Rogue Thoughts

I have moved to http://opaqueparcels.wordpress.com/

alisdorf — Wed, 21 Jul 2010 21:13:07 +0000

Why doesn’t severe punishment seem to have any effect in pro cycling?

alisdorf — Sun, 06 Jun 2010 23:27:41 +0000

Recently I discussed an interesting topic related to professional cycling with one of my friends. The advent was the new “heat” generated around the American bicycling legend Lance Armstrong alleging, yet again, continuous doping abuse, or at least an “uncanny” depth of knowledge of use of substances considered doping.
Being a bicycling enthusiast since my childhood, this is a subject I have given some thought. In the beginning I used to be annoyed at people claiming that everybody in the Peloton is doped (because of the obvious counterargument that then they would have no advantage). Now, however, it is also the feeling I get.
But, why is it that doping is so tied up with professional cycling and why doesn’t the severe punishment like being stripped of your victory have any effect?

Let us be clear about one thing: doping, or the use of performance enhancing drugs, is an arbitrary measure. The actual substances are medication that saves lives and are for the greatest part not dangerous (or at least not more dangerous than sprinting downhill with 100 kilometers per hour unprotected alongside a hundred similarly racing and unprotected bicyclists). The only extra risk professional riders take, wen they use doping, is the risk of being caught. Proper medical counseling makes sure that the use of doping is safe.
Let us be clear about another thing: Doping REALLY helps. Depending on how much you train and what sort of substance you use, you could increase your performance by at least 10 percent.
And a third thing is that the recent wave of confessions to extensive substance abuse by retired riders testifies to the fact that statistically the odds of being caught are low.

This combination seems to be driving the adoption of doping in professional cycling:

Professional cycling is thus a game where cheating gives a huge advantage. The punishment for being caught is severe but cheating carries no extra cost, such as damage to your health. The odds at being caught are low (I wonder if it is possible to frame it in more precise game theory terms)

Every time someone is caught it has consequences, not only for the perpetrator, but also for the whole sport. A typical conclusion is that everyone is doped. Consequently, for every high profiled conviction the prestige of the sport is lowered and thereby the earning potential, hence there is a double angle to the doping story, since the motivation to police is inversely proportional to the number of perpetrators caught. Remember that policing is the only thing that has any effect on doping

It seems that there are two equilibria here: one in which policing may be low, under the doctrine of “see no evil – hear no evil”. In this scenario doping is widespread, because the odds at being caught are low. It may demand one or two random convictions to set an example – TO THE OUTSIDE WORLD
The other equilibrium features extensive policing, and may similarly catch a few, but doping is not widespread, because odds at being caught are high. Here, however, convictions set an example TO THE INSIDE WORLD of the sport.

The severe punishment of taking away the victory and banning athletes for years, does not work, I believe, in the first scenario, because the punishment is felt as a stochastic side effect of doing business; just like you may crash when racing downhill. It is not really felt as something significantly tied to the individual dispositions of the athlete himself, since cheating is part of the game.

In the second scenario, doping is not widespread and therefore the choice of taking doping IS tied to the individual dispositions of the athlete, and hence the punishment is felt as such. Here it would work to punish.

Now, the million dollar question is how do you go from the first equilibrium to the second? Unfortunately I have to stop here, but feel free to continue…

Why “wikinomics” seems revolutionary and why it will certainly fail

alisdorf — Tue, 01 Jun 2010 23:11:57 +0000

“What has been will be again, what has been done will be done again; there is nothing new under the sun. Is there anything of which one can say, “Look! This is something new”? It was here already, long ago; it was here before our time.” (ecc.1.9-10)

Recently I had an interesting conversation with a colleague of mine. He teaches entrepreneurship at a business school. He told me about his students’ approach to business and entrepreneurship which is completely different from classical economic theory otherwise taught at the Business School.

The claim went something like this: Usually when you as an entrepreneur come across a new idea or make an invention, the first thing you will do is to protect it with a patent. Young people act quite differently. They publish it on the internet, develop it through different means of crowd sourcing, and then they offer it for free, but make money on associated services. The model, sometimes called wikinomics, has been hailed as revolutionizing and changing the economy fundamentally.

I admit that it is a new model, which is interesting. I also do admit that young people are inventive, and that I am getting older. BUT it is kind of not the first time we heard that the young generation is changing everything. The same was said of my parents generation in 68 doing away with war and stuff. Then came generation X (which I think is me or I just missed it) that was just overeducated and underpaid (yeah, I think that IS me), then came generation Y, who demands that the workplace adapt to their life.

The sad fact though is that, nothing is ever changed fundamentally by the current “young” generation. I will explain why the “young” generation is always revolutionary why it always comes to nothing eventually

The apparently innovative behavior of the young generation is powered by an intergenerational dynamic – expressed by the new categorical imperative “Act only according to that maxim whereby you can at the same time be sure your parents wouldn’t do it”, when this oscillation meets a certain economical climate (up or down) fun things happen: up led to hippie culture, down lead to punk – that is the reason why young generations are always revolutionary.

The reason why nothing ever comes of it is that being young carries less obligations and therefore much more individual maneuverability with it. Eventually young people grow up and face the same conditions that shaped their parents lives.

To return to the alleged fundamental change in entrepreneurship heralded by the new “young” generation: The apparently innovative behavior comes from behaving differently than their parents: Well, their parents certainly wasn’t on the internet, so that an easy one. They sit in their dorm typing on the Mac their parents’ economic fortune has equipped them with. They have no real need for money, either because their parents saved money for college or because they are fortunate enough to get state subsidies (as we do here in Denmark) or loans. Either way, they have no fear that they will not somehow be able to get money for the next pizza. This allows them to develop apparently radically new business models. These business models, however, are only competitive because the businesses only need to make sure that the next pizza can be paid when it arrives. They are not faced with greedy middle aged shareholders (such as their parents – whom they already moved away from!) demanding profit from their investment every quarter.

When this young generation grows up they will realize that a normal family life with wife kids and a house (as is still the most common) does not run on 14 hours per day spent writing wikipedia articles and an endless supply of pizza: you need to supply Nintendos and vacations to Disneyland and listen attentively for hours to your spouse and even occasionally engage in real conversation!

I may of course be completely off track, but can you really mention more than a handful of successful businesses founded on wikinomics?

Scientific to the bone

alisdorf — Tue, 18 May 2010 22:14:45 +0000

It is rare to find people who really live by their ideals and cut to the bone when they want to know something. I just read about one such person: Donald L. Unger.
Haven’t heard of him?
Donald L. Unger is a doctor. When he was young his mother told him, as mothers do, that he should stop cracking his knuckles, because he would get arthritis. Come adolescence, and little, or not so little anymore, Donald catches a bit of defiance. He begs to differ on the putative cause of arthritis expounded by his mother. So, he decides to put it the claim to the test and cracks his knuckles on his left hand every day.
He has been doing so for more than 60 years now. An estimate is that the knuckles on the left hand have been cracked approximately 36.500 times. It turns out, that today even after this he still has no arthritis in either hand.
As much as I admire this man for his perseverance and contribution to our knowledge of arthritis, I admire him even more for the conclusion that he decides to take away from his research: “This result calls into question whether other parental beliefs, e.g., the importance of eating spinach, are also flawed. Further investigation is likely warranted.”
’nuff said….

Relevance equals demand in internet advertising – a quick guide for suppliers

alisdorf — Mon, 26 Apr 2010 21:29:11 +0000

If you figured that the internet is probably here to stay, and that, although what you offer this internet, is of supreme quality, you DO have to advertise it..there… on the… internet. Here is a quick guide to how to get the best of it.

There are several different ways to do online advertising. But first it is important to keep one single thing in mind: When you want to communicate something it is always important to realize what the recipient finds relevant. Relevance is the ‘demand’ of internet advertising, and guess what…on this..internet it is a buyers market.

Today there are three payment models in use. Let us take a look at these, to find out which one suits you best
CPM – Cost Per Mille: The advertisers pay per thousand expositions. Typically used by newspapers, radio and television. For this to be effective the advertising company must make sure that the segment of users (newspaper readers for example), matches the intended target segment. A hit-list pop station may be good for soft drink adds while the jazz station may be better suited for bourbon. The relevance thus depends on how well the target segment fits the actual recipient segment AND how well there is an actual fit between the stipulated relevance of the add for the target segment.
Pros: Advertisers know exactly how many he will pay for
Could be cheaper per exposure than the alternatives
The model is intuitive
It is more diffuse
Cons: Low correlation with sales
Lower probability that it will actually be relevant for the individual recipient
When/when not to use: When you have a wide segment or want to control the context in which your brand appears. If you have a small donut shop and want to move in to the snobbier abodes of the pastry market an add in Gourmet Magazine. You wold probably not want to use this model if you are running a campaign for your new size 1/8 drill for hardened steel

CPC – Cost per click: the advertiser pays per click, this is perhaps the most common model. It is used by Google’s AdWords, but yahoo and Microsoft also offer the service. CPC billing is found in two forms, the original one was the “Flat-rate model”, where every click is billed with the same price. Then google invented the “Bid model”. A good overview can be found in this article. The bid model essentially bills the advertiser according to a bid on search key words. This is currently the most widely used. The relevance depends on how well the “Key word” matches the content of the Add.
Pros: It is targeted according to the action of a user, ie. he/she has typed a word stating intention.
Cons: It is susceptible to fraud, because of false clicks
It is not very intuitive (the bid model)
it is difficult to figure out when, where, why and how many times it will be featured
When/when not to use: When you have a very well defined product that easily translates product offering into search phrases. If you have niche offerings, the further out the long tail they are the more effective the CPC model will be. So if you specialize in pink bicycle apparel, you know what to do. If you are at all concerned about effectiveness and would not want your new line of farmers-market-type-dairy products ending up in any search for jugs, you may want to choose something else.

CPA cost per action: The advertiser pays for every time an add results in some action, like buying a product. An example of this is a company called jellyfish, which was acquired by Microsoft October 2007. It is now part of Bing known as cashback. They only charge when the customer actually makes a purchase, or some other agreed upon action.
Pros: Very transparent and intuitive for the advertiser
No possibility for fraud.
Cons: Difficult to use if you are not an online merchant, or cannot sell your product immediately (like for example an insurance policy or a bank loan)
When/when not to use: When you have a very well defined objective, like signing up members for a political party like, say, “The Official Monster Raving Loony Party”. It may not be of much use if one day you find yourself the Chief of Tourism in Butan.

Burka rapportens videnskabelighed

alisdorf — Sun, 24 Jan 2010 16:36:43 +0000

“Burka rapportens” formål er at undersøge hvor mange, der går med Burka i Danmark. Er det besvaret godt nok? Jeg er selv uddannet som ph.d. fra Institut for Tværkulturelle og Regionale Studier, afdeling for religionshistorie i religionshistorie og har arbejdet der, så jeg har mødt nogle af de implicerede forskere, men har ikke i dag noget professionelt eller privat forhold til dem og har dermed ikke nogen interesse i at støtte dem.
I den religionssociologiske del af rapporten mangler refleksion omkring forskerne, som har udført rapporten og de kilder, som de bruger.
De forskere, som har gennemført den religionssociologiske analyse, er ikke på nogen måde ”internationalt velanskrevne” forskere indenfor området. Men de ved naturligvis en del om Islam. Alle jeg har mødt som studerer Islam på instituttet har en tendens til at forsvare muslimer (med rette eller urette). Mig bekendt, er de involverede forskere ingen undtagelse. Ikke at det ændrer ved noget, men nogle kunne dog mene, at det kunne medføre en speciel interesse i hvorvidt det billede rapporten skaber er et skræmme billede.
Lad os gå videre til kilderne. Husk på, at antallet af burkaklædte kvinder er ligefrem proportionalt med, hvor skræmmende billedet er. Det ved imamer såvel som almindelige muslimer også.
Kilderne til antallet af burkaklædte er: Imam Abu Bashar og tre anonyme, ubeskrevne informanter (kun 7 i Odense), 16 muslimske friskoler (6 i København), Sundhedsplejerskerne på Vesterbro, Nørrebro og Bispebjerg (20-25), en optælling foretaget af forskerne selv i Club Danmark Hallen til Eid festen, hvor der angiveligt skulle have været 15000 (12). På landsplan vurderer talsmand for Islamisk trossamfund Imran Shah, at der er (30-40), Imam Abdul Wahid (højst 100), og Zubair Butt Hussain (50-15), altså nærmest udelukkende vurderinger foretaget af imamer og muslimer.
Når man iagttager disse kilder er det bemærkelsesværdigt at se, at kun en kilde ikke har nogen direkte interesse (sundhedsplejerskerne). Samtidig er det den kilde, som giver det højeste estimat. For den kritiske læser er det, lad os bare sige, kritisk, at der i det mindste ikke reflekteres over dette.
Hvordan har forskerne sikret sig, at de har talt alle 15000 og ikke kun set et udsnit på f.eks. 2000? De kan meget vel have falkeblik, men der er ingen refleksion over det i rapporten.
Personligt har jeg svært ved at tro, at konklusionen er væsentligt forkert, men den sjuskede forskning åbner for allehånde angreb, som totalt underminerer rapporten og reflekterer uheldigt tilbage på den institution, som har forfattet den.

Direct marketing based on adopter categories

alisdorf — Sat, 16 Jan 2010 13:49:18 +0000

In an earlier blog I discussed recommendation engines on the internet such as amazon’s and Itunes’ and the algorithms behind it. They usually do not take into account metadata about persons, such as demographic or psychographic data. That makes a lot of sense, because it would require the customers to fill in a lot of information about themselves.
Some effort is usually made in a classic segmentation paradigm (such as the one championed by Tesco) to incorporate demographic characteristics into marketing. That is the typical strategy of loyalty programs.
Not much has been made of psychographic data. First of al because it is more sensitive, but also because it is more intangible. However, the psychological science of today is reasonably agreed on the basic personality traits: the “Big Five” factors and there are standard tests to asses the score on these factors. It should be a reasonably straightforward hypothesis that your personality is a powerful predictor of your shopping habits such as product preference. Consequently the return on the marketing spend depends on the fit between the product and the personality traits. I would think that there is potential here.
One way to do it is to look at adopter categories. We know a good deal about these from the diffusion of innovations literature (the central work is Everett Rogers’ “Diffusion of Innovations”). According to this literature there are 5 basic types of persons related to the adoption of a new product: Innovators, Early adopters, early majority, late majority, and laggards.
The article in Wikipedia describes them thus:

Innovators
Innovators are the first individuals to adopt an innovation. Innovators are willing to take risks, youngest in age, have the highest social class, have great financial lucidity, very social and have closest contact to scientific sources and interaction with other innovators.
Early Adopters
This is second fastest category of individuals who adopt an innovation. These individuals have the highest degree of opinion leadership among the other adopter categories. Early adopters are typically younger in age, have a higher social status, have more financial lucidity, advanced education, and are more socially forward than late adopters (Rogers 1962, p. 185).
Early Majority
Individuals in this category adopt an innovation after a varying degree of time. This time of adoption is significantly longer than the innovators and early adopters. Early Majority tend to be slower in the adoption process, have above average social status, contact with early adopters, and show some opinion leadership
Late Majority
Individuals in this category will adopt an innovation after the average member of the society. These individuals approach an innovation with a high degree of skepticism and after the majority of society has adopted the innovation. Late Majority are typically skeptical about an innovation, have below average social status, very little financial lucidity, in contact with others in late majority and early majority, very little opinion leadership.
Laggards
Individuals in this category are the last to adopt an innovation. Unlike some of the previous categories, individuals in this category show little to no opinion leadership. These individuals typically have an aversion to change-agents and tend to be advanced in age. Laggards typically tend to be focused on “traditions”, have lowest social status, lowest financial fluidity, oldest of all other adopters, in contact with only family and close friends, very little to no opinion leadership.

Now, my point is that just as it is possible to make tests to ascertain the big five personality tests, it is possible to make tests to ascertain which of the five adopter categories you belong to.
With this knowledge it is possible to market new stuff to the innovators, who will appreciate it a lot more than the laggards, and old stuff to the laggards, who will appreciate it much more than the innovators. I guess what I’m trying to say is send me an offer for the new google phone and not the Nokia 6700…

The marketing effect wold be much greater for the company and the customers who will receive the marketing will get experience more relevant offerings.

Robot retail

alisdorf — Sat, 19 Dec 2009 12:16:42 +0000

If you have purchased something on Amazon, iTunes or the like you will already have come expect that they will recommend something else, which you might like. Would that work in the supermarket? Superficially a yes seems probable, but does it take the same kind of intelligence to shop books as bananas?.

Recommendation engines are a simple form of artificial intelligence. (at least compared to the more illustrious versions known from popular culture – such as the film “Bladerunner” or “I Robot”). The term artificial intelligence was originally coined in 1956 by John McCarthy and has since then followed two branches. One branch is philosophical and ponders the nature of the human mind, while the other is more technical and tries to apply artificial intelligence in technological solutions.

The technical branch has brought us the algorithms that exist today on the web in the form of recommendation engines like Netflix, jjester, stumble upon, last.fm etc.. This particular type of algorithms are called collaborative filtering algorithms. They have been fine tuned in the extreme, but they are all aimed at one particular problem: recommending a limited number of new items that the customer/user has not yet tried from a vast repository. One key characteristic of these items, whether they be films (Netflix), jokes (Jester), websites (Stumble upon) or music (Last.fm) are that they are only meaningfully used once (music is a partial exception, but still bought only once).

Now, working for a food retailer, we wanted to apply the same kind of AI to retail, but found ourselves faced with the problem that the repository of possible items was not so vast, since a typical large retailer will have between 20000 and 40000 products at most. The typical consumer would buy mostly the same products repeatedly, which clearly differs from movies (I know I saw Star wars episode 5 more than thirty times, but I was young and it is an exception), jokes (although some people really do tell the same joke over and over again, that is also an exception), websites (this one is more tricky, but when you want stumble upon to recommend a website, it is a new website you haven’t used before), and music (again I must admit that I have bought Aphex Twin’s “Selected ambient works Vol. 2″ five times, but it was because it kept getting lost for some reason).

Of course, you can use the collaborative filtering algorithms to introduce a customer to new products, but often the customer may be more interested in locating offers on stuff he usually buys, or just stuff that he ran out of and has to replenish. Another thing would be seasonal recommendations, like offers on fresh berries when they are in season. I still haven’t found the optimal solution for this type of problem where you want recommendations for items that are used repeatedly and selected from a limited base where novelty is not the key driver of customer appreciation.

Obviously some sort of time dimension has to be woked in to it. There are at least two sorts of time here, the cycles that drive humans, such as the day, the working week, the year and the life cycle. But there is also the product life cycle; how often the customer needs to refill his olive oil stash and coffee stock.

I believe the solution points to a more fundamental challenge for the AI powered marketing of the web 2.0 and beyond: introducing the cyclical time dimension that drives every cell in our body, that is the diurnal, and the more cultural ones like the working week/year. That is the next big challenge for artificial intelligence of web 3.0 and beyond.

Halal, Kosher, Organic, CO2 friendly?

alisdorf — Wed, 09 Dec 2009 00:45:17 +0000

Working in a company obsessed with Corporate Social Responsibility and having a green profile, I recently had the good fortune of being exposed to a sparkling soft drink offered without charge at my job. Perhaps out of boredom or natural inquisitiveness, I suddenly felt compelled to study the label. I found that organic fruits had been used, but more disturbingly I found that it contained organic CO2! after recovering from the out of place conjunction of these terms, this situation made me think a bit about why we act “responsibly” and consume products that are organic and CO2 friendly.

Personally I prefer to buy organic food, and have a generally low level of energy consumption etc. But I am becoming increasingly weary of the tone and manner in which these things are communicated.
I don’t like the fact that I am expected by the community to buy a product based on some clean/unclean distinction reminiscent of religious discourse. People who eat food labeled kosher or halal do it because of a religiously constructed demand for such products. A large conglomerate of ‘civil servants’ called priests or imams, have branded their club and have a constitution embedded in a deliberately cryptic piece of literature, called the Quran or Thorah, that is, their canon. In this canon there are prescriptions about a good many things, such as how to behave responsibly and what to eat and not to eat.

These clubs have been very successful in their marketing efforts and brand positioning, because they have been able to include the brand in a narrative. As has been convincingly argued by professor of literature Mark Turner, pychologist Jerome Bruner and philosopher Daniel Dennet, the human mind is a literary mind, in the words of Turner; life is a narrative, in the words of Bruner and the self is a center of narrative gravity in the words Dennet.
The Jewish club has embedded the rules for proper food and conduct (kosher) in the narrative of the Torah, where it plays a part in the future demise of the world as we know it and the ultimate rescue of the individual club members abiding to these rules to a cooler place than here, that is heaven, at some undisclosed point in the future.
The Islamic plot is disappointingly similar: act as prescribed pray to Allah five times a day, go to Mekka plus eat halal and you will transpose your existence into a much cooler place at some undisclosed point in the future.
Religious clubs’ marketing strategy is mimicked by an emerging club of what we could call ‘climate guardians’. This club’s canon is the IPCC’s report on climate change. This canon also has a base narrative about the impending doom, and possible rescue of the members of the club to a literally ‘cooler’ place at some undisclosed point in the future, rather than the hellish prospect offered by the present prognosis (granted, this club is more democratic since it saves everyone, not just the members). In this club you also have to eat properly, that is organic or CO2 friendly, and act responsibly, fore example by turning out the lights, buying small cars etc…

Personally I will continue to buy halal, kosher, organic and CO2 friendly food and turn out the lights, I will just not be coerced into it by narrative hypnotists, such as imams or climate evangelists. I will do it because, or in so far as, there is persuasive evidence that it has beneficial economic or health effects to my family and me, or because it tastes d*** good…

That was not the case for the sparkling soft drink, so I will leave it for a dedicated club member.

Windows 7 – a Kafkaesque-50-First-Dates-ish Vibe

alisdorf — Wed, 02 Dec 2009 21:01:49 +0000

Recently I was forced by powers out of my control (family pressure) to buy a new computer, primarily because I needed one that would run a Windows operating system. Since I had an early traumatic experience with Windows Vista, I would rather buy a typewriter than buy a computer with that installed.
I had read several not too negative reviews of Windows 7, so I was hoping that this would give me an operating system that would be close enough to Windows XP, which is bearable to work with.
Searching for a cheap computer, my choice landed on a Toshiba satellite l450d-10z.
So, I bought the computer, came home and opened it with trepidation. I reckoned that finally I would be hailed as a decent father, supplying the hardware and software necessary to play the Sims, and a good husband that would make everything work.
I pressed the on button. I selected Danish language, since my oldest daughter still prefers that. Then the computer carried on with its installation duties. Eventually a screen came that informed me that it would now run Windows for the first time. I thought that was fair enough. A new screen came and I waited while it was seemingly working on something. Then the computer restarted by itself. I was a bit suspicious. The message about running Windows for the first time was repeated. The computer restarted. And the message repeated. And the computer restarted… after three times I thought it was time to go for a walk with the dog. I went, came back and it was still starting Windows for the first time.
Eventually I succeeded in interrupting it and installed updates and some other programs. It worked fine. I felt happy. Restarted the computer. And it informed me that it would now run Windows for the first time…
I felt like being in the movie “50 first dates”, just without the romance and the Hawaiian scenery, although I could have probably used someone like Adam Sandler to cheer me up.
The diagnosis was obvious: the system forgot everything when it restarted. I decided to try to with strategy and decided to hibernate it instead of restarting. Filled with high expectations and satisfied with my superior intelligence I went for a cup of coffee. I came back and turned it on, convinced that I had finally outsmarted the system. I pressed the button with the active user and a message came that my account had been deleted and I should contact my system administrator. I looked slowly around in my living room although I was pretty sure I had not let in any system administrators this particular evening. Where was my system administrator? Now I was getting more of a Kafkaesque vibe.
I pressed restart, but the system informed me that another user was using the system. I looked around the room again…turned it off and went to bed.