Let’s Shed Some Light on This

Do you know what would be really great, when I fly in from a time zone quite a few hours misaligned with the local one? It would be fantastic if you all did not immediately lock me in a room without any windows for several hours. I mean, that’s bad enough for people who have to work there all the time, but it’s murder for those of us whose home timezone is many hours adrift from local time.

Meeting rooms with actual daylight would be a great help to the much-abused circadian rhythms of overseas visitors. That way we can avoid coming down with a bad case of SAD.

Meanwhile, I will continue to self-medicate with my own cocktail of melatonin and gin&tonic. Cheers!

Image by Rawpixel via Unsplash.

The Road To Augmented Intelligence

A company called Babylon has been in the news, claiming that its chatbot can pass a standard medical exam with higher scores than most human candidates. Naturally, the medical profession is not overjoyed with this result:

No app or algorithm will be able to do what a GP does.

On the surface, this looks like just the latest front in the ongoing replacement of human professionals with automation. It has been pointed out that supporters of the automation of blue-collar jobs become hypocritically defensive when it looks like their own white-collar jobs may be next in the firing line.

For what it’s worth, I don’t think that is what is going on here. As I have written before, automation takes over tasks, not jobs. That earlier wave of automation of blue-collar jobs was enabled by the fact that the jobs in question had already been refined down to single tasks on an assembly line. It was this subdivision which made it practical for machinery to take over those discrete tasks.

Most white-collar jobs are not so neatly subdivided, consisting of many different tasks. Automating away one task should, all things being equal, help people focus on other parts of the job. GPs – General Practitioners – by definition have jobs that encompass many tasks, and requiring significant human empathy. While I do therefore agree with the RCGP that there is no immediate danger to GP’s jobs, that is not to say there is no impact to jobs from automation; I’d hate to be a travel agent right now, for instance.

Here is a different example, still in the medical field: a neural network is apparently able to identify early signs of tumours in X-ray images. So does that mean there is no role for doctors here either? Well, no; spotting the tumour is just one task for oncologists, and should this technology live up to its promise (as yet unproven), it would become one more tool that doctors could use, removing the bottleneck of reliance on a few overworked X-ray technicians.

Augmenting Human Capabilities

These situations, where some tasks are automated within the context of a wider-scoped job, is augmented intelligence: AI and machine-learning enabling new capabilities for people, not replacing them.

Augmented intelligence is not a get-out-of-jail-free card, though. There are still impacts from automation, and not just to the X-ray technicians whose jobs might be endangered. Azeem Azhar writes in his essential Exponential View newsletter about a different sort of impact from automation, citing that RGCP piece I linked to earlier:

Babylon’s services were more likely to appeal to the young, healthy, educated and technology-savvy, allowing Babylon to cherry pick low-cost patients, leaving the traditional GPs with more complex, older patients. This is a real concern, if only because older patients often have multiple co-morbidities and are vulnerable in many ways other than their physical health. The nature of health funding in the UK depends, in some ways, on pooling patients of different risks. In other words, that unequal access to technology ends up benefiting the young (and generally more healthy) at the cost of those who aren’t well served by the technology in its present state.

Exponential View has repeatedly flagged the risks of unequal access to technology because these technologies are, whatever you think of them, literally the interface to the resources we need to live in the societies of today and tomorrow.

My rose-tinted view of the future is that making one type of patient cheaper to care for frees up more resources to devote to caring for other patients. On the other hand, I am sure some Pharma Bro 2.0 is even now writing up a business plan for something even worse than Theranos, powered by algorithms and possibly – why not? – some blockchain for good measure.1

Ethical concerns are just some of the many reasons I don’t work in healthcare. As a general rule, IT comes with far fewer moral dilemmas. In IT, in fact, we are actively encouraged to cull the weak and the sick, and indeed to do so on a purely algorithmic basis.

It is, however, extremely important that we don’t forget which domain we are operating in. An error in a medical diagnosis, whether false-positive or false-negative, can have devastating consequences, as can any system which relies on (claims of) infallible technology, such as autonomous vehicles.

A human in the loop can help correct these imbalances, such as when a GP is able to, firstly, interpret the response of the algorithm analysing X-ray images, and secondly, break the news to a patient in a compassionate way. For this type of augmentation to work, though, the process must also be designed correctly. It is not sufficient to have a human sitting in the driver’s seat and expected to take control at any time and with only seconds’ notice. Systems and processes must be designed in such a way as to take advantage of the capabilities of both participants – humans and machines.

Maybe this is something the machines can also help us with? The image above shows a component as designed by human engineers on the left, side-by-side with versions of the same component designed by neural networks.

What might our companies’ org charts look like if they were subjected to the same process? What about our economies and governments? It would be fascinating for people to use these new technologies to find out.

Photo from EVG Photos via Pexels

  1. One assumes that any would-be emulator of Martin Shkreli has at least learned not to disrespect the Wu-Tang Clan

Unintended Consequences

One of the unintended consequences of the GDPR is that some US websites have decided that the easiest way to comply is simply to ban all European IP addresses. Now, IANAL, nor do I even play one on the Internet, but my understanding is that this does not shield them from liability, should a European citizen browse one of these sites while travelling in the US.

Regardless of the legal niceties, these restrictions get annoying when I want to browse one of these sites. Most of the time, the correct response to a site that does not want my business is to take said business elsewhere, but it does happen on occasion – and possibly by accident – that something interesting is published on one of these sites. For instance, Mike Godwin, of the eponymous Law, wrote a piece in the LA Times, clarifying how Godwin’s Law should be applied in our day and age.

The LA Times is owned by the company until recently known as Tronc, all of whose web sites block European browsers. Yeah, their policy is about as silly as their name. Normally I can ignore their entire stable of papers without missing anything, but on this instance I did actually want to read the article.

My solution involves the TunnelBear, a VPN service with a cute bear theme and, more importantly for my purposes, a useful free tier. 500 MB per month is a perfectly usable amount for the sort of usage I have, which is mostly text and the very occasional YouTube video which is inexplicably “unavailable in your country".

If you’re interested, check out the TunnelBear here. By using that link, we both get free data – pretty cool!

Why Sales People Will Always Be With Us

I was looking for something else in my drafts folder, and I found this older thing: Sales People: Do We No Longer Need Them?:

We all know that the internet, in particular, has made us – the customers - more savvy and more able to easily see when we're being “sold." Instead, many now believe traditional sales people should go the way of the dodo bird, and companies should offer something else — experience and customer service.

I think there is a kernel of truth here, but the argument goes too far with it. As in every market transition, there is a sieve effect at work. The good sales people were already operating in a way that is compatible with educated and demanding customers. This breed of sales person leads a diverse team and maintains the context, preventing discussions getting lost down rabbit holes. Bad sales people who rely on ill-informed customers and add no value will fail to make the transition – and indeed are already doing so. However, to bridge from the self-service check-out at the supermarket to the imminent extinction of car sales people – or enterprise sales teams – is too much of a stretch.

Example: my wife just bought a car, and sure, we had done our research, researched alternatives, built several configurations online, and she was pretty sure of what she wanted. She had narrowed it down to three alternatives, so we visited the three dealerships. Two of them had ignorant, pushy and unhelpful sales people - who did not get the sale. The third had a courteous, well-informed sales person who was passionate about his product, and helped us navigate the various finance options to get a deal that worked for everyone – including getting a notary to come to the dealership after hours so we could take advantage of a deal that was about to expire! That sales person earned his commission – and the sale.

Sales is definitely changing, but it's not going away. The only sales people who will lose their jobs are the ones who fail to adapt and evolve.

If this sounds familiar from similar screeds I have written about how AI and automation are not going to take away sysadmins’ jobs, you are exactly right. The fact that one task goes away is only a problem if that single task defined your entire job. Sure, it sucks if you’re a supermarket checkout person, because that automated checkout lane is definitely taking your job scanning barcodes by hand. On the other hand, the introduction of ATMs increased employment in banks for a long time (although it is now declining due to branch closures).

I would like to close from a quote that came up in conversation earlier today:

Ultimately, everyone’s job is sales – or we’re all out of a job.

Photo by Fredrick Kearney Jr on Unsplash

Privacy Policy

Short version: I don’t have one.

Long version: I don’t gather any data, I even turned off Google Analytics (and not just because it was depressing me with its minuscule numbers!), and I don’t have access to the server logs even if I wanted to look at IP addresses or whatever. This blog’s host, Postach.io, have their own privacy policy here.

Regarding analytics specifically, I am somewhat curious about how many people read individual posts, but I’m not going to sell you out to Google so you can see adverts for whatever you read about here following you all over the internet for the next two weeks. Neither of us gets enough benefit for that to be worthwhile.

The Driver Behind The Curtain

Truly autonomous driving is an incredibly hard problem to solve. It would be hard enough in controlled situations, but in uncontrolled ones, where other road users may or may not be respecting the rules of the road1, it’s pretty close to being impossible to achieve a perfect solution. The best we can hope for is one that is better than the current state of affairs, with distracted human drivers taking an incredible toll on life.

That is the promise of self-driving cars: get the dangerous, unpredictable humans out of the loop. Getting there, however, is tough. It turns out that the tragic death of a woman in Arizona due to a failure of an Uber experiment in autonomous driving may have been caused by the uncanny valley of partial autonomy.

Let’s take it as given that fully-autonomous (Level 5) vehicles are safer than human-driven ones. However, nobody has built one yet. What we do have are vehicles that may on occasion require human occupants to take control, and to do so with very little warning. According to the crash reports, the Uber driver in the Arizona crash had no more than six seconds’ warning of an obstacle ahead, and perhaps as little as 1.3 seconds.

Contrary to some early reports, the driver was not looking at a smartphone (although more time for our phones is one of the benefits to be expected from actual self-driving cars), but at "a touchscreen that was used to monitor the self-driving car software":

"The operator is responsible for monitoring diagnostic messages that appear on an interface in the center stack of the vehicle dash and tagging events of interest for subsequent review," the [NTSB] report said.

The Uncanny Valley

I wrote about this uncanny valley problem of autonomous vehicles before:

as long as human drivers are required as backup to self-driving tech that works most of the time, we are actually worse off than if we did not have this tech at all.

In the first known fatal accident involving self-driving tech, the driver may have ignored up to seven warnings to put his hands back on the wheel. That was an extreme case, with rumours that the driver may even have been watching a film on a laptop, but in the Arizona case, the driver may have had only between four and one seconds of warning. If you’re texting or even carrying on a conversation with other occupants of the car, four seconds to context-switch back to driving and re-acquire situational awareness is not a lot. One second? Forget it.

Uber may have made that already dangerous situation worse by limiting the software’s ability to take action autonomously when it detected an emergency condition:

the automated braking that might have prevented the death of pedestrian Elaine Herzberg had been switched off "to reduce the potential for erratic vehicle behavior." Such functions were delegated to the driver, who was simultaneously responsible for preventing accidents and monitoring the system’s performance.

In other words, to prevent the vehicle suddenly jamming on the brakes in unclear situations like the one in Arizona, where "the self-driving system software classified the pedestrian as an unknown object, as a vehicle, and then as a bicycle with varying expectations of future travel path", Uber simply opted to delegate all braking to the "safety driver" – while also requiring her to "monitor the system’s performance". This situation – distracting the driver who is also expected to take immediate (and correct) action in an emergency – could hardly have been better designed to produce the outcome we saw in Arizona.

This is exactly what I predicted in my previous post on Uber:

Along the way to full Level 5 autonomy, we must pass through an “uncanny valley" of partial autonomy, which is actually more dangerous than no autonomy at all.
Adding the desperate urgency of a company whose very survival depends on the success of this research seems like a very bad idea on the surface of it. It is all too easy to imagine Uber (or any other company, but right now it’s Uber), with only a quarter or two worth of cash in the bank, deciding to rush out self-driving tech that is 1.0 at best.
It’s said that you shouldn’t buy any 1.0 product unless you are willing to tolerate significant imperfections. Would you ride in a car operated by software with significant imperfections?
Would you cross the street in front of one?

What Next?

Uber has now ceased tests of self-driving cars in Arizona, but it is continuing the work in Pittsburgh, having already been kicked out of San Francisco after one of its self-driving cars ran a red light right in front of SFMOMA.

Despite these setbacks, it is however continuing work on its other projects, such as flying taxis.

Thats seems perfectly safe, and hardly at all likely to go horribly wrong in its own turn.

Drone crash during ski race

GIF is of a drone almost crashing into a skier during a race in Madonna di Campiglio.

  1. Such as they are, yes, I am familiar with The Invention Of Jaywalking

The VP of Nope

I have a character in my head, the VP of Nope. This is pure wish-fulfilment on my part: when everyone was in the room taking an utterly wrong and bone-headed decision, I wish there had been someone present who was sufficiently senior to just say "nnnope" and move on.

It seems I’m not the only one to feel that way, judging by the reactions to my tweet where I mentioned this:

(Scoff all you want, but those are pretty big engagement numbers for me.)

The VP of Nope has to be a VP in order not to have to get bogged down in particulars. Software engineers in particular are very susceptible to getting ideas into their heads which are great in a small context, but have all sorts of problems if you take a step back and look at them again in a wider context.

Here’s an example from my own history. I used to work for a company whose products used fat clients – as in, natively compiled applications for each supported platform. This was fine at the time, but web applications were obviously the future, and so it came to pass that a project was initiated to write thin web clients instead. I was part of the early review group, and we were all horrified to find that the developers had opted to write everything in Flex.

If you are not familiar with Adobe Flex1, it had a very brief heyday as a way to write rich web interfaces, but had the very significant drawback of running on top of Adobe’s late, unlamented Flash technology. There were several very significant problems due to that dependency:

  • Corporate IT security policies almost never allowed the Flash plugin to be installed on people’s browsers. This meant that a Flex GUI was either a complete non-starter, or required exceptions to be requested and granted for every single machine that was going to connect to the application back-end, thereby losing most of the benefits of moving away from the fat client in the first place.
  • Thin clients are supposed to be less resource-hungry on the client machine than fat clients (although of course they are much more dependant on network performance). While web browsers were indeed lighter-weight than many fat clients, especially Java-based ones, the Flash browser plugin was a notorious resource hog, nullifying or even reversing any energy savings.
  • While Apple’s iPad was not yet nearly as dominant as it is today, when it is the only serious tablet, it was still very obvious that tablets and mobile devices in general were The Future. Every company was falling over its feet to provide some sort of tablet app, but famously, Steve Jobs hated Flash, and articulated why in his open letter, Thoughts on Flash. All of Steve’s reasons were of course in themselves valid and sufficient reasons not to develop anything in Flex, or indeed to require Flash in any way, but the fact that Steve Jobs was committing to never supporting Flash on Apple devices killed Flash dead (and there was much rejoicing). Sure, it took a couple of years for the corpse to stop twitching, but the writing was on the wall.

Building any sort of strategic application in Flex after the release of that letter in April 2010 was a brain-meltingly idiotic and blinkered decision – and all of us on the early review programme said so, loudly, repeatedly, and (eventually) profanely. However, none of us had sufficient seniority to make our opinions count, and so the rough beast, its hour come round at last, slouched towards GA to be born into an uncaring world.

This Is Not A Rare Event

I have any number of examples like this one, where one group took a narrow view of a problem, unaware of or even wilfully ignoring the wider context. In this particular case, Engineering had determined that they could develop a thin web client more quickly and easily by using a piece of Adobe technology than by dealing with the (admittedly still immature) HTML5 tools available at the time. Given their internal metrics and constraints, this may even have been the right decision – but it resulted in an outcome that was so wrong as to actively damage the prospects of what had been until then perfectly viable products.

In such situations, the knock-on effects of the initial fumble are often even worse than the immediate impact, and so it was to prove in this case as well. First, enormous amounts of time, energy, and goodwill were wasted arguing back and forth, and then the whole GUI had to be re-written from scratch a second time without Flex, once it became apparent to enough people what a disaster the first rewrite was. Meanwhile, customers were continuing to use the old fat client, which was falling further and further behind the state of the art, since all of Engineering’s effort was being expended on either rewriting the GUI yet again, or strenuously defending the most recent rewrite against its critics. All of this wasted and misdirected effort was a major contributing factor to later strategic stumbles whose far-reaching consequences are still playing out now, nearly a decade later.

This is what is referred to as an omnishambles, a situation that is comprehensively messed up in every possible way – and the whole thing could have been headed off before it even began by the VP of Nope, quietly clearing their throat at the back of the room and shaking their head, once.

Their salary would be very well earned.

Photo by Vladimir Kudinov on Unsplash

  1. Originally developed by Adobe, it now seems to be staggering through an unloved half-life as an open-source project under the umbrella of the Apache foundation. Just kill it already! Kill it with fire! 

Apple Abroad

I am broadly bullish about Apple’s purchase of digital magazine subscription service Texture. I do however have concerns about Apple’s ability and willingness to deliver this service internationally. This concern is based on many past examples of Apple rolling out services to the US (and maybe UK) first, and the rest of the world only slowly, piecemeal, and according to no obvious or consistent logic.

Subscription hell is a real problem, and it creates a substantial barrier for users considering new subscriptions. Even if the financial element were removed, I have had to adopt a strict one-in, one-out policy for podcasts, because I simply don’t have enough hours in the day to listen to them all. (It doesn’t help when The Talk Show does one of its three-hour-long monster episodes, either.) Add a price component to that decision, and I’m even more reluctant to spend money on something I may not use enough to justify the cost. I would love to subscribe to the Financial Times and the Economist, but there is no way I could get through that much (excellent) writing, and they are pretty expensive subscriptions.

On the other hand, the idea of paying for one Netflix-style sub that includes a whole bunch of magazines, so that I can read what I want, seems pretty attractive on the surface. Even better if I can change the mix of consumption from one month (beach holiday) to the next (international business travel) without having to set up a whole bunch of new subs, with all the attendant friction.

Here’s the problem, though. Apple has form in releasing services in the US, and then only rolling them out internationally at a glacially slow pace. I realise that many commentators may not be aware of this issue, so let’s have a quick rundown, just off the top of my head.


Apple’s News app is still only officially available in the US, UK, and Australia. Luckily this restriction is pretty easy to fool by setting your iOS device to a region where it is supported, and there you go – the News app is now available on your home screen. Still, it seems an odd miss for what they regularly claim as a strategic service.

Siri on AppleTV

I have ranted before about the shameful lack of Siri on AppleTV, but this issue still hasn’t been resolved. Worse, the list of countries where Siri is available on AppleTV makes no sense. What concerns me, obviously, is the absence of Italy, especially when much smaller countries (the Netherlands? Norway?) are included, but there are other oddities. For instance, French is fine in France and Canada, but not in Belgium. Why? Quebec French is far more different than Belgian French. Also, Siri works just fine in way more countries and languages than are on that list, so it’s far from obvious why it’s not available on tvOS.

The worst is that it is not possible to get around this one, as the restriction is tied to the country where the user’s Apple ID is registered, and that in turn is tied inextricably to the credit card’s billing address. Short of registering a whole new credit card, if you live outside one of the blessed countries, you’re not going to be able to use the Siri remote for its intended function. Given that nobody likes that remote, and fully 20% of its button complement is dedicated to Siri, this limitation substantially detracts from the usage experience of what is already a pretty expensive device.

Apple Pay in Messages

As with Siri on tvOS, this is a weird restriction, given that Apple Pay works fine in many countries – but is not available in Messages. I could understand if this were a banking restriction, but why not enable payment in Apple Store vouchers? Given my monthly spend, I’d be happy to take the occasional bar tab in store credit, and put it towards my iCloud, Apple Music, other subscriptions, and occasional apps. But no, I’m not allowed to do that.

TV app

Returning to the TV theme, if you’re outside a fairly short list of countries, you are still using the old Video app on iOS and tvOS, not the new TV app. Given that the TV app was announced in October of 2016 and launched at the end of that year, this is a pretty long wait. It’s especially annoying if you regularly use both the iTunes Store and a local iTunes library, as those live in separate places, especially in light of the next item.

iTunes Store

Even when a service is available, that doesn’t mean it’s the same everywhere. One of the most glaring examples is that I still can’t buy TV shows through the Italian iTunes Store. I’m not quite sure why this is, unless it’s weird geographical licensing hangovers. Cable TV providers, Amazon, and Netflix all seem to have worked out licensing for simulcast with the US, though, so it is possible to solve this.

Movies are another problem, because even when they are available, sometimes (but not always!) the only audio track is the Italian dubbed version, which I do not want. Seriously, Apple – literally every DVD has multiple audio tracks; could you at least do the same with Movies in the iTunes Store?

And sometimes films or books simply aren’t available in the Italian store, but they are in the US store. It’s not a licensing issue, because Amazon carries them quite happily in both countries. A couple of times I have asked authors on Twitter whether they know what is going on, but they are just as mystified as I am.

It Works In My Country

There is a more complete list of iOS feature availability out there, and I would love if someone were able to explain the logic behind the different availability of seemingly similar functionality in certain countries – and the different lists of countries for seemingly identical features! Right now, Apple’s attitude seems to be a variation of the classic support response, “it works on my machine": “but it works in my country…".

And that’s why I worry about Apple’s supposed Texture-based revamp of Apple News: maybe it gets locked down so I can’t have it at all, or maybe it’s neutered so I can’t access the full selection of magazines, or some other annoyance. I just wish Apple would introduce an “International" region, where as long as you accept to do everything in English, they just give you full access and call it good, without making us jump through all these ridiculous hoops.

Needy Much, Facebook?

This notification was on my iPad:

A HUNDRED messages? Okay, maybe something blew up. I’ve not been looking at Facebook for a while, but I’ve been reluctant to delete my account entirely because it’s the only way I keep in touch with a whole bunch of people. Maybe something happened?

I open the app, and I’m greeted with this:

Yeah, no notifications whatsoever inside the app.

Facebook is now actively lying to get its Daily Active Users count up. Keep this sort of thing in mind when they quote such-and-such a number of users.

To Facebook, user engagement stats are life itself. If they ever start to slide seriously, their business is toast. Remember in 2016, when Facebook was sued over inflated video ad metrics? Basically, if you scrolled past a video ad in your feed, that still counted as a “view", resulting in viewer counts that were inflated by 80%.

Earlier this year, Facebook had its first loss in daily active users in the US and Canada. They are still growing elsewhere, but not without consequences, as the New York Times reports in a hard-hitting piece entitled Where Countries Are Tinderboxes and Facebook Is a Match.

At this point, I imagine anyone still working for Facebook is not nearly as forward with that fact at dinner parties or in bars, instead offering the sort of generic “yeah, I work in IT" non-answer that back-office staff at porn sites are used to giving.