behind. Nougat on the OnePlus 5T is simply excellent. Apple’s processors are the most powerful on the market, but the OnePlus feels faster in normal use. It’s got a ridiculous amount of RAM and can keep lots of apps in memory, so switching between them is really fast. Even better, those long standing issues with lag and dropped frames are no longer part of Android due to serious effort put into fixing this with 7.1. Even games run great.

So what’s the major drawback? It’s the camera of the 5T. I knew this was a problem area prior to buying, but it was hard to judge just how good or bad it would be. Professional reviews warned it was pretty bad. Other reviews thought it was ok, and of the sample images I checked, quite a few pictures looked perfectly reasonable. Now I own the device, I really dislike the camera. Outside shots are passable, but inside shots are really bad, being gloomy and underexposed even in bright light. And if you zoom in even slightly on them, there’s tons of noise. That ugly digital watercolour effect is very strong.

The other downside is that I can’t use my Apple Watch any more, which is a disappointment. But, that’s not Android or OnePlus’s fault, it’s Apple’s. In my five years of iPhone ownership I brushed off the whole story of Apple having a closed ecosystem or walled garden. But it’s true. It’s just that you don’t notice it from the inside.

The future of Windows: Linux and a built-in Xbox?

A few weeks ago I wrote a post about how Windows could move to a Linux kernel, as a cost-saving strategy for Microsoft. Windows would be rebuilt as a custom GUI layer and set of services that runs on top of Linux. Legacy Windows applications would run under virtualisation, marking the end of development of the Windows kernel and OS. After giving it some more thought, there is one rather large problem with an otherwise plausible idea – gaming. The PC has an enormous catalogue of Windows games, and this is a major benefit that keeps people on the platform.

Gaming apart, there are far fewer blockers for moving to Linux than there were ten years ago – we rely less on Office and other productivity applications because we have alternatives such as cloud services. But games do present a problem – Steam tried to build a gaming platform on Linux but the size of the catalogue is tiny in comparison, and a lack of investment in drivers means there is usually a performance gap. For a huge library of games that can run on a massive array of hardware configurations, Win32/64 and the DirectX graphics library are unparalleled.

So how could our idea of a new Linux platform with a Windows GUI and Win32/64 compatibility layer accommodate DirectX games? Well it turns out that a very similar problem has already been solved – by Microsoft themselves. The Xbox One console makes clever use of virtualisation to run apps and games in separate virtual machines on top of a lightweight hypervisor. It enables the impressive trick of Xbox games running live in a window on the Xbox home screen. When this hypervisor solution was first announced it sounded like a solution that could incur a performance penalty but it is now clear that  Microsoft’s engineers were able to sidestep this.

Taking the hypervisor idea a step further, the Linux Windows of the future could use virtualisation to divide and conquer the problem of migrating to a more secure Unix-based kernel while still keeping compatibility with DirectX games.

There are two different ways to solve this problem:

  1. A hypervisor boots the new Linux-based Windows platform, which is solely for new apps, and a separate virtual machine runs DirectX games and older Win32/64 applications.
  2. A hypervisor boots the new Linux-based Windows platform which runs new apps and contains an embedded Win32/64 virtual machine for older applications. A separate native Win32/64 instance runs purely for gaming, which is very similar to the game virtual machine that the Xbox One already runs. It’s an elegant solution that cleanly separates gaming and applications at run time.

The second option looks preferable because it cleanly separates the application and gaming sides of Windows, and it aligns the PC nicely with the Xbox platform, which is a direction Microsoft have been moving in for some time. This option also appears more flexible for future rewrites and consolidation – the kernel of the gaming VM could be stripped back and re-engineered because games use a smaller subset of the Windows API than regular apps.

In the future, perhaps all PC’s will be Xboxes?

How long will Trump last in office? 3 years at most

Trump has only been in office for a few weeks and people are already wondering how long he will last. Will he resign, be impeached or even assassinated? Or will he defy his critics and last the full two terms as he suggested in that dire press briefing at Trump Tower back in January?

Let’s get the easy one out of the way – he is extremely unlikely to be assassinated. Just like his predecessors he will have excellent security, and the kind of people likely to bear arms probably support him anyway. During the presidential campaign, Trump gained support from gun owners by saying Hillary wanted to take away guns, and even made a vague suggestion that someone should stop her by shooting her.

Impeachment is harder to predict, although even at this early stage there are possibilities. Investigations into his business interests show that they might not have been severed as cleanly as promised. There is also the more visible matter of his disregard for the US Judicial System which goes as high as the Supreme Court. To make matters worse, he has made enemies within his own Republican party, and government departments such as the CIA.

What is more likely is a resignation before the end of his first term. His approval rating a few weeks after entering office is just 41%. That’s especially bad given that a massive campaign to secure voter approval and win the election has only just finished. Usually a president starts with a high approval rating and it tails off as the term progresses. Going south from 41% is going to be unpleasant. There are tough times ahead for both the United States and Trump.

The most likely reason for Trump to resign is because of who he is and the life he has led. His story is not one of rags to riches, it’s riches to riches. He’s led a sheltered and privileged life, living in a bubble of ‘yes men’. Criticism or even constructive feedback is not something he will have had much exposure to. As a TV personality he understands that you live and die by your popularity ratings and whether people are talking about you. He shows clear signs of not being able to ignore or deal with criticism. And Trump has been the target for significant criticism, protest and negative TV coverage. This was happening before the election result but it’s intensified since he moved into the White House. His social media platform of choice is full of angry tweets from all over the world.

It’s easy to think of the President of the United States as a very powerful man, and Trump certainly is, but there is also a lot of evidence that the man behind the job title is insecure and craves acceptance. Some of the opinions and policies that are making him unpopular aren’t even his own, but originate from people like Steve Bannon. That wont make it any easier to deal with criticism and a dwindling approval rating.

Trump currently appears to be a man in a living hell, where a dream to become the most powerful man in America somehow came true and he has woken up to a reality that is far removed from the dream. He’s isolated in the White House, and is apparently obsessed with watching his critics on CNN. If you don’t like him, take some comfort in the fact that he now exists in his own personal Twilight Zone.

Assuming Trump is still in the job after two years, he will have to deal with mid-term elections. It’s normal even for a popular president to lose ground in the mid-terms, but Trump is likely to suffer a huge blow. A good proportion of his voters were sold on pledges to end corruption, bring jobs home and take on Wall Street, but these already feel like distant memories. His more extreme policies and erratic behaviour will cost even more votes.

After the mid-terms, the job of president will become even more challenging if the majority in congress is lost, as Obama found during both his terms. A year of roadblocks in Congress on top of being chipped away at by his critics on Twitter and numerous mainstream media outlets is going to take it’s toll. 

By the time his third year starts he will have had enough.

Fixing British Politics Part 2: Enough of the broken promises

During an election campaign, a political party will set out it’s vision of how the country will be transformed when they get into power. As you would expect, it’s a positive message, containing the political equivalent of palm trees and sandy beaches. Nobody wants to hear about how they might find it harder to get a job, or have to pay more taxes. The problem with this whole charade is that the vision is usually rather non-committal. It might sound impressive but the use of weasel words ensures that the number of genuine promises being made is surprisingly small.

By the end of the term in office, there is often a significant disparity between what was promised and what was delivered. A memorable recent example is the broken commitment of the Liberal Democrats to scrap university tuition fees, and during the Obama presidency in the US there are many more examples.

In the competitive business environments where many of us work, this sort of behaviour is unacceptable. We all know of someone who found themselves out of a job because they failed to deliver a critical project or manage a department properly.

While it’s good for government to have a vision, it’s less important today because our society has already been built. We are now in the process of incrementally improving it, so the focus should be on delivery and management. Massive changes on the scale of the welfare state and the national health service are almost unheard of.

So why are governments not held accountable to their election manifesto in a formal way? And is it even viable to do so? Running the country is a difficult task and unexpected events take place that could significantly impact the priorities of a government. Would it really be fair to tear apart the Japanese government for failing to deliver on promises when they had to deal with a massive nuclear catastrophe? Also, no government works in isolation – it has to work with other nations and organisations like the EU and the UN.

Clearly, it’s not as simple as creating a checklist from an election manifesto and blindly assessing a government against it without applying some judgement. But even so, is it appropriate to have no accountability whatsoever, to allow manifestos to say things that politicians know have little chance of becoming reality?

I believe a party running for election should include a concrete list of the projects and programmes it intends to deliver. And at the end of the term of office, that list should be independently scored. An overall score should be lenient – given the constraints and shifting priorities faced by any government, delivering even 60% of the original plan is probably a commendable achievement. But over time, the rating should improve.

If shifting priorities mean that new projects and programmes must be started, they should be grouped into a referendum vote because they diverge from what the public originally voted for. Obviously, any emergency measures to deal with natural or humanitarian disasters would have to be exempt from this process.

While it is possible to find web sites that rate presidents and prime ministers on their record against election promises, I want government itself to embrace this rather than leaving it to third parties. They should rate themselves accurately and take pride (or disappointment) in their scorecard.

Fixing British Politics Part 1: More Referendums, not less

In July 2016 we held a major national referendum on whether to leave or remain in the European Union. This was an political gamble which attempted to settle a lengthy and bitter divide in British politics that was being exploited by niche parties like UKIP. The somewhat unexpected result was a narrow majority in favour of leaving. In the aftermath of the referendum, we can point to many issues, but the two most obvious flaws were:

  • Facts – both sides ran their campaigns poorly, with highly negative campaigning from the Remainers and a lack of vision from the Leave side. As a result I don’t believe the public had a real grasp of the situation and what was at stake. We were fed soundbites such as “£350 million a week goes to the EU”, and “economic ruin if we leave”. Neither were true. We had endless TV debates where politicians argued over the case for staying in the EU, but there was always someone from the other side arguing just as passionately to leave. Then there was the press – an ever-present blight on British politics, with bitter and misleading reporting from the likes of the Daily Mail. Were the British public really equipped to take in all of this and filter the lies from the truth?
  • Voting strategies – some of the Leave votes were protest votes against the Government, the EU, or even the establishment in general. Given the disenfranchisement and trust issues already explored on this blog, this is quite understandable. The trouble with protest voting is that it’s somewhat lazy because there is no requirement to think about the issue in any detail, and as a result those voters failed to accurately answer the question that was asked of them. I suspect that our political establishment will deal with this by problem by never having another referendum, eliminating the chance of another result that goes against establishment expectations. The excuse will be that the public simply cannot be trusted to make complex decisions.

I believe that the answer to the woefully-executed 2016 referendum is to “lean into the pain” and have more of them, not less. Holding regular referendums on a wide range of issues must surely be the single biggest thing we could do to re-engage people with politics. The possibilities are endless – imagine a Conservative government is in power but holds a referendum on whether to increase taxes to fund the NHS properly. That could lead to a number of interesting outcomes, such as right-wing voters being able to “switch sides” for something they believe in, and support a choice which might traditionally be perceived as socialist. The opposite is of course true – left wing voters might opt to privatise an industry because they’ve had enough of government bureaucracy and inefficiency.

It sounds like a much more robust decision-making process than lurching between left wing and right wing governments every ten years or so, and letting them spend years undoing each others work. Frequent referendums would mean that it would be less important which party was in power. When the party we don’t like wins an election, we usually retreat into a sulk for four years and hope for a different result next time. Frequent referendums would keep us all engaged in the running of our country.

Getting people comfortable with regular voting on a wide range of issues will have the benefit of taking political bias out of referendums. Instead of “sticking it to the man”, the chances of rational voter thought on the issue increase. There should be other positive effects such as people taking more personal responsibility for their opinions and the way they vote. Sooner or later, everyone will feel the effects of the decisions they make at referendums, no matter how small.

It’s unlikely that this process would need significant oversight or regulation, but any proposed referendum must be reviewed independently before being launched, to ensure it is a sound proposition and the arguments are clear and factual.

The EU referendum was a major decision, so the process ran for a long time and was expensive. A more frequent referendum schedule sounds costly, but each instance would be on a smaller scale than the EU one. Switzerland has held an average of nine referendums a year over the last 20 years.

Just imagine how different the world could be today if there had been a referendum on going to war in Iraq without a UN resolution.