Archiv für den Monat April 2014

Googles latest chapter for the self-driving car: mastering city street driving

Jaywalking pedestrians. Cars lurching out of hidden driveways. Double-parked delivery trucks blocking your lane and your view. At a busy time of day, a typical city street can leave even experienced drivers sweaty-palmed and irritable. We all dream of a world in which city centers are freed of congestion from cars circling for parking (PDF) and have fewer intersections made dangerous by distracted drivers. That’s why over the last year we’ve shifted the focus of the Google self-driving car project onto mastering city street driving.

Since our last update, we’ve logged thousands of miles on the streets of our hometown of Mountain View, Calif. A mile of city driving is much more complex than a mile of freeway driving, with hundreds of different objects moving according to different rules of the road in a small area. We’ve improved our software so it can detect hundreds of distinct objects simultaneously—pedestrians, buses, a stop sign held up by a crossing guard, or a cyclist making gestures that indicate a possible turn. A self-driving vehicle can pay attention to all of these things in a way that a human physically can’t—and it never gets tired or distracted.

Here’s a video showing how our vehicle navigates some common scenarios near the Googleplex:

As it turns out, what looks chaotic and random on a city street to the human eye is actually fairly predictable to a computer. As we’ve encountered thousands of different situations, we’ve built software models of what to expect, from the likely (a car stopping at a red light) to the unlikely (blowing through it). We still have lots of problems to solve, including teaching the car to drive more streets in Mountain View before we tackle another town, but thousands of situations on city streets that would have stumped us two years ago can now be navigated autonomously.

Our vehicles have now logged nearly 700,000 autonomous miles, and with every passing mile we’re growing more optimistic that we’re heading toward an achievable goal—a vehicle that operates fully without human intervention.


BitCoins 3 Fatal Design Flaws

For those that don’t know, BitCoin is a digital currency (known as a cryptocurrency) that is not issued by governments or banks. Instead the currency uses some complicated programming to limit the amount of money that can be created. Only 21 million BitCoins will ever be created, and there is no  human decision maker who can influence that. For advocates of the currency, this is a major advantage, as it prevents the abuse of the power to create money. It is easy to see why this would be so appealing – after all, we have recently seen the damage that can happen when commercial banks have the power to create hundreds of billions of pounds in just a few years.

But there are serious problems with BitCoin. This was highlighted most recently when one of the largest exchanges MtGox, revealed that it had lost around $350 million of customer’s money after hacking incident. “Lost” in this sense doesn’t mean they made bad investments that went bad; the BitCoins were literally stolen, now exist on somebody else’s computer, and the exchange has no idea where they are.

I want to look at BitCoin’s design flaws here, so if you want to know more about the details of the currency itself, read How to Explain BitCoin to your Grandmother by Brett Scott or this Chicago Federal Reserve paper for a central bank perspective.

BitCoin is a prototype

The key point to note is that BitCoin is a prototype for what is now known as crypto currency. It was the first of its kind, an experiment designed by someone (or a some group) going by the name Satoshi Nakamoto. The original paper that outlines the proposal for a currency is well written but has the tone of a working paper – an initial proposal, not fully thought out, rather than a fully worked out master plan.

What usually happens with a new idea or product is that you try it out, find that it’s inherently flawed, and then you alter the design to make it work better. Orville and Wilbur Wright’s original plane flew just a few metres. The first bicycle, designed in 1817, involved sitting on a saddle whilst pushing the bike along by running with your feet on the floor:

First Bicycle

The fanaticism of some BitCoin enthusiasts, along with the claims that BitCoin – specifically – will become the currency of the future, is a bit like someone in 1902 insisting that in the future we’ll all be flying across the Atlantic in individual gliders that look like this:

Wright Brothers Initial Plane

Of course we won’t. The first prototype of something should be a test case, which reveals the design flaws then gets discarded in favour of something better.

I believe there are two  design flaws that are fatal for BitCoin.

Design Flaw 1. The rate of money creation

BitCoin is designed so that new BitCoins are created (‘mined’) at a predetermined and gradually decelerating speed. Around half the BitCoins that were ever designed have been created already. The money supply will increase by another 66% between now and 2025, but by then the rate of creation of new BitCoins will have slowed to a negligible amount, essentially making it a fixed money supply by 2025.


This limited supply was supposed to be a clever design feature, but actually it’s turned BitCoin into a speculative asset. The problem with this is that the amount of the currency doesn’t increase in line with the number of people using it. Economists from the Austrian school would argue that this is fine: just allow prices to fall relative to the currency. Indeed, that’s what has happened with BitCoin – each BitCoin now buys you more real “stuff” in the economy than it did in the past.

The problem comes when the limited supply affects the way people use the currency. BitCoin users who have seen the currency go from 1 BitCoin = $5 (in 2011) to 1 BitCoin = $445 (as it currently is) don’t think “Great, the price of a Coke is falling in terms of BitCoin”. Instead they think, “If I sit on the BitCoins that I own, in 1 year they might be worth 10 times more. So I won’t spend them.”

This means that BitCoin users don’t want to pay using BitCoin. In other words, they want to use BitCoin as a speculative investment, rather than as a means of payment. 

The only way to avoid this is to ensure that the supply of the currency increases in line with how much it is being used, so that the exchange of BitCoin to other currencies or of BitCoin to real goods and services is broadly stable. Without this design feature, a currency that consistently and rapidly appreciates relative to other currencies will be held as an asset rather than being used to make payments.

This is a design flaw specific to BitCoin. Other cryptocurrencies have different ways of regulating the creation of the coins.

A Note on Volatility

BitCoin is also highly volatile, having jumped from $13.36 at the beginning of 2013 to $1,124.76 in November 2013 – an 8,313% increase – and then back down to $445 today. I don’t list this as one of the currency’s design flaws as it’s largely to do with the fact that BitCoin is new, uncertain, and that the authorities aren’t quite sure how to deal with it, so volatility is a result more of the speculation about whether BitCoin will be banned or accepted, rather than the fundamental issue of the rate of money creation.

Design Flaw 2: BitCoin rewards the adopters and speculators

As with the current monetary system, BitCoin rewards the creators of the currency (the ‘miners’ who use their computers to do complex calculations to create the currency). The early adopters have become very wealthy, along with speculators who sit on their coins rather than spending them. Again, this means that those who benefit from the currency are not those who use it to trade in the real economy i.e. people who actually produce real value and make BitCoin a viable and usable currency. Instead, the benefit goes to those who sit on the currency (which prevents it functioning as a currency and makes it a speculative asset).

I would prefer to see a cryptocurrency that rewards those who use the currency as a means of payment, rather than as a speculative asset. So the more you use the currency to buy goods and services from the real economy, the more you would get rewarded with a portion of any newly created currency, whereas those who sit on their coins and use them as a speculative asset would get no share of the newly created money.

My knowledge of computer science and maths aren’t sufficient to say how this could be programmed, but it doesn’t appear to be too complicated. (There would need to be some kind of check to ensure that you don’t end up with people gaming the system, for example two users trading the currency between themselves at high speed in order to ‘earn’ more of the newly created coins.)

Design Flaw 3: BitCoin is LESS secure that national currencies

Because of the design of BitCoin, each coin should be seen as a physical unit that exists on a specific computer hard drive. In the same way that a house burglar could steal gold coins (which I’m sure you have lying around the house), a computer hacker can steal your BitCoins.

One user left his coins on a hard-drive which went to landfill, and then saw the value of the coins appreciate to hundreds of thousands of pounds. The exchange MtGox has alleged ‘lost’ 650,000 BitCoins as a result of hacking.

Anyone holding a significant amount of BitCoins is advised to transfer them to “cold storage” –  a hard drive or USB disk that is disconnected from any computer connected to the internet, and hidden somewhere secure (eg. a physical safe).

For all the arguments that BitCoin is ‘safer’ because it has no central authority, it certainly isn’t safer in practical terms.

The Way Forward

Cryptocurrencies are fascinating. We’ve made a very clear argument that the current monetary system, in which most money is created by banks when they make loans, has been a disaster. But at the same time, when states have used their power to create money, such as through QE, they’ve used it to inflate financial markets (enriching the already wealthy), rather than benefitting the real economy and ordinary people.

We’re obviously campaigning for national currencies to be created and used in the public interest, but it’s still possible that national currencies might be bypassed completely if a currency comes along that is stable, works in the interest of ordinary people, and prevents abuse of the power to create money.

Since BitCoin was established, literally hundreds of other cryptocurrencies have been designed and released. One of them already out there might have the right design features to make a stable currency that can be a real benefit to society and the economy. Cryptocurrencies have only been around for half a decade; there will be a lot of innovation over the next 5 years and it’s possible that we might see something genuinely socially useful come out of it.

But with regards to BitCoin, it’s time to let it die to make way for something better.


PS. This reddit thread by people who lost money when the MtGox exchange shut down shows how BitCoin has become a speculative asset bubble similar to the dot com bubble or any stock market bubble. There are stories of people taking their  kid’s education fund, or partner’s life savings, and investing them entirely in BitCoin. One guy even claims his friend committed suicide after investing – and losing – over $900,000 in BitCoin.

But this is not the fault of BitCoin, or a disadvantage of BitCoin. It’s more a fault of a lack of general financial literacy, in particular an ignorance of the basic point that you should never invest all of your wealth in one single asset, whether it’s BitCoin, or RBS shares (or property for that matter). Many of these people had no concept of risk management. I’m not sure we can blame them – an understanding of money and financial literacy is not something that most people acquired at school.

There’s also the desire to “get rich quick” or even just boost your income beyond what you can earn from working. Again, I’m not sure how much we can blame people for that. When the current monetary system is making it harder and harder for people to save anything after paying the mortgage and the costs of living, it’s natural to look for other ways of making money. If the guy mentioned above genuinely believed that investing in BitCoin would mean that his kids could go to university whilst avoiding being saddled with the debt, then it’s natural for him to take that option. It was the lack of understanding of money, finance or risk management that led to him making such a bad decision.


Job Search Tactics – with highest probability for success

You’ve been on the job hunt for weeks.

You’re applying immediately to every job you come across that’s remotely related to your field. You’re getting your resume in the hands of anyone you meet. You’refollowing up with hiring managers like your life depends on it.

And still? Nothing. Nada. Zilch.

Well, I’m going to tell you a little secret.

It might be you that’s the problem.

I know—before you get all ready to tussle with me, let me assure you that I realize that most people are smart and motivated and have all the best intentions when it comes to landing that next big thing. The problem is that most of us don’t have much training on how to not suck at the job search. Which means—we’re bound to make some gaffes along the way.

So let’s change that. ASAP.

Rule #1: If you’re using any of these (very-common) job search tactics, you must change course immediately.


1. Spending 100% of Your Search Time Submitting Online Applications

If trolling the job boards is your primary search tactic, you’re looking at a long road ahead. Realize that, for every job you pursue, at least one or two people are going to find an “in” at that company. And they’re going to use that “in” to get a direct introduction. Would you rather be the one with the “in,” or one of the other 20, 80, or 400 contenders coming in via the automated “clump” of applicants?

Instead: Even if you apply for the job online, the moment you hit “send,” head over to LinkedIn and see if you have a first- or second-degree connection at that company. Reach out, stat. Your goal is to be the one who gets the direct introduction.


2. Applying for Jobs (Blindly) When You’re Not an Obvious On-Paper Match

Nobody’s sitting around deducing what you might be good at or why you might make sense for any particular job. Read: When you apply online, if your resume and cover letter don’t speak to the specific needs and deliverables of the job—and spell out exactly how you are going to meet them—no applicant tracking system is going to even find it.

Instead: If you’re not an obvious match (on paper) for a job, you either need to figure out a way to make yourself one (i.e., gaining new skills, taking on volunteer opportunities or freelance work to boost your resume), or find an opportunity to explain your rationale for applying directly to a hiring manager (i.e., show how your previous work experience in your current field would translate seamlessly to this new job).


3. Expecting “I’m a Fast Learner” Will Clinch Anything for You

Unless you’re applying for a job that is, by nature, entry level, you should pretty much assume that the decision makers are on the lookout for someone who can hit the ground running. Does this mean you’ll never land a job in a new industry? Not at all. But if you’re pressed in an interview on why they should take a chance on you, don’t think for a moment the hiring manager is looking for “Because I’m a fast learner.”

Instead: Think about how the aggregate of your skills and experiences (no matter how unrelated) may actually make you a great candidate for that role. If you’re clear on why you’d be perfect for the job, it’ll be a heck of a lot easier for the decision makers to feel confident about hiring you, even if you’re a bit green.


4. Foisting Your Resume on Strangers Before You’ve Spent 10 Seconds Building Some Rapport

Would you ever walk up to a stranger and propose marriage? Of course you wouldn’t. So why do you think it’s remotely OK to find someone who works at your dream company and—before you’ve even gotten to the “How about that crazy weather?” stage of small talk—shove your resume at him, with a plea to take it on over to the manager? That’s not networking, that’s ambushing.

Instead: If you meet a contact or find a great connection on LinkedIn, look for ways to build a relationship before you ask for a job. Think: “Hi Jill, You and are both members of the Dallas Market Researchers group here on LinkedIn. I notice that you’re an analyst with Fort Knox Inc. I’m a research analyst, too, and I’ve heard great things about your firm. May I ask you just two quick questions about your role?”


5. Calling the HR Person, Recruiter, or Hiring Manager with Ridiculous Frequency

Yes, I know. The squeaky wheel gets the oil. Fortune favors the bold. Ask and ye shall receive. All sound mantras. But there is a very fine line between “confident, proactive professional” and “desperate dude who will not stop calling us.”

Instead: If you haven’t heard back about a position, follow up nicely by email after your original thank-you note: “Hi Mary, Just a quick note–you mentioned that you’d be firming up hiring plans this week. I’m very excited to help you bring the Canyon Product Line to market in 2015. No response needed, but please let me know if I can provide any additional information to aid you in your final decision.”

Job searching isn’t easy, nor can it be boiled down to a single, perfect formula. But if you eliminate the tactics that don’t work (or make you look flat-out foolish), and start replacing them with more effective alternatives?

You’ll probably start seeing progress. And progress gives you momentum. And momentum?

That’s what allows you to steamroll your way to greatness.