Medusis tech blog

1985: Polaroid wins its suit against Kodak

A descendant of Jewish immigrants joins Harvard; there, he thinks about a new technology, develops it, and drops out of college to found a company to exploit it. Mark Zuckerberg? No, Edwin H. Land, inventor of the polarized film and founder of Polaroid.

Land-Wheelwright (named after its two founders) was created in 1932 and renamed Polaroid in 1937. The first Polaroid camera was offered to the public in 1948, and propelled the runaway success of the Polaroid corporation.

Polaroid had invented a new product category: "instant photography", and maintained leadership of that category for a long time; in fact, it was the sole contender.

But then competitors entered the market, esp. Kodak who by the mid-seventies, had gained nearly 30 percent of the instant photography market.

Then what happened? In 1976, Polaroid filed suit against Kodak, alleging patent infringement; nine years later, it won, driving Kodak out of the market.

And of course, in 2001 Polaroid filed for bankruptcy and has since been a shadow of itself (then a shadow of a shadow).

2012: Facebook buys Instagram for $1 billion

Why did Facebook pay so much for a two-year-old startup with just 13 employees and zero revenues? Instagram's 30 million users don't seem to amount to much when compared to the 850 million active users of Facebook.

The conventional answer to that question seems to be that Facebook intends to keep the social space all to itself:

1. A big part of the Facebook experience is about sharing pictures
2. Instagram let users share pictures better than Facebook, and its growth was threatening Facebook's dominance
3. If you can't beat them... have them join you

Will Oremus from Slate says as much about the acquisition:

Instagram may not be worth $1 billion to anyone else—indeed, it was valued at half that price just last week—but it's worth that and more to Facebook to head off a potential challenge to its core business. Let Google or Twitter or Pinterest buy Instagram, and suddenly they're great alternatives for sharing photos. And if there's one thing Mark Zuckerberg wants, it's to make sure there are no great alternatives to Facebook.

The problem with that line of thinking is that, contrary to popular wisdom, competition helps the leader because it helps grow the category. If you try to stifle competition you also suck air out of the category, and therefore hurt yourself (first and foremost). This is what Al and Laura Ries wrote about Polaroid suing Kodak out of their core market:

One of Polaroid's biggest mistakes was forcing Kodak out of the instant-photography market. Although it won a few millions in its lawsuit, Polaroid effectively removed a competitor that could have greatly expanded the market. (A Coke/Pepsi advertising war benefits both brands (...)[,] expands the consumer's interest in the cola category.)

The 22 Immutable Laws of Branding

Of course in the case of Facebook one might object that there may not be much to expand; but that's never true, even in the case of an already very dominant player:

1. Even when the total number of users seems to have peaked, user engagement with the brand can always grow
2. If you have to play defense, it's better to attack other competitive categories than spend energy or cash to reduce competition in your own category

If you guys were the inventors of Facebook, you'd have invented Facebook

Unsolicited advice is often preposterous—even more so coming from someone who didn't invent Facebook, and directed at the guy who did. But the question remains: what will the purchase of Instagram accomplish for Facebook?

I don't know if it will make much of a difference, positive or negative. But I do think it offers a window into Facebook's soul. It tells us Facebook is more like Polaroid than like Coca-Cola. They will spare no expense to make sure no one else has a chance of sharing the social space with them.

Sharing: the one thing Facebook will not do.

When I was in elementary school, I founded the "Students in Revolt" party. I made some letterhead and went looking for like-minded individuals who would join this promising initiative. It turned out that not many of my fellow students were in "revolt"—zero being a more correct approximation. I even tried to enroll my younger brother, who was probably 5 at the time. I offered him the title of Vice President; he declined. I offered him money; he said he diddn't care about money.

The "Students in Revolt" party died for lack of members, or purpose. My revolt was still there, though, and it felt real and legitimate.

In reality, I was bored. I needed meaning and purpose in everything, which fueled my unhappiness. This lasted a very long time: from elementary school to business school and my first couple of jobs.

But it progressively dawned on me that what I really liked was to make things, and that programming is a fantastic way of making things. I became an independant contractor and have been very happy ever since.

And yet something was missing: I had no one to talk to. Amongst my friends or family there were no programmers; at clients' sites, most developers didn't seem to care much about their job.

I was looking for a guild: a place to meet skilled craftsmen in the same trade as I, interested not only in improving their craft but in learning about all sorts of topics.

Of course, Hacker News is one such place. Discovering it felt incredible.

And yet... something was still missing. Online communities give you the illusion of being gregarious, when in fact you're sitting alone behind a computer screen, most of the time not even fully dressed.

This is where Hacker News Meetups come in; they are "Hacker News in the flesh", and something else entirely than lurking on an online forum. But it seemed improbable there would ever be one in Paris.

And then the improbable happened—not just once, but four times, and counting.

4th Paris Hackers Meetup

Last Wednesday was the 4th Paris Hackers Meetup, held for the second time in the new office of Joshfire near République.

This time about a hundred people attended. Energetic, competent people (still mostly male, unfortunately). To a large extent, Hacker News Meetups seem to be take place in an alternate universe: everyone's happy, everyone speaks decent English (this is Paris!?!)... and every company is hiring,

There were two presentations: in the first one, the Mozilla team showcased new products and versions, including Boot2Gecko (new mobile OS based on the rendering engine of Firefox) and a "Tilt 3D view" that let one see the structure of a web page in 3D.

The Q&A session was a bit strange, as some people seemed to blame Mozilla for its very existence; but the Mozilla people answered gracefully. They are still hiring like crazy.

The second presentation tried to answer the question: how do you scale to absurd numbers of page views in the most efficient manner?

Michel Bartz was Lead Software Engineer at Manwin Canada and had to deal with over 5 billions page views per month. The software stack was a pretty common LAMP setup, but also included Sphinx for search, which can plug itself directly to any ODBC data source, and Redis as an in-memory key-value store. He seemed especially fond of Redis which he said never ever crashed under his watch (unlike, say, Memcached).

At this point (if you weren't there) you may wonder how anyone who's not Google or Youtube ever gets to 5 billion page views per month. The answer is porn, of course (Pornhub and Youporn). Pornhub itself incurs charges of over $600,000 dollars a month on bandwidth alone.

At the end of this talk there were many questions, mostly technical. But what Sylvain (our host) really wanted to know was if there were naked women running around in the office. The answer is no; it's a pretty normal working environment, save for the perks ("350 TB of porn... beat that!").

After that Sacha Greif and I said a few words about the latest Hacker News Meetup in London that we attended together; it was huge (500 people) and nice: many talks, including one by Joel Spolsky of Joel on Software fame and another one by Sugru maker Jane ni Dhulchaointigh (yes, that's her name). Sugru is a silicone-based special rubber with which you can fix anything. It was listed as one of the 50 best inventions of 2010 by Time Magazine. I use it and love it.

The downside, it seemed to us, was that there was not enough time or space for schmoozing. Also, London is a very strange place where people drive on the wrong side of the road... and were trees come dangerously close to the earth.

Anyway. Then Sacha demoed an ebook he's writing about "Step by Step UI Design" (no link since it's kind of a secret for now ;-).

And after all that came the beers and pizzas, and socializing. Watching a hundred introverts engage in passionate conversation, and being part of it, is quite an experience.

All in all, a most excellent installment of the Paris Hacker News Meetup—best one so far, for my money. Next time a larger venue will probably be needed, as well as corporate sponsors for the food and drinks...

Right now we're sending big thanks to Joshfire for the great initiative and great execution.

The Lemon effect

The Lemon effect happens when, in a given market, sellers know more about the products being sold than buyers. (Lemon is an American slang term for a car that is found to be defective only after it has been bought).

If people wanting to buy a used car don't know which cars are bad and which aren't, they assume every car is a potential lemon... so they're only willing to pay lemons' prices for any car. Sellers, who know the difference, are not willing to offer good cars for the price of lemons.

Therefore, sellers of good cars withdraw from the market, creating a vicious circle: the greater chance one has to buy a lemon, the more weary buyers become. Eventually there are only lemons left.

Once you start thinking about the Lemon effect, you see it everywhere.

For instance, the Lemon effect explains the success of fast-food chains: it's not about quality or convenience (or speed) as much as it's about consistency. When you're out of town, a local restaurant is a gambit. McDonald's isn't.

And by the way, the whole point of the "methodology" of making Big Macs isn't to allow "idiots" to make acceptable hamburgers; it's to ensure consistency — which isn't the same thing. If McDonald's started to hire actual chefs in all of its restaurants, and let them be creative, they would be worse off, because patrons would have no way to tell which ones are good and which aren't, and going to McDonald's would involve the same uncertainty as going to the average local burger joint.

In fact, the Lemon effect is so important that simply removing it creates markets where none existed before: that's what Airbnb did.

Buyers and Sellers

But if certain institutions can help alleviate the Lemon effect (brands that offer a consistent and predictable experience, licences issued by a trusted authority, etc.), what are individual buyers and sellers to do?

Buyers can educate themselves, but it takes a lot of time and energy. It may be worthwhile if it's your job to buy a specific category of goods, but if you're in the market for just one item, it's unrealistic.

In that case you'd rather simply look for external clues that are fast to identify and process.

What about sellers? (Honest sellers, obviously.) For one thing, sellers can't educate buyers. If you're hoping to sell your car you cannot offer to turn even one prospective buyer into a mechanic (more than one, even less so).

Sellers can do two things:

1. Send signals that the things they're selling are actually good (not lemons).
2. Look for competent buyers

Signals are effective because buyers are actively looking for them. If you can build a brand, or offer the recommendation of a trusted authority, etc., more power to you. (Price is a signal, too.)

But signals hard to get right, because they are the first thing dishonest sellers will try to manipulate (to the point where too many good signals may be a bad sign). Therefore, valuable signals are the ones that are very hard to manipulate (those for which manipulation, even if possible, is so difficult that it's probably not worth the effort).

This is why we rely so much on "word of mouth" for advice on what to buy or which movie to see. We value the opinion of our friends not because they are our friends, but because we think they're hard to game. And sometimes we even value the opinion of strangers more than that of experts, because strangers are perceived as "real people" who are probably telling the truth, whereas to assess an expert's opinion you need to be one yourself.

Looking for competent buyers is also an excellent strategy for sellers, and one the most overlooked ones.

Intuitively, we tend to believe that the less the buyer knows, the better. We'll be able to show them the shiny paint and have them appreciate the leathery smell. This is only true if you're selling lemons, and even then, you will only be able to sell them at a lemon's price.

If you have a Picasso in your home, and you know it's the real thing, you won't put it up in a garage sale; but other times people do exactly that. Case in point: the labor market.

Don't be a lemon

Every year brings new "marketplaces" for IT contractors or developers, each with its own set of rules. Inevitably, after a while, all that's left are people selling their time for 5 bucks an hour against people looking to build whole dynamic web applications for $300.

Why does this keep happening? (This isn't "commoditization". Commoditization would require standards, and having many sellers providing the exact same product. Commoditization can't happen in a prototype economy.) These sites are the epitome of the Lemon effect... a case of Lemon cancer if you will. They attract incapable professionals and clueless buyers into an orgy of incompetence, while keeping everyone else out.

In the same vein, dealing with an HR person who isn't a competent professional in your field often feels very unpleasant. Why? Because they don't have a clue. One would think that their cluelessness makes them look at you with deference, but the opposite is true: because they can't evaluate your abilities, they assume you're a charlatan and treat you accordingly. This doesn't make them bad people; it's human nature. Think of yourself buying a car from a complete stranger.

There are many posts going around on the web about why you should try to negotiate a better salary and how to do it (see this for a recent example).

All sound advice, I guess; but the question is, if prospective employers (or clients) know how much you're worth, is there much to negociate about? If they don't, how will some negotiation tricks make them suddently change their mind? When they suggest a low price for your services, they're not only thinking you may be a lemon, they're assuming you actually are one, since if you weren't, you would laugh in their face and leave.

Let's end this post

So my point is this: lemon markets are not your problem; they're okay, you know... for lemons.

If you're not a lemon, you should be very careful not to appear as one — or near one. Incompetent buyers will always assume every item for sale is a dud. It's understandable: they've been burned before. In the secret of your heart, you should forgive them, but still avoid them like the plague. Say you're currently unavailable. Change numbers. Speak a different language.

Look for people who understand the quality of what you're selling. There's nothing better in this world than doing good work for people who can appreciate it.

Further reading

See the original paper from George Akerlof that would eventually earn him a Nobel Prize. The paper uses the secondary car market only as a simple example, and alludes to many other cases of the Lemon effect at work, including health care, or how to counter it (fast food chains). It's a short and great read.