Algorithms to Live By

Algorithms to Live By: The Computer Science of Human Decisions by Brian Christian, Tom Griffiths.

Cover page of book Algorithms to Live By: The Computer Science of Human Decisions by Brian Christian, Tom Griffiths

All models are wrong. But some are useful.“–George Box

This book provides detailed exploration of some models or algorithms as applied to daily problems–ranging from marriage decision to design of parking lots. Simultaneous exploration from two standpoints–computational and psychological–brings two interesting insights.

First, it makes very clear that all theory and practice could be very different. As authors noted “Communication is one of those delightful things that work only in practice; in theory it’s impossible.” It doesn’t stop here and go in details what and why differs. For instance, in chapter on explore-exploit dilemma it turned out that people tend to over-explore. We switch to explore mode faster than algorithm suggests. However, it makes perfect sense if one take into account that model assumes unchanging world. If environment changes–as it constantly does–switching to explore is a good approach.

Second, it factors in computational burden. This could be crucial for explaining and design. The trick is seemingly similar things use different algorithms with different computational burden. For instance, it turned out that people are more likely to be available when you request meeting with specific time (or a number of options), rather than “at convenient time next week.” The former constrained problem is easier to solve. The latter launch resource consuming scheduling algorithm.

Overall the book is an excellent reading. It combines in academic arguments, everyday observations, anecdotes and philosophical pieces.

P.S. I don’t remember exactly how this book came to me. I believe, it is coming from someone reading list.

Sarcastic one-liners vs Deep feelings

Sarcastic one-liners are the trademark of Chandler Bing in the “Friends” series. Under this rough and spiky skin Chandler hides the heart of gold. His epic bromance with Joey is one story. Remember, how he invented the game “Cups” to ensure that Joey get money without hurting his feelings. His romance with Monica is full of tender moments

Arguably, the most emotional moment of the whole series is the scene where Chandler convince Erica to choose him and Monica to be her baby’s parents (Season 10, Episode 9 “The One with the Birth Mother”). It shows very different Chandler, and how he truly feels and sees Monica. And this is a great example of persuasive speaking.

“My wife’s an incredible woman. She’s loving and devoted and caring. And don’t tell her I said this, but the woman’s always right. I love my wife more than anything in this world. And it kills me that I can’t give her a baby… I really want a kid. And when that day finally comes, I’ll learn how to be a good dad. But my wife… she’s already there. She’s a mother… without a baby… Please?”

Why did the chicken cross the road? To lead by example

Humour is often used as a communication device, to charm and to offend, to mock and to encourage, but ultimately to persuade. Despite the common usage, the question “to joke or not to joke?” still provoke heated debates. Scientists cannot miss this research opportunity. A team from University of South California conducted an analysis to better understand the contingencies of humour effects. The results are summarised in a paper “A Priest, a Rabbi, and a Minister Walk into a Bar: A Meta-Analysis of Humor Effects on Persuasion“. They scrutinized some hundred studies and find three interesting things about effects of humour on persuasion.

Cartoon. Chicken at Psychotherapist: "Why do _you_ think you cross the road?"

First, indeed humour has effect on persuasion. Overall this effect is weak, but robust. But, humor has a moderate-level influence on knowledge, but only a weak impact on attitudes and behavioral intent. In other words, use jokes to support learning. But don’t expect gigs would significantly change attitudes and behaviuor.

Second, jokes should be relatable. The study found that the most effective is humour central to the message, addressed to the topic important for people (or, as authors put it related-humor for highly-involved individuals). When humor was central to the message tended to exert more impact on knowledge, attitudes, and behavioral intent, compared to messages that used humor only peripherally. What is is more important, effects were stronger when the topic of the message had direct consequences for the participants’ lives (what authors call highly-involved individuals). Hence, surprisingly, humorous instructions on important things work better. Safety procedures during a flight or a message advocating against drinking and driving addressed to students, just to name a few.

Third, use it but don’t abuse it. No jokes is bad, too many jokes is even worse. The results revealed an inverted U-shaped effect of humor intensity on persuasion. Small amounts of humor may not be enough to draw attention. Too much humor overwhelm the processing of information, things getting “too funny”. One need to calibrate and moderate. The good news is that one should just avoid extremes. There is a broad range of humour intensity, where it works well.

The plotted regression of humor intensity on persuasion.
Figure. The plotted regression of humour intensity on persuasion.

Pronunciation made easy

Pronunciation‘ is one of the most commonly mispronounced words. This fact should alert you about particularities of English. During its history English language accumulated words of different origins–Germanic, French, Roman and Greek to mention a few. That is why we have to learn both spelling and pronunciation. Because all ‘c’ in Pacific Ocean sounds different, and because ‘Ghoti’ is a creative respelling of the word ‘fish’.

Crazy English: ghoti by
Crazy English: ghoti by

Technologies make pronunciation easier, if not easy. Two could be particularly useful when you prepare a speech:

Recorded pronunciation in dictionaries. Oxford Learner’s Dictionaries offer pronunciation of words in both British English and American English. Merriam-Webster is an alternative for American English. Simply click on a speaker icon and hear how the word sounds. Try to pronounce it. Hear again. Pronounce again. Repetition makes perfect!

Pronunciation of pronunciation in Oxford Learner's Dictionaries and Merriam-Webster dictionary,

Use Google Translate to listen the whole sentence. There is also a ‘Listen’ button in Google Translate, available for both source text and translation. Select appropriate language, click ‘Listen’ button and hear how your sentence sounds. This could be a great training tool for your speech. (By the way, you could even beatbox in Google Translate.)

Using of Listen function in Google Translate

How I stopped worrying and learned to love the Grammarian role

What could be worse than a Grammarian role?! Poking finger into people’s mistakes? Sounds dull and unexciting. Moreover, for many Toastmasters English is not a native language. Personally, I’ve learned it in a very strange way–practice separately, grammar separately. And I don’t shy away from creative English. I routinely use the verb ‘to guineapig’ for testing done scientifically. Once I used ‘unmisoverestimateable’ to slow down one crazy daisy manager. (If you wonder what does it mean–something so grandiose, that it cannot be overestimated by mistake. I warned you about my English!) So, during two years in the club, I took the Grammarian role only once. It was impossible to dodge it in the legacy Competent Leadership Manual.

Ancient Grammar Difficulties

This time it was different. This time I jump on the role of Grammarian. I need it to collect the flush royale of Toastmasters roles. No way to dodge. But, could I make it a useful and fun experience? What could be great in this role? Why does this role exist at all? It struck me that I was looking at this role from the wrong standpoint. I was pondering “Whats?” of the role, while the right question to ask was “Why?” In essence, the Grammarian is an educational role. Its scope is to help people improve their English. This “Aha!” moment brought me “Why?” Then “How?” and “What?” came easy.

I decided to make the experience of English grammar fun and educational. Recipe? Take a Word of the Day. Mix with spotting predictable mistakes, season with an education mini-session. Top up with good examples from participants. My Word of the Day session was longer than usual. I linked it to the speech topics–persuasive speaking–and chose the word “coax”. I employed a dynamic Powerpoint Slide to bring pronunciation and examples. That was an hors d’oeuvres. The main course was spotting mispronunciation and running an education mini-session. While “foreign accent is a sign of bravery,” it is also quite predictable, especially when you know who is talking. It was very easy to capture a couple of examples. Next, I did a trick. Instead of playing the blame games and lambasting mistakes, I offered a way to improve. Dictionaries offer great pronunciation functionality, both in Oxford Learner’s Dictionary for British English and Merriam-Webster for American English. There is a wonderful poem “The Chaos,” which highlights all complex pronunciation cases. Last but not the least, I hunted for good expressions used by speakers and paraded my trophies in the final evaluation. “Motivation gets you going, but discipline keeps you growing” was one example, which resonated strongly with the speech topics–persuasive speaking.

It worked amazingly well! People were quite happy to learn something new. I reached the goal of my role–to help people improve their English in an encouraging and supportive manner. I learned a lesson–start with “Why?” and frame it properly. Choosing a particular means of “How?” and “What?” is easier when “Why?” is clear. And I have just Toastmaster of the Evening to complete my flush royale of roles.

To Sell is Human by Daniel H. Pink

“To Sell is Human” by Daniel H. Pink is very readable and practical book about sales. He argues that most of us (well, US workforce, and relatable top many in post-industrial economies) are involved in sales, including “non-sales selling”–coaxing people into something, not necessarily meaning immediate movement of cash. Hence, argues Pink, selling is very human activity, although in XXI century it requires some new approaches. ABC of “Always Be Closing” is now replaced by ABC of “Attunement, Buoyancy, Clarity”. In information-rich society caveat emptor give a way to caveat venditor;.

The central part of the book for me is Chapter 7 on Pitch. I find it very useful and practical. The rest is nice salad for this main course–well prepared, well researched, very convincing, and addition to the something central. The positive thing about the book is its practicality. It includes many advice and techniques (based on well documented research), which you could start putting in practice immediately.

P.S. I’ve got this book from some internet list “20 Sales Books You Need to Read in 2020”. However, I’ve heard about it previously, it was sort of hanging on my interest list for some time.

Reps Reps Reps!

This is a draft of Arnold Schwarzenegger‘s speech at UN. Each tick mark at the top mean one rehearsal–I counted 55. How many time you rehearse your speech?

Reps Reps Reps by Schwarzenegger

Reps Reps Reps by Schwarzenegger

No matter what you do in life, it’s either reps or milage. […] [The practice of putting doiwn check-mark for each rep] had a huge impact on my motivation. I always had the visual feedback of “Wow, an accomplishment. I did what I said I had to do. Now I will go for the next set, and then the next set.” Writing out my goals became second nature, and so did conviction that there are no shortcuts. It took hundreds and even thousands of repetitions for me to learn to hit a great three-quarter back pose, deliver a punch line, dance the tango in True Lies, paint beautiful birthday card, and say “I’ll be back” just the right way.

The power of improvisation. Directed improvisation.

There is no shortage of ‘best practices’ or ‘one-size-fits-all’ solutions in a development economy. Free trade, democracy, institution building, you name it. The Washington Consensus has reigned in the decades since the 1990s. Now the Beijing consensus is emerging.

Yuen Yuen Ang dismisses such solutions and suggests that there is no universal prescription. Her one-sentence summary is “Poor and weak countries can escape the poverty trap by first building markets with weak institutions and, more fundamentally, by crafting environments that facilitate improvisation among the relevant players.” She explores this idea through studying how China managed to achieve and sustain economic development after Mao. The country employed this approach, which allowed the to achievement of economic growth in complex circumstances. Yuen Yuen Ang call this approach ‘directed improvisation’ where central reformers direct and local state agents improvise. It taps local knowledge and adapts to the local circumstances, while aiming at an overarching goal. The resulting transformative process has displayed three distinct patterns. It is broad, bringing systemic changes despite incremental reforms. It is bold, unusually entrepreneurial but also attracts corruption-prone bureaucrats. Finally, it is uneven, with wide regional disparities coexisting with national prosperity.

The Government nurtured what Yuen Yuen Ang calls ‘directed innovation’ through variation, selection and niche creation. To promote variation, central reformers allowed local agents to flexibly implement central mandates according to local conditions. This has been done through deliberate creation of grey zones, as too much leeway could create chaos. Hence, Central bodies clearly delineated these zones of local improvisation. They imposed red lines around local administration, denoting things which are prohibited and risk very severe punishments; and black lines for things which must be delivered, again at the risk of severe punishment in the case of non-performance. The rest was in a grey zone, open for innovation and adaptation to local conditions. Selection was promoted by clearly defining and rewarding success within bureaucracy, of the type in the black lines discussed above. Central reformers clearly communicated the criteria for success to lower levels and ranked localities, and closely looked for what has worked and what didn’t. Successful models and approaches then became central policy, scaled up and replicated throughout China. For instance, the famous Township and Village Enterprises (TVEs) were neither prescribed, nor anticipated by central reformers, as Deng Xiaoping himself admitted. They grew up out of local experimentation at that stage of the reforms, as best fit to local needs and conditions, to produce growth spurts, and centrally imposed restrictions for the of non-acceptability of private property. The diversity of China provided the raw material for innovation, resulting in niche creation for different localities. Regional diversity thus turned from liability into a collective advantage.

The methodology Yuen Yuen Ang used is a mapping of the ‘coevolutionary process’, with whole Chapter (1) and Annex (A) devoted to the description of the methodology. She does not engage into construction of sophisticated regression models, torturing data in the elusive quest for causality out of correlation. Neither does she stick to small N approach, looking through messy and overcomplicated set of variable for a single case (By the way, enquiring reader could find great discussion of cultural differences between small N versus big N approaches in “A Tale of Two Cultures: Qualitative and Quantitative Research in the Social Sciences” by Gary Goertz and James Mahoney). Hence, no oversimplification and no messy non-reductionist approach to complexity. Rather, a complex approach, which captures a non-linear, co-evolutionary process, in reduced form. She tracks changes in the related systems of economy and institutions, over the time. Throughout the book, she dives into examples of Forest Hill, Blessed County and Humble County in China, which are archetypes of various types of localities in China. In her methodological annex she extends this approach to two additional cases of tax-less finance in United States of America in the 1880s and raise of Nollywood in Nigeria.

The three ‘I’s haunting development economics, according to Esther Duflo, are a conceiving Ideology, often derived out of Ignorance, that is perpetuated as a result of Inertia (see also “Poor Economics: A Radical Rethinking of the Way to Fight Global Poverty” by Abhijit Banerjee and Esther Duflo). More and more we do recognize the complex nature of issues we face with, the inadequacy of opaque models pushing correlation for causality, and the lack of Silver Bullet solutions. However, we are yet to find the instruments for handling complexity in a meaningful way. The book by Yuen Yuen Ang is an excellent starting point for this intellectual inquiry.

Keep it simple: Complexity and the SDGs

As Einstein says, “Everything should be made as simple as possible, but not simpler.”

Grey wolves in Lamar Valley. Photo: Yellowstone National Park

Grey wolves in Lamar Valley. Photo: Yellowstone National Park

Can wolves change a river flow? Hard to believe, but yes.

In 1995, grey wolf packs were reintroduced in Yellowstone National Park, with remarkable consequences for the entire ecosystem. Elks, afraid of being eating by the wolves, began avoiding open regions such as open meadows and gorges. New plant growth increased, provoking subtle changes in the park’s waterways. An increased number of trees attracted various bird species. The beaver, previously extinct in the region, returned and their dams attracted otters, muskrats and reptiles. The rivers meandered less, reducing erosion and forming more pools. So ultimately, the wolves helped stabilize river banks and fix the rivers in their courses.

This story resonates strongly with my experience with the Sustainable Development Goals, especially as we at UNDP help countries to embrace and domesticate Agenda 2030. We are living in ecosystems, not hierarchies. Our relationships in the world are circular, not linear.

Linkages are more important than the elements of the system themselves.

For example, in Bosnia and Herzegovina some 40 percent of companies have vacancies hard to fill due to applicants’ lack of skills. But at the same time, nearly 25 percent of its youth aren’t working or in school or training. People complain that not knowing the right people is a major obstacle in getting good job (56%), not just lack of jobs (35%) or inadequate or irrelevant education (14%). These relationships are clearly non-linear and cannot be solved by traditional approaches, assuming that a good education will find you a job.

In this instance, UNDP looked at Bosnia and Herzegovina’s local employment “ecosystem” and saw that one of the big issues is a disconnection between social assistance and job placement systems. UNDP supported establishing links between the two, so people who are receiving social assistance also have chances at job offers and are helped transition to jobs. And UNDP brokered partnerships with private companies to provide internships and apprenticeships to provide practical skills demanded by the labor market.

This is just one example of the broader SDG issue. We need tools to make this type of complexity manageable, and we are actively designing and testing them:

1) Tools to zoom out and zoom in

System Dynamic Map maps out all problems and interactions between them. For instance, youth employment is linked not only with enrollment to universities, but also with skills for employment, economic development policies, and accessibility of transport. The whole system map is a spaghetti of interactions. You can easily spot “economic”, “social” or “environmental” corners and dig in for more details, or you can zoom out and see the overall picture. Such maps are also a great tool for building partnerships. A similar map in Tajikistan revealed one village self-support group of migrant wives left behind organized a community kindergarten to free up time for business.

2) Better tools for Sense Making and Foresight

In a complex system, you cannot really categorize things into neat boxes and use “best practices”. Regular indicators only tell part of the story. You need triangulation, i.e. looking at the same issue from different perspectives, combining qualitative and quantitative data.  Micronarratives, which combines narrative with quantitative description, is one example of sense-making tools we are using. As testified by stories told by the Roma, the main reason to embark on an uncertain – and often unregulated – migration to the EU is the search for mere survival in terms of income and physical and emotional security. Those who returned from the EU feel rejected and alienated with limited support for reintegration. Mere data from our recent study could not tell us these nuances.

Equipped with these tools, we can better handle the enormous complexity of SDGs. We could go from unmanageable complexity—everything is related to everything—to well-informed Theories of Change.


Originally posted at UNDP Europe and Central Asia blog.

0. Making sense of Elephant

Hokusai’s Blind Men Examining an ElephantAn Elephant in the Dark

Some Hindus have an elephant to show.
No one here has ever seen an elephant.
They bring it at night to a dark room.

One by one, we go in the dark and come out
saying how we experience the animal.
One of us happens to touch the trunk.
A water-pipe kind of creature.

Another, the ear. A very strong, always moving
back and forth, fan-animal. Another, the leg.
I find it still, like a column on a temple.

Another touches the curved back.
A leathery throne. Another the cleverest,
feels the tusk. A rounded sword made of porcelain.
He is proud of his description.

Each of us touches one place
and understands the whole that way.
The palm and the fingers feeling in the dark
are how the senses explore the reality of the elephant.

If each of us held a candle there,
and if we went in together, we could see it.