What Transport for London can learn about us from our mobile data
takes.jamesomalley.co.uk
takes.jamesomalley.co.uk
The other people are afraid of a totalitarian state having control of such data because obviously it would be super bad.
The first group might argue, "As this is in theory a democratic society, can we just not voluntarily elect awful people would would make a totalitarian government? And then in the meantime we can actually use this to do good, like catching actual dangerous criminals?"
The second group clearly believes the descent into totalitarianism is inevitable and just wants to hobble those future villains as much as possible (A noble idea if you believe this).
I think the first group could argue that if a truly evil dictator does come to power, they would have zero qualms about imposing the really scary surveillance stuff themselves anyway, so what we're doing now doesn't matter as much as we think.
I don't think either is automatically wrong, but I wish both sides would acknowledge that the other has a point.
Specific "tasks" are bounded by their requirements, we've seen none of that in the take-all and leave no stone unturned approach to electronics.
There is a false calm in not taking this seriously enough. The default reaction of many is to avoid hurting feelings, at the cost of western liberty.
Your choice.
Imagine instead the damage an all-powerful government could do when people you disagree with are in charge of it.
Because, with probability 1.0, they will be someday.
Sir Thomas More: “Yes! What would you do? Cut a great road through the law to get after the Devil?”
William Roper: “Yes, I'd cut down every law in England to do that!”
Sir Thomas More: “Oh? And when the last law was down, and the Devil turned 'round on you, where would you hide, Roper, the laws all being flat? This country is planted thick with laws, from coast to coast, Man's laws, not God's! And if you cut them down, and you're just the man to do it, do you really think you could stand upright in the winds that would blow then? Yes, I'd give the Devil benefit of law, for my own safety's sake!” ― Robert Bolt, A Man for All Seasons: A Play in Two Acts
Again we limit power because it will corrupt and that's how we avoid totalitarian states. "This power grab is fine because we're not a totalitarian state" is utterly and dangerously moronic.
Historically this has generally been the case. E.g. Rome becoming more authoritarian as it collapsed, or the liberal Song dynasty degenerating into authoritarianism from which China still hasn't recovered.
Investigating on rape may get you arrested, speaking up about the govt decisions will get you jailed. As opposed to, you know, actually performing those exactions.
Proposing some sort of tangible benefit is only marginally less bad than “think of the children” as a justification. But the tyranny is here today, if you’re unlucky your government will gladly trample all over your life, and if they ever do it’s unlikely many people will even take any notice, let alone care.
Ironically, when you visit Central Europe, for instance, you can use similar transport services just fine with cash. Furthermore, the city's mobile app hasn't really done anything for the tax payers; in fact, the fees have only increased. But I'm sure someone somewhere benefits from the personal data collected.
A cross-reference:
Any system that relies on a UID is trackable, it just depends how much legwork you're prepared to put in.
That still requires more resources than mobile app tracking, and it's easier for people to disguise themselves against a camera rather than mobile app tracking, so mobile app tracking is the greater current threat.
But CCTV identification will improve and it will be harder to fool, so I don't want that either.
What I man by (probably) unlawful is not that they wouldn't have the right to collect that data due to public interests. And I am sure many locals wouldn't bother sharing their data with the city if it helps to improve its and hence theirs transport system (like it probably does). But you'd have to reverse engineer the app to see whether the data goes also to ad-tech and who knows who else. That's one thing.
And also a telecom has the data due to different obligations. But they also do business with it. That's the second thing. Also the smartphone manufacturer probably has it, and that's the third interesting thing.
But, now, under the GDPR, none of them "own" the data because it is your personal data. With this point, please also stop using the term PII. This covers anything and pretty much everything; like the author, I am interested in the geolocation data but I'm sure there's a lot more interesting stuff.
Then, under the GDPR---because it is your personal data and it "belongs" (for a lack of a better term) to you---you'd want to exercise the many rights granted to you by the laws, the GDPR and the Charter. Particularly, you'd want to exercise your rights to have your personal data and transfer it to another system in a portable manner. In fact, you want it real-time through a 24h API because it is your fundamental right.
Then, assuming you have your personal data from the city, from the telecom, and from the smartphone manufacturer, you can compete with them, do AI/ML with your personal data, do crowd-sourced science with others, do open data with open source, donate it to your favorite city in Europe or even the EU, or, hell, even sell your personal data if you want to.
That's why I wrote unlawful (probably).
The law assigns personal ownership of your data, to control you.
The farcical idea that some law will allow the individual to reverse decades of progress towards electronic identity systems (even for themselves) is a joke.
It's over for the British, the culture won. You cannot oppose cultural movements anymore, because they happen behind a computer screen. Britian is dead.
You can't say no to electronics. The cat behind the curtain could be anybody. A state, a corporation, a theocracy, ect.
Xi Jin Ping (or a Jihadist) could hypothetically set all the electronic rules in the UK and nobody would know the difference.
When a cultural revolution happens behind the electronic curtain, your conscience can't know it, or oppose it.
Severing control of one's behaviour, from the individual conscience and placing it in the hands of electronics.
Brings into existence total cultural control of Britian, without the ability for individuals to oppose it or understand it.
Our world used to run on written agreements and Kantian-esque moral emotions with Duty. None of that matters or has any influence over an electronic life. All that matters is the will, so told by the unseen electronic controllers.
Britian is dead and Britian is not allowed to know it.
Oh bullshit. You can buy and use an oyster card anonymously. I do. Stop spreading FUD.
In many cities across the Western world, and even here in Eastern Europe, it has become impossible to "hide" your movements by car inside big cities because of the authorities forcing mobile parking apps down people's throats. In other words it has become almost impossible to pay for parking by cash, anonymously, to a machine installed at the side of street, you have to pay via those mobile apps (or through SMS, which is basically the same thing).
They got Oysters, they’re not just seafood
So the boy-dem can follow and see you
Oh, that's nice. How lovely that by being an O2 customer I'm able to contribute in so many different ways as a cash cow profit-making machine.
I'm sure they'll be sending me my cut of the extra profits any day now, perhaps as a discount off my bill?
What's that? They're increasing prices above inflation year-on-year? Well, I am shocked!
Not to worry - I'll show them! I'll just switch to one of their competitors! The free market in action!
What's that? Their competitors are all doing the same things? And many of their competitors actually use O2's network infrastructure?
Well, guess I'm fucked.
Yes it's creepy that O2 has pretty tight demographic data and shares some of that information with TfL, and we have to assume they're doing that for other more nefarious customers with fewer privacy restrictions. But telecoms being involved in that kind of shady business is old news.
I'm all for improving the information (given it's private) that public transit service lines and scheduling planners have access to if it means smoother transit and especially less dependence on cars.
My dream is a CityMapper style app used to offer "virtual public transit" as the means to both collect this kind of info and actively use it to make transit better.
That is, say TFL were to offer up "virtual bus routes" that may or may not coincide with the real ones, and that feed other transit hubs.
Where they coincide with regular routes, the selling point would be that they'd hire cars to handle excess demand or gaps in service so you can guarantee pickup from any bus stop within x minutes as long as you plan a journey in the app. What's great about the tube is knowing you can just show up and there's "always" a train soon. Getting something like that for busses would be fantastic, and you can offset some of the cost of guaranteeing more frequent service by actually cutting the frequency of full-size scheduled buses on less used stretches as long as you guarantee on average shorter waits when people need it.
Maybe with some limitations - e.g. if not a "real" bus you might be shunted onto regular services at the nearest transit hub.
Where they don't coincide with regular routes, you use first/last mile data from the planning to try to guess at new routes to offer, and so when people book, they might get offered to go to a nearer pickup point that isn't a bus stop but where you're guaranteed to be picked up. There are some services sort-of like this, but usually as a separate thing, and it'd be great if it's fully integrated as a means to plug holes in service and test route changes.
I think there's a lot of chicken and egg going on with transit where likely viable routes aren't available because testing them out is difficult and expensive and gathering data on which routes might be useful likewise. If you can fill them with smaller, cheaper cars and dynamically test, I think it'd have the potential to get a lot more people to try to stick with public transport. Especially if it also offer "fallback" or combined options involving booking a taxi for all or part of it when even the virtual routes can cut it.
CityMapper is great for the planning in London, but that ability to offer "virtual routes" at scale is what would really change things.
for the most part mapping services like Google Maps often provide the public transport times, bike share, ride share, etc. within the same page and query where I live.
An example near me is that the main bus routes follow one of 3 two main roads between two major stations near me. If you're in the middle or want to cross between those two, it's usually faster to walk. Especially as the buses are often full during rush hour. If you were able to both offload a portion and offer a more attractive alternative for those living in the middle, even just between those two stations, it'd plug a hole. That route would break even if you managed to put on average 4 passengers per minibus paying the same as the bus fare based on Uber and minicab prices. Given bus routes are subsidised, that's a pretty decent starting point even if you drive increased demand to other parts of the network.
On the other end of the scale, there are places a bit further out to me where the buses run every half an hour, if you're lucky, and winds it's way across a large area to cover a much as possible with as few buses as possible where the demand is at the same time so low that you could meet it far better with near-on-demand minibuses that'd skip.
I think you could do a lot here without actually increasing budgets, by scaling down services that are currently kept going because they're seen as necessary, but replacing them with something more dynamic.
> for the most part mapping services like Google Maps often provide the public transport times, bike share, ride share, etc. within the same page and query where I live.
Google Maps is very sub-optimal in London compared to the Citymapper model which is to hand-hold you through the entire journey (tell you which platform to go to, vibrate when you're close to a stop etc.), but that's not really the point. The point would be to be able to plan your journey in a way that ensured the data could feed straight into bookings as well as into end to end demand forecasting of location pairs instead of just which journeys people take when they don't have any other alternative.
The title: What Transport for London can learn about us from our mobile data
I don't see what about the title is so emotionally charged. It only asks the question what London transport can learn from mobile data and the article answers precisely that.
It definitely makes me think though that GCHQ certainly have the deanonymised stream of data from all cellular providers - only for use in “matters of national security” of course
It generally gets you where you need to go, but often via roundabout routes. Both buses and trains are often delayed and sometimes don't arrive at all, so you best not be in a hurry. Trains are so expensive that a car ride is usually more cost effective, even with the congestion fees.
Definitely don't try and use public transport anywhere else in England then!
> TfL is unique compared to other transport authorities in the world in that it is mostly dependent on fares for its income. This is largely down to the fact that the Government took the decision, in 2015, to remove TfL's £700 million annual operating grant from April 2018 onwards... The Government’s punitive approach to TfL’s financial future not only jeopardises London’s economic recovery, but that of the country. Ministers know that investment in London’s transport system creates jobs and generates dividends for the rest of the country
[1] https://www.london.gov.uk/tfl/tfl-funding-settlement-another...
Public transport is for “poor” people. Ergo, public transport isn’t important
Plus at the moment they’re trying their best to leave the country in a state that for which they can spend years heavily criticising the next government (which will likely be the opposition)
They’re reallocating funds to improve motoring in a vote buying spree too
Then the fares crashed and needed bailing out by the central government who don't like public transport or the London mayor. The government are now much more car boosting than they were so anything that decreases the voters' perceived need to use cars or restricts their use is cut back.
Myth. By nature of the routes they serve, trains and metro systems tend to be used disproportionately by office workers with above-average incomes. In most western countries, subsidies for trains and metros are highly regressive. Buses do tend to be used by people on below-average incomes, tend to deliver disproportionate social benefits and are richly deserving of subsidy.
Even if this was all there was to public transit, if you shifted all of the train and metro use onto the roads - even in buses - in a city like London it'd mean traffic would grind to a total halt.
Those who don't use these systems still benefit from how they shift demand away from the roads.
> Buses do tend to be used by people on below-average incomes, tend to deliver disproportionate social benefits and are richly deserving of subsidy.
So in other words public transit is a universal benefit. Transit includes buses.
You've delivered a good argument for considering which types of transit we subsidise at what level, not against the notion that public transit is a universal benefit.
At the same time I think you're also missing something: In London at least, a not insignificant reason why the bus is used more by lower paid people is not down to routes but cost. That may not account for all of the difference, but even with the hub and spoke nature of London transport that does favour those with higher paid jobs in the centre, a large portion of routes between places not ideally placed for it will still be faster by train.
As such, the problem often isn't subsidies for trains and the underground, but that the subsidies are too low to make these services financially viable for lower paid people.
E.g. I live in zone 5. I know people who commute by bus from here to zone 1 because a monthly travelcard costs 267.30 for zones 1-5, while a monthly bus and tram pass costs 94.90.
The effect is often a significant "time tax" on poorer people. One woman I know used to spend two hours each way on a commute that'd take her via zone 1 on the train but only take one hour each way because she couldn't afford the travel card. There are many people in that position.
So by all means subsidise buses more, but that is not a reason to not consider that higher subsidies of trains and metro systems would also significantly benefit poorer people.
And that’ll get you infinite journeys from zone 1-6, (£14.90 daily cap for that distance with oyster).
That same journey by car, if we count fuel (average of 52mpg and 162p per litre of diesel) - it’ll cost you ~£5 each way from Epping to Tottenham Court Road.
So £25/potentially £37.50 with ULEZ vs £14.90. And then I hope you have parking space paid for you in zone 1…
FWIW, TfL has been dealing with funding crises for years. I believe it is far less subsidised than public transit systems in other major cities. There was talk a while ago about TfL officially going into a state of "managed decline", where the priority would be attempting to mitigate the impact of funding shortfalls on service levels. Like, that was a phrase used publicly at the highest levels. I don't know if they ever technically activated that scenario but they have been on the brink financially at least since Covid.
As for the specific criticisms, I mostly use the tube which is probably the best of the TfL services. I am on the Northern line which is super frequent and reliable so I can't complain. The hub and spoke model does mean it's a pain getting from one non-central area to another non-central area. Buses are better for that but they do have issues with delays in my experience.
I moved here from a city with notoriously awful public transit so to me it's great, but I guess there are probably cities in continental Europe with far superior systems. Honestly though, it's 2023 and I still need to buy paper tickets for the Paris metro...
4 out of 10 of the proposed routes have come online by virtue of renaming previously existing express routes.
There's a whole lot of existing track that could serve as the basis to make it viable to commute without going via central London between a belt of commuter towns outside of, or on the boundaries of, London that are in far better positions to handle "cheap growth".
My pet example, largely because I live in Croydon and a past employer wanted to relocate the office from the centre to Guildford, is Guildford to Croydon. Both have fairly large centres and scope for significant increases in density near their transit hubs.
There are several viable routes direct between them, but the shortest, potentially fastest routes along existing rails only have a smattering of trains that go the whole distance, as part of trains providing some basic level of service to intermediate stations.
The current fastest routes you can depend on for commutes are both significant detours North to Clapham or South to Redhill.
The most direct routes via Epsom and Leatherhead are few enough or stop enough that they are almost never worth considering.
Committing to maintaining commuter-level direct services between bigger towns that already has track connecting them in an outer loop, or arc at least, would do a lot to make it more attractive for companies to consider moving out, and pulling demand out with it and offload the centre.
My boss mentioned earlier had to abandon his plan to move the company in part because a large portion of his staff would walk over the extra commute - for me it'd increase the average commute from an already tedious one hour to 1h30 -, but for many of us it wouldn't take much longer if only there were direct connections instead of the hub-spoke.
The speed of the services also matter as the Clapham Junction to Guilford direct train is significantly than the Clapham Junction to Guildford via Epsom, due both to track speed and number of stops.
The extra milage here is cut by the speed of the longer part and the CJ-G runs on 4 track rail so allowing it to miss stations whilst the other lines have to stop everywhere.
Much better examples of hub-spoke issues are north of London between East and West Coast main lines.
I can find no indications that the lines in question are anywhere near capacity. Even if they were, adding new track to create such an arc around London would be far cheaper than continuing the crazy efforts add capacity in the centre.
> also your case would not produce a viable service as not many would use it.
This is entirely backwards thinking. It's how we've ended up in this mess in the first place. You need to guarantee commuter level service as a means to drive economic activity and demand for housing, then you can expect use. That requires a 20+ year commitment to get people to take the routes into consideration when moving or setting up a business.
This, much more so than the specific path, was the crux of my argument: It's stupid to induce further demand in the centre rather than investing in inducing demand in places where further expansion is far cheaper by reacting to congestion instead of planning ahead. Don't fix congestion where it occurs when it is expensive to do so. Instead route around it and make it attractive to avoid.
> The speed of the services also matter as the Clapham Junction to Guilford direct train is significantly than the Clapham Junction to Guildford via Epsom.
It sounds like you've misunderstood me. Clapham Junction to Guildford via Epsom is certainly not what I was arguing for. That'd entirely defeat the point of connecting towns outside or on the edges of London while bypassing the overloaded central routes and stations. If you're already on the South Western Mainline I fully agree it makes no sense to go off it before you can get directly onto the line to Guildford.
The point of what I suggested was to provide an example of avoiding a detour in towards the centre entirely in the first place. Hence the example of avoiding Clapham entirely when coming from Croydon, as one example of a way of improving connections between towns big enough to form viable commuter paths given time and opportunity to for people to take it into account when moving.
The key point of that example is that the travel time from West Croydon to Epsom + travel time from Epsom to Guildford is already competitive with the typical (but not the fastest) travel with interchange at Clapham Junction, but the trains run rarely enough on those routes that the change at Epsom will almost always make it significantly slower not due to the track or the stops, but due to the lack of frequent enough and direct trains.
A direct service would cut the average time via that route significantly enough to be competitive on time even without any track improvements even if still stopping at every stop. Even just that would be an improvement by removing the annoyance of a change.
> due both to track speed and number of stops.
As pointed out, on the line I actually talked about, track speed is sufficient, though not great. But gradually upgrading tracks on a relatively lightly loaded line is also a lot easier and cheaper than building new lines - the point was not to avoid the need for upgrades, but that it is possible to start a service without having to build entirely new lines.
> The extra milage here is cut by the speed of the longer part and the CJ-G runs on 4 track rail so allowing it to miss stations whilst the other lines have to stop everywhere.
There's no inherent need to stop everywhere even if fast trains and stopping services runs on the same tracks. Plenty of places run express trains and local trains on two tracks with careful timetabling. E.g. ensuring stopping services wait at stations if necessary. I grew up with that. Yes, it has challenges and causes delays at times, and yes it sometimes makes the stopping services even slower, but it works up until a service frequency well above the current one on the lines in question, at least for parts of the route. As a starting point for services you'e pointed out not many would use at the outset it's perfectly fine.
> Much better examples of hub-spoke issues are north of London between East and West Coast main lines.
Great, but that's entirely besides the point, which was not the specific route, but the overall idea of connecting towns on the edges of London together to reduce the need to go further in.
>> The problem for your case is that there are not enough train paths to add a new train, >I can find no indications that the lines in question are anywhere near capacity. Even if they were, adding new track to create such an arc around London would be far cheaper than continuing the crazy efforts add capacity in the centre.
The limit is where does the train go on a Guilford to Croydon run - there is no room in Croydon for it to turn round - see London Overground neededing to create a third line in West Croydon.
>> The speed of the services also matter as the Clapham Junction to Guilford direct train is significantly than the Clapham Junction to Guildford via Epsom.
> It sounds like you've misunderstood me. Clapham Junction to Guildford via Epsom is certainly not what I was arguing for. That'd entirely defeat the point of connecting towns outside or on the edges of London while bypassing the overloaded central routes and stations. If you're already on the South Western Mainline I fully agree it makes no sense to go off it before you can get directly onto the line to Guildford.
No you have misunderstood me - because the CJ->G journey is over straighter tracks it can run faster than on slow lines elsewhere and mis out stops. The main point is that time on a railway is not only determined by track miles but also type of track and other trains on the tracks.
>> due both to track speed and number of stops.
>As pointed out, on the line I actually talked about, track speed is sufficient, though not great. But gradually upgrading tracks on a relatively lightly loaded line is also a lot easier and cheaper than building new lines - the point was not to avoid the need for upgrades, but that it is possible to start a service without having to build entirely new lines.
For upgrading costs see both East coast mainline and Great Western - which is why HS2 is not an upgrade of the West Coast Mainline.
The speed between CJ->G is not really due to better tracks it is due to having 4 tracks so fast trains can overtake slow ones. Now there is no way to put 4 tracks in Croydon to Epsom and I suspect to Effingham Junction there are houses backing on to the line, many bridges etc.
Timing on railways in not just down to direct lines but you have to plan Southern Region as a whole.
I also think Croydon to Guildford would never be sustainable - there are many other places that need upgrading first. Especially as the link is already reasonable - the speed up would not be sufficient.
Trains already follow the entire route, which is sufficient proof that this problem is solvable. Croydon doesn't need to be the first/final stop - that's entirely besides the point. Indeed, this specific journey doesn't need to be viable - that too is entirely besides the point.
> No you have misunderstood me - because the CJ->G journey is over straighter tracks it can run faster than on slow lines elsewhere and mis out stops. The main point is that time on a railway is not only determined by track miles but also type of track and other trains on the tracks.
That's fine but irrelevant. As I've pointed out, journey times for the individual legs of the route are fast enough to make it viable. And there is no need for it to be able to run faster to miss out stops. It'd be a nice bonus to be able to, but it's not needed for the argument I've made at all.
> I also think Croydon to Guildford would never be sustainable - there are many other places that need upgrading first.
It certainly isn't sustainable. That is exactly the point I was making: Planning based on what is sustainable is entirely backwards thinking when optimising transport. You need to plan based on what you can make sustainable as part of driving larger changes. And again, the specific journey is irrelevant. Not the point.
> Especially as the link is already reasonable - the speed up would not be sufficient.
... but this is just not true. The link is not reasonable. That was my starting point: The existing connection is enough of a hassle that it stopped my former boss moving his company to Guildford because half his staff would leave if he did. I was one of the one who would have quit over that. Had there been direct trains from Croydon that were were regular enough and direct I'd have been fine. Even without trains going any faster or skipping any stations based on existing travel times on existing trains between Croydon and Epsom and Epsom and Guildford.
ALso in South London there are orbital rail and tram routes so your case does not seem to happen in South London.
The biggest problem with transport in the UK in general tends to be cost more than anything.
E.g. there are regular examples of people doing stupid shit like buying a car for a single journey for less than the train ticket, or some student who took a detour between Sheffield and Essex by plane via Berlin[1] because it was cheaper than the direct train even factoring in travel to/from the airport (though marginally enough that presumably his motivation was more wanting to actually go and because it was ridiculous than the few pounds he saved).
While TfL is not that extreme, they're still not cheap either. I think the biggest problem with TfL pricing is that the discounts for passes is very minimal compared to many other places.
> Honestly though, it's 2023 and I still need to buy paper tickets for the Paris metro...
It's been a few years since I went to Paris, but the contactless Navigo Easy pass was introduced in 2019. Are there places in Paris you still can't use it?
[1] https://www.independent.co.uk/travel/news-and-advice/teenage...
* Paris
* Amsterdam
* Brussels
* Berlin
* Vienna
* Budapest
And probably some others I'm missing. TfL isn't as special as some people think, but it's still pretty damn good.
At some point, it will be trivial to force people to act against their conscience, en masse, with electronic precision.
Computers don't take no for an answer. They are the horrifying interface to tyranny.