Analytics for trains

Like Jeremy Corbyn(sic), I was consigned to the floor of a Cross Country train for 3.5 hours today. For a sizable fee of £175, I was locomoted between Reading and Manchester in some discomfort. And it really gets you thinking.

How can a fixed capacity train service be sold to overcapacity? The demand is of course peaky, very peaky. But then so are networks of parcel distribution (at Christmas), mobile phones (on new years eve) and swanky bars (on a Friday night).

Despite the advanced analytics on offer today, these capacity-critical services all seem to apply the same rule... "Deliver the worst service we can get away with".

Mostly this is driven by good business sense, but also by pure supply and demand. Train Operating Companies (TOCs) are contracted under the 'Cap and collar' model whereby if they lose money the government bails out a good proportion of their losses or, if they .ake money, takes a cut of their profits. Profits or losses vary widely across the rail network. In 2015 South Eastern Trains amassed £803 million in profit. Paying their shareholders a dividend of £25 million. And this was achieved with an average seat occupancy of 26%. Imagine how their profits would look if they could bump that up to 30% of even a third.

The problem here is that the contracts TOCs sign include a minimum service provision, which forces them to run off-peak services at a loss while trying to make up for that by stuffing peak time services to the gunwales. This means misery for daily commuters and the occasional traveller-on-a-schedule.

But is there a better way? Erm... have you been on a train in Germany? German trains run like clockwork and everyone has a seat. And, the cost of this experience ? Well.. that's where the problem lies.

Germany spends €17bn on rail subsidies per year compared with €4.5bn in the UK. That extra €12.5bn is equivalent to the UK government spending across roads, buses, trams and railways combined.

If we want a pleasant rail service we either have to pay for it, at the expense of the NHS, social security, defense, education or pensions, or we'll have to accept a service which more closely matches supply with demand.

Very basic analytics could allow us to optimise the off-peak services. For example, instead of providing time-boxed services (e.g. hourly), we should work on functional services (e.g. demand-driven). This might mean a train service between towns and cities twice or thrice in a day.  We could gradually reduce the public expectation of trains running when we want throughout the day, to having a reduced but reliable and comfortable service in the mornings, middays and evenings.

Sure, people would have to organise their activities differently to adjust to this new paradigm. But that, to me, is progress and good money sense.

So let's tear up the TOC contracts and deliver some societal change in the form of a better rail service for all.

The search for Smarts

I'm in the market for a smart watch. After 4 years of loyal service, my Garmin Fenix 2 has expired, leaving me with only two other ways to track my running and cycling activities.

But rather than mourn the loss of a dear technological friend, I feel it's only right to commence the most comprehensive evaluation of all possible replacement devices.

Which watch? 

I run, cycle, work out, and occasionally swim. So speed and distance measurements are important, as is heart rate monitoring. I couldn't give two hoots about stride length, cadence, vertical oscillation, VO2 Max or recovery periods. I'm not Mo Farrah. Oh, but it has to Bluetooth to my  phone, obviously.

So I set off on my epic campaign to review all the information available to a would-be watch buyer. And, wow! What a choice there is. Even if I stick with good 'ol Garmin, there are 28 different models to choose from, not counting the different colour schemes.

Nearly all of them do what I want. And they range from the £849.99 Fenix Chronos, down to the £89.99 Vivofit. Garmin's UK website doesn't allow filtering by functionality, only by 'activity', and not by more than one activity at a time. I don't want to be pigeon-holed into just being a runner, or just doing 'tactical training', whatever that is.

The only other option is to compare devices. There's a limit of 4 devices on their comparison tool which, given I haven't narrowed my selection down from 28 yet, seems inadequate.

I'm done with the Garmin website. Next up is the excellent www.dcrainmaker.com. DC's website gives phenomenal in-depth reviews of most of the latest sporty watchy gear. He leaves no stone un-turned in the pursuit of feature and performance testing. Both the watch and the software get a thorough going over, and useful comparisons against other devices and his annual favorites. Some of the reviews run to 'dissertation' length. And honestly, I don't have the time!

I checked out his '2016 Best Buys' list, and found the Garmin Fenix 3 topping the list, alongside the Garmin 920XT. Looks like a simple choice... I liked the Fenix 2, why not buy the Fenix 3 and get on with my life?

BUT WAIT....

What about a proper smart watch... you know, like the Apple Watch, but not so w[SENSORED]ky ?

Since I bought my trusty Fenix 2, the market has exploded with smart watches from all the phone companies, most of the watch companies, and a host of other people too... Mont Blanc eStrap-on? Huh?!

And do I want a smart watch, or just a smart strap? Or connected trainers? Or just use my phone - which does everything a smartwatch does better.

I'm going for a lie down. This is all too much. I pity the general public in making these kinds of decisions, except for the ones who like making comparison spreadsheets, and the ones who just buy whatever's on offer in Argos. For the rest of us, it's time to embrace the impending shoppers' remorse, and 'chose-the-wrong-device' anxiety.

Now... where's my spreadsheet from 2012 ?

IOT or IORT (Internet Of Rubbish Things) ?

The Internet Of Things (IOT) is coming! Along with 5G, SDN & NFV, the IOT has been driving industry article column inches for a couple of years now. And what do we have to show for this IOT hype? To most people's minds... not a lot.

IOT breaks down into a few areas: wearables, automotive, home, medical, industrial, transport. And reports suggest anywhere from 10 to 50 billion 'things' will be embedded with the ability to connect to a network. That's 5 to 10 times more devices than there are smartphones on the planet.

A year ago I thought those claims were wildly off the mark. No one really needs a bluetooth toothbrush, and all those internet-connected fridges simply haven't happened. So I lay back and imagined that the IOT was another industry fad, consigned to promise much but deliver little. This isn't the IOT ... the Internet Of Things, it is the IORT ... the Internet Of Rubbish Things.

In the history of our telecoms world, there is an enjoyable tendency to get over-excited about something new, then forget about it and focus on the next 'something new'. And then, a few years later, let the world catch up and hey-presto... that promised 'something new' finally happens, to dramatically change the world, but without too much fan-fare

I was at the Mobile World Congress in 2007 the year after iPhone launched, and remember the buzz around 'Mobile TV'. I remember the quirky Japanese phones with screens that rotated on their unwieldy soap bar handsets to show video in landscape mode.

There was an expectation that, with the advent of 3G, video and TV services would launch across the mobile carrier world and be the next big growth driver for data usage. Vodafone duly launched video-calling and showed at least one Cricket test series on their free trial of mobile TV. 3 Months later the trial had ended with no commercial proposition resulting. 2007 happened to be the year that Netflix began it's streaming service and began it's move away from posting DVDs to the public.

Cut forward 9 years, through 4G, fibre-optic, wifi-n, and we arrived at the mobile-video-enabled Rio 2016 Olympic Games. Almost half of all content streams from Rio 2016 were made to mobile devices.

This dramatic change of behaviour is what we expected in 2008. It just took a little while to get there. To go from a failed mobile TV pilot to this point, where mobile video is something that not-just-millenials have wholeheartedly adopted, indicates what the future may hold for IOT.

IOT is best when the customers don't notice it's there. "Any sufficiently advanced technology is indistinguishable from magic." So said Arthur C Clarke. And this is where IOT is the magic.

These tiny embedded sensors that track your movement and heart beat, and share them with your smartphone. The SatNav in your car which shows live traffic information. The thermostat that warms your house when it sees that you're coming home. All happening magically without us having to think about it.

So the IOT revolution will happen not with a fan-fare, but with a billion high-frequency whispers. And before you know it, everything will be connected, and the 'everythings' will be making your life that little bit better.

The challenge of Authentic PR

A recent Telecom.com podcast featured Paul Nolan from CC Group, who quote-un-quoted "Authentic" as a recent buzzword too often used by PR agencies. Equating it to IOT, future-proof and 5G-ready - the current favorite buzzwords of the Telecoms business.

This made for a fascinating comparison between industries. In the Telecoms world we have no end to the deluge of latest innovations and the buzzwords that accompany them. In PR, it seems the diatribe is also alive and kicking. 

I quite enjoyed seeing that the world of PR has just as many infuriatingly meaningless terms as Telecoms. Have we had enough hotly-anticipated, paradigm shifting, market-optimized, horizontal, well-positioned, marquee solutions...?  Probably not.

Trevor McDonald, our favorite barrier-breaking newsreader, would be turning in his grave over the lack of plain English and specificity displayed by so much of the Tech company PR content.

But, to my mind, "Authentic" is what we should be shooting for. We should all be authentic in our interactions with other humans. As Patti Azzarello suggests in her excellent managers book 'Rise', being authentic is fundamental to developing effective relationships.

So let's all agree to put a little more authenticity into our communications, whether we're lowly middle-managers or high-flying PR execs. We'll all be better off, in time, and better informed, in information.

... and if you are in the business of writing copy for yourself or your business, here are some great guidelines taken from Apple's approach to clear, compelling messaging.

Who do you trust ? Tesla autopilot .. or .. the distracted teenager behind you ?

Following on from today's post on driver-less cars, I remembered watching a great TED talk by Jennifer Healey at Intel.

She points out that a human driver has two eyes with which to scan the surroundings, and those eyes are not designed for the task. For starters, they only see clearly in one direction. To check behind, the driver stops looking where he/she is going. Not looking where you are going whilst driving at motorway speeds surrounded by other large metal objects is frequently fatal.

So what if cars could talk to each other and say things like... "I'm slowing down rapidly" or... "I'm changing lane". And other cars could listen and respond immediately and in a safe manner. That sounds like a better plan that relying on my two eyes and poor reaction time, or even worse, the concentration and lack of experience of the distracted teenager in the car behind me.

And Jennifer points out that only cars within say 100 meters of each other would need to talk, so they wouldn't require a cellular connection, just a white-space radio messaging function.

Today's more advanced cars use arrays of sensors to detect their environment and they do a pretty good job of looking at the car in front or behind. Top-end cars already have traffic sensing cruise control. But each car is an island on its own, detecting it's own necessary actions. With car-to-car communication, a car two or three cars behind on the road would already be taking action before the car immediately in front of it had reacted, cancelling out the 'traffic wave effect'. 

It's simple, it's intelligent, and because it doesn't require expensive arrays of sensors in every car... it would be cheaper and more accessible to lower-end vehicles.

Let's hope the auto industry take a pragmatic view of these possibilities and deliver a solution in the not-too-distant future.

References:
Jennifer Healey - If cars could talk, accidents might be avoidable
CDC - Teen Drivers: Get the Facts
New Scientist - Shockwave Traffic Jams

Driverless cars? UK maybe, India no hope.

The buzz around driver-less cars is hotting up. Suddenly Google are being made to seem slow-to-market by Tesla and Uber. It seems the machine-driven car is here already and aside from a few years of human-aided operation, 'the machines' will have a big impact on our modes of transport.

I never understood why trains needed drivers in the first place. They run on tracks, the junctions are controlled by points operated by machines, and there is minimal traffic to worry about. I suspect it may be customer fear and major reluctance from laggard trade unions that have restricted the automation of what seems to be a much simpler transport problem. At least we are seeing a gradual introduction on newer lines (Docklands Light Railway) and some older lines (Victoria line), albeit with human 'train captains' still on board to handle emergencies.

The easier train lines to automate are new ones, which don't suffer from the same key problem as driver-less cars do. The problem is humans. Wherever automation and humans need to coexist there is more complexity and a much higher degree of risk.

This is the key problem when putting driverless cars on the roads. It's not the computers that are the problem but the illogical, unpredictable, and frankly hazardous humans.

On 30th June 2016 Tesla reported that it had made the NHTSA (National Highway & Traffic Safety Association) aware of a fatal accident. A Tesla Model-S collided with a tractor trailer on a divided highway as the tractor was crossing perpendicular to the road. The Tesla autopilot failed to identify the trailers white side panels against the sky and did not detect the chassis of the trailer. This tragic event left a family without a father and served as a stark warning to the driver-less car industry.

Even in countries with established and, by-and-large, good adherence to road safety and conduct rules, there are between 5 (UK) and 12 (USA) deaths per 100,000 vehicles. that works out to 3 (UK) and 7 (USA) fatalities per billion km's driven. The Tesla fatality (US) was the first in over 200 million km's, indicating a better safety level than that of human-driven cars. 

In India 130 people die per 100,000 vehicles. In Bangalore, where I find myself this week, 2 people die on the roads every day. I wish I'd have known this figure before I rented a motorcycle for the day. I might not have been so keen.

The reasons for this alarmingly high death toll are numerous. There is appalling traffic congestion, there are overcrowded and poorly maintained pavements, there are no traffic lanes, huge numbers of scooters and small motorcycles. The clearance between vehicles is often inches rather than feet. Nervous drivers go nowhere, and traffic signals are like the pirates' code - more guidelines than rules. Add into the mix a few thousand tuk-tuk taxis, poorly maintained vehicles and roads, animals running loose in the road, and you need a vastly more complicated compute model for driver-less vehicles.

If the Google/Uber/Tesla software wants a real test, they should be (and probably are) testing here in India. If they can drive us safely in Bangalore, Mumbai and Delhi, then Los Angeles and London will be child's play for 'the machines' to navigate safely.

Net Neutrality vs Utilities

We often hear that Internet Access is now considered a basic human right, and should be thought of in the same way as other utilities such as water and electricity. Dumbing down the argument to this level is fine for the tabloid press, but let's take a more detailed look, and decide for ourselves.

What is net neutrality? It is the principle that ISPs and governments should treat all data the same, not discriminating or charging differentially by any one of many factors e.g. type of content, which website is being visited, what device is using it and for what purpose.

Most ISPs and mobile carriers have the ability to throttle (i.e. reduce) speed and bandwidth in their network, in order to provide reliable service, even-out peaks in traffic and provide a better service. If you use a 4G mobile phone, your voice calls are converted into packets of data and will more than likely have used the internet to send those packets to another 4G mobile user that you are calling. And yes, those packets will have had priority within the carrier network, to make sure that the call doesn't break-up, and provides real-time two way conversation.

What carriers are not allowed to do is to treat Voice IP traffic from competing companies different to their own traffic. They are also not allowed to throttle traffic that they don't like the look of. Peer-2-peer file sharing networks which carry a vast amount of illegally shared content can't be throttled in preference to perceived 'legitimate' traffic.

Without net neutrality, ISPs could deliver tiered services, to content providers in addition to those tiers offered to the consumers. So a provide a real-time video e-healthcare services (think remote midwifery), could secure a high quality video feed, whilst an e-book download service could secure a cheaper less performant access because they don't need ultra-low latency and high bandwidth.

Why net neutrality is GOOD:

1. Equal access to the internet is a human right. The internet has been a mostly free and open place for all of humanity. Charging "different rates to different website" would destroy that principle

2. It is good for competition and innovation. It protects smaller companies from the leverage of larger companies who could pay more to deliver content and services.

3. It fosters fairness in ISPs, who currently are paid by consumers to provide access. Without net neutrality, the ISPs could charge content providers for priority delivery, limiting the consumers' overall level of service.

Why net neutrality is BAD:

1. If ISPs can charge more from content providers, they could improve the network faster and provide better overall service, depending on the traffic being carried.

2. Other parts of the network already deliver tiered services across the internet. Amazon charges more for faster processing and better storage, which already impact on the consumers' experiences.

3. Many consumers can choose between Broadband providers with competing networks. In UK, between BT hosted services and Virgin's cable network. If consumers don't like how their ISP is providing differentiated service, they can opt to move to a competitor.

Conclusion

Internet access is 100% not like a utility. No one is offering faster, better electricity. It's all you can eat electric until a circuit breaker blows. You can fill your bath over and over, and nobody has better or more water to offer. This is an infrastructural limitation. There would be little point laying two sets of water pipe, or two electric lines into a house.

However, the analogy breaks down a little when you consider that it's perfectly possible to buy extra water to drink when the need arises, or special water for a special occasion (Perrier anyone?)

Internet access is the same. We can choose how to access the internet, be it at home, work or mobile. We already pay differentiated access fees. But would we suffer if say Netflix could pay more to TalkTalk to secure a buffer-free experience for its users? 

I suspect that we would. Corporate decision making is seldom driven by what's best for the customer. If an edge can be gained over a competitor, then a large company typically will do what it takes to get ahead. Those competitive decisions would have the potential to limit choice, increase cost and stifle innovation in the services that we consume.

A policy for a government not to pursue net neutrality would have to be regulated by even more complex limitations. Sticking with net neutrality seems to be one control mechanism that protects the consumer, and simplifies the governance of internet access to all.

Reference material:

https://en.wikipedia.org/wiki/Net_neutrality

http://plugandplaytechcenter.com/2014/05/16/four-and-against-the-net-neutrality-debate/

http://www.callforwarding.com/blog/net-neutrality-voip-need-know/ - Caitlin Burge