Apple shouldn’t be working on a car

18. February 2015 · Categories: Apple

With Apple apparently hiring a lot of people with car backgrounds, we are seeing speculation that Apple is working on a car.

While there are areas in the car experience where Apple could help with creating a better experience, I do not believe that it makes business sense to become a car manufacturer itself.

What Apple could improve

A major sore point with modern cars is the bad integration of the additional complexity modern computers bring into the car. The problem here is that the car companies have not yet fully realized the extent to which software is a fixed cost, and so continue to have a quite modular approach while software would work best with an integrated approach: you should have one software interface, and use consistent UI controllers for every car. No steering wheel options with different controls, no choice of screens, no choice of input controllers. All of this makes the design work more difficult and so increases the risk that the result is mediocre.

Car companies can actually learn from the Apple Watch the core lesson: provide options in the physical, haptic, emotional world, but keep the software interface identical across all variants.

One of the important tasks for an improved interface on the car would be to clearly separate the driving and the standing interface: reduce what you can do while driving, essentially only allow selections from lists, and use a touchscreen with a nice keyboard for configuration while standing ( or make this available via an app on the phone)

Apple cannot do wonders

But this is also why the user interface is not really that important: as the driver, you cannot do any involved tasks without becoming so distracted that it could kill you. This severely limits the interaction you will have; when you only interact once every half hour, the interface quality matters a lot less. Of course there are a lot of people who see it as their right to text while driving, but this is so dangerous behavior that it should be banned, not encouraged.

This will only change when the car will drive itself on at least the easy route parts, and so allows the driver to do other things. But when this happens, we can as well use our phones directly, we no longer need a special interface optimized for low attention.

Electric cars are the best bet

For a new entrant, electric cars are the best bet to get started, as the greater freedom to design a car, and the newness of the technology means that they can enter a less crowded market, and it means the incumbents are likely hampered by old assumptions. But they utterly depend on their batteries: currently they are only viable in the luxury segment, where you can spend $80000 on a car. The battery cost is still eating up the advantage from having a much simpler motor, and we are some way from batteries being cost competitive with petrol. A Nissan Leaf battery will set you back $5500 plus installation, and is guaranteed for 60000 miles. This comes to 5¢/km ( 8¢ per mile). A Tesla with its larger battery currently costs twice as much for a 8 year prepaid option, and easily 6 times as much when needing replacement today. So everybody is working on getting the battery price low enough to become price competitive. It will need to reduce by a factor of 2 to 3 until we are there. And this could be hard: Not only do batteries only improve at a rate of 7% annually, and we are already able to achieve way more than half the theoretical limit in capacity, it is also that more than half of the price of the battery are for the raw materials, and you would need more than 50 times as much lithium as currently produced worldwide were all new cars suddenly electric.

Given these constraints it is rather unlikely that battery technology will become price competitive for the mass market within the next 15 years or so. And until then the market for electric cars will be a luxury market without massive subsidies, and we are not affluent enough for many people to be willing to part with a $10000 surcharge for a better car. Sales of high end cars are currently in the 1 to 2 million annually range, compared with 85 million total production.

Is Apple really working on a car?

This makes the market quite small by Apple standards: they realized a bit more than $200 per iPhone, assuming they could make $5000 per car they would need to sell more than 8 million cars to match the iPhone.

Given these economics, I doubt that Apple is actually intending to become a car company, I find it much more likely that Apple wants some car prototypes they can use to create an optimized experience, so that they can sell to car companies an outstanding car multimedia module. This is a much better strategy for Apple: they do not burden themselves with a lot of low margin manufacturing investments, they do not risk the brand for car related problems, and they concentrate fully on the part where they can deliver the best value. Considering Tim Cooks vow to double down on secrecy, I suspect that these rumors are a smoke screen to throw us off what Apple is actually doing.

Regenerative Braking

15. February 2015 · Categories: General

One of the ways to make cars more efficient is to keep the combustion engine running at optimal efficiency, and use an energy storage system to provide the actual, variable demand needed for driving. This works because the efficiency ranges from 10% to 35% depending on the actual load. (The theoretical maximum is around 50%, leaving not much room for further efficiency gains from the motor)

When we use an electrical drivetrain for this, we have a couple of losses:

Efficiency	Caused by
92%	Mechanical to electrical
60%–95%	Battery storage
92%	Electrical to mechanical
50%–80%	Total

The biggest problem is the battery¹, which gets significantly less efficient as we increase the current we are drawing from it.

We have a strong design conflict here: the kinetic energy of 2 tons moving at 226 km/h is 1 kWh, so a small battery is enough to smoothen the load, but to retrieve the 25kW we want for a smooth acceleration efficiently, we can either have a large battery or expensive capacitors. Such a battery would would need to support a lot of cycles: 1KWh are roughly 15 accelerations to city speed, so a thousand battery cycles per year seem a reasonable lower bound.

In essence, we have some expensive tech to improve city mileage to the 60 mpg (4l/100km) range, but otherwise we have come so close to the ideal, that another 50% in efficiency gains can only be had by changing the car to have radically less weight and drag.

some background on batteries can be found here ⏎

Let Kids Vote

14. February 2015 · Categories: Politics

In our modern democracies we typically give all citizens from age 18 the right to vote, until they die. Even when they have dementia, they keep the right, and their legal representative may vote for them. Predictably this has led to policies that favor the old and gets the young shafted. Therefore it is time to extend the vote to every citizen. Since 18 is a good age where you are informed enough to vote, parents should cast the vote for the younger ones.

I believe a reasonable set of rules for the kid vote would be the following:

you get the vote when you are 6 months old
This is long enough to make sure registration has happened, and prevents differences because some bureaucracies are more efficient.
only a legal representative who also has the vote itself may cast the kids vote

This is to prevent an outcry for children born to foreign parents. These can designate a citizen they trust to cast the vote for their children.
if there are two caretakers, typically mom and dad, then each gets a half vote for each kid

A bit more complicated to count, sure, but you prevent fighting between the parents how to cast the vote. Especially you prevent strong arming to twist the vote in one partners favor, or that the courts need to intervene to get to a vote when there is a stalemate which prevents the vote from happening.

The risk with this is of course that we get even more votes for people that are not working, which could increase the pressure to implement costly programs such that the tax load becomes suffocating. But I believe that we have strong enough safeguards against discrimination for that to become a big problem.

Open Source Founding

11. February 2015 · Categories: Software

We see that for open source projects, appropriate founding can be difficult to secure, even though people derive a lot of commercial benefit from it. It might be a good idea to get people selling stuff that includes an open source component to sell two variants: the normal variant and one which includes an open source contribution. This would make it much easier for people to support open source as it removes any payment friction: there is only the hassle to decide whether to support, not a new payment to be set up. And it could help that we ask for support at a moment where a customer is willing to part with money already.

iPhone Going Forward

28. January 2015 · Categories: Apple

In the earnings conference call we were given two interesting bits of information:

The number of iPhone owners who upgraded to the 6 / 6P is barely over 10%
The number of switchers from Android was significantly higher than for the last three years

Both numbers indicate that for a large number of people the phone size is more important than improvements in technical capabilities. This points to enough residual demand for a smaller iPhone that we will see a return of a small model, probably with a slightly larger screen but the same overall dimensions.

And it points to another question that is bugging me: When will the phone be good enough? The current replacement cycle of roughly 2 years is fueled by the large improvements we have seen in processing power as the mobile chips have vastly outpaced the improvements seen with PCs. But the gap is now down to one generation, the camera already delivers a remarkable quality even under indoor lighting, and almost nobody feels that their phone is too slow. Just as with the iPad ¹, the signs are there that people will start to keep their phones for longer; the question is now whether the carriers will continue to force device subsidies onto their customers, or will they switch to those 24 months device installment plans to effectively end the forced subsidies?

While a longer replacement cycle will reduce sales, it will help Apple: the longer you intend to keep your phone, the more important it becomes that it will have the quality to last, and active support over the entire lifetime. We already see this in the renaissance of the Mac: now that our computers last, the willingness to invest into a premium product increases.

The iPad still sees a lot of first time buyers, and a slow replacement cycle. This makes me believe that the main driver for phone updates, apart from size, was the camera ↩

Apple Earnings

27. January 2015 · Categories: Apple

In order to embarrass myself a bit, these are my estimates for the numbers Apple will announce today:

Units	Item
69m – 71m	iPhones
24m – 28m	iPads
5.9m – 6.3m	Macs

Earnings will take a hit from the Euro, given that the iPhone no longer sells at a nice premium in Europe, but it is modest as most of the deprecation came at the end of the quarter.

Edit: now the results are in, and I was wide off the mark: 5.5m Macs, 21.4m iPads, 74.5m iPhones. The irony is that from the U.S. mix I had an estimate of 74.3m iPhones, but it just felt too high.

Securing Web Accounts

26. January 2015 · Categories: Software

One problem we have with more and more services moving into the cloud is to provide appropriate security. The goals for the user are simple, if a bit contradictory:

her data must not be disclosed against her will,
no one else should be able to pretend to be her to others,
it should be easy to authenticate herself to the service, and
she does not want to loose anything should she forget her authentication.

The ideal solution would be a token, equipped with biometric sensors, to be your key to all your accounts, with a backup stored somewhere, which uses your DNA to authenticate you. Currently this is not feasible, and even the intermediate step of having a key storage that would do manual DNA checks is too expensive.

So we are stuck with using passwords, ungainly constructs that need almost 50 letters and/or numbers to encode a password of 256 bits strength. This can be reduced to 90 bits, or around 16 symbols, if you encode the password using a key derivation function set to 10ms, but this remains too unwieldy to remember for all but your master key.

If we could make the assumption that there is a secure storage for all our complicated passwords, then the ideal model for security would be simple: have one regular password, and a backup. The backup could only be changed with the knowledge of the current backup, and email addresses would no longer be sufficient to demonstrate your identity. One would move to this model by sending a conformation link to your email when you first enter the backup password, and publicizing the fact that you should add the back up password soon.

The closest we are currently are with two factor authentication methods that generate a recovery key, and where you can disable the confirmation channel to enforce using your recovery key. Unfortunately this does not work for Apple IDs: you always require two of account password, recovery key, and a trusted device to authenticate, and once you are authenticated you can reset the third factor. Given this weakness I’d assume that iCloud content can be accessed with with a subpoena. That the recovery key is a bit short with 14 characters or roughly 80 bits of entropy is a minor issue in comparison.

A Larger iPad?

21. January 2015 · Categories: Apple

There are currently rumors that Apple is planning to release a larger iPad, at 12.2″ or 12.9″, with an optional stylus. If we interpolate the weight from the difference between the Air and the mini, we arrive at 650g to 700g, compared to the 437g of the Air 2. So we are back at the weight of the original, which was best used sitting on a chair, with the iPad ideally lying flat on a table. While I see some usage for this “Air Plus”, I feel that for most the Air will remain a better compromise between screen size and weight. The main benefit of the 12″ size could well be that it allows enough space to effectively support multitasking, which could make it into a viable alternative to a MacBook Air.

Actually the Air Plus will closely resemble the Surface 3, which has its fans, for example for being an excellent drawing surface. Both these computers, as well as a super light MacBook Air, will compete to become the standard choice for normal computer users, and it will be interesting to see how their software will develop, and if touch based text entry will become good enough to replace the keyboards.

Apple Watch Depends on the iPhone

14. January 2015 · Categories: Apple

I believe that the killer app for the Apple Watch will be authentication. And therefore the success of it will depend on a wide adaptation of all locks to support it, exactly like Apple Pay needs to be widely accepted. In both cases the watch is easier to use, but they need the iPhone as a steppingstone to reach the critical mass to convince people to make the effort to support it.

And this means that the watch will be likely off to a relatively slow start, until in a few years the ecosystems have grown enough for it to displace wallets and keys.

What Use For The Apple Watch?

07. January 2015 · Categories: Apple

I wonder about the Apple Watch, as I find it difficult to imagine what it will be good for. It has to make do with an extremely small output area, which means that it can only compete against the iPhone in cases where the interaction is so short that the convenience of having a 2 to 3 second faster access wins out against the much better experience you could provide with the iPhone, or the information that needs to be processed is so small that it can comfortably fit onto the watch.

These activities are pretty limited:

Authentication is probably the killer app: it could replace all the keys and the credit cards you own with an inherently more secure solution. Ideally it would also replace our IDs, and it could replace all our computer logins.
“What’s up?” is also an important task to master: seeing a reminder for your next appointment, for puplic transport times, the weather, or walking/cycling directions are quite helpful. For some seeing who is calling/texting might be essential, but for me it would be just annoying, while the reverse would be brilliant, having the watch figure out whether it is good time to interrupt me and tell my callers.
Remote control for small things: controlling the TV, changing the temperature, preheating the car, …
fitness tracking is an obvious point, including replacing the iPod as your music source during exercise. A smallish, but lucrative niche to get the watch of the ground, with some nice halo effect as well.
And of course telling the time, and the other functions of the clock app.

It will be interesting to watch this evolve, as I feel there is a huge risk of providing too many small functions cluttering the interface, making access so slow that you’d rather use the phone. Also the interface itself looks quite limited: you have the touchscreen, with your thumb obscuring most of the screen, or the crown which is only one dimension and could be difficult to use when strapped on the wrist. It seems to me that having more than a dozen apps to choose from will be painfully slow. All of this points to a very restrained role the watch will be playing compared to the phone, a role which will only require quite limited hardware resources, and so see much longer upgrade cycles than the iPhone.