An example of great, no-clickbait journalism

This Financial Times headline and the lede:

“Uber IPO woes stem from a lack of innovation”

“Replacing Travis Kalanick as chief sapped the company’s relentless drive”

And then the article begins:

“When Uber hired Dara Khosrowshahi in 2017, the company board was solving the wrong problem.”

Because its readers have paid for access online, or have bought its paper in print, its editors have no incentive to resort to the sort of attention-thuggery that most of the internet now relies on.

I’m no fan of a paywall-ed internet, but the ad-filled eyeball-grabbing model isn’t working.

“Can caffeine improve your exercise performance?”

Yes. Though not that much.


“During waking hours, adenosine slows down brain activity and results in feelings of fatigue. When we have caffeine, the caffeine binds to the adenosine receptors and has the opposite effect of adenosine. It reduces fatigue and our perception of effort (for example, how hard it feels to perform an exercise).”

How much?

“Experts believe caffeine doses between 3 and 6 mg/kg are needed to improve performance. That’s 210 to 420mg for a 70kg person, or about two cups of coffee.”

“Those who respond most strongly to caffeine might see improvements of around 16%, but this is unusual. For the average person, improvements will likely be between about 2% and 6%.”


“Experts believe caffeine doses between 3 and 6 mg/kg are needed to improve performance. That’s 210 to 420mg for a 70kg person, or about two cups of coffee.”

Notes from 21 Lessons for the 21st Century

Recently read Yuval Noah Harari’s book 21 Lessons for the 21st Century (he is also the writer of Sapiens and Homo Deus). Part of it on the Kindle, part on audio.

The essence of the book is that our models for making sense of the world are no longer adequate in the current age and of complexity, and that we have a finite amount of time to find and adopt new models.

Our present ways of organising society through religion, nationalism, fascism, capitalism worked fine when the impact of such societies was something humans could imagine. The impact of the actions of a local business are limited to the area it serves, of a group of villagers to the bounds of the village, a king’s decisions to his kingdom, and so on. But multi-national corporations, AI systems, the internet and social media are entities in a society with global reach and whose actions have repercussions far beyond the point of decision making. Calling Apple an American company makes as much sense now as calling an AI system Buddhist.

The underlying message in most lessons is that humans have failed to properly identify and price externalities. We have evolved to use extremely complex, global systems but still view their costs through a narrow lens with little consideration of second and third order impacts.

For example. We see Uber mostly as a city-specific public vs private transportation issue, with its impact on local congestion and on taxicab drivers. But its popularity also means people from outside the city will migrate to this new job opportunity. That it may skew car sales in favour of one manufacturer as Uber cuts a financing deal with one of them, causing plant shutdowns elsewhere in the country. That a large pool of people could be rendered unemployed if Uber decides a city is not profitable. In fact there is an even bigger risk of unemployment with the inevitable move to self-driving vehicles. Then there is the impact of mapping data and user ride data owned by foreign companies with no incentive or obligation to provide it to the city itself. Of user data being stored outside the country if it is to be used along with other global ride data to improve routing and others algorithms.

Consequently, problems seem intractable because we present ourselves with narrow choices, neither of which are palatable to all parties involved. Also, solving one problem throws up another, then another in turn, just like optimising a traffic system too narrowly creates flying bottlenecks.

The writer presents what is effectively trans-humanism as a model to tackle problems such as income inequality, immigration, ecological destruction and climate change. He suggests, as the final lesson, mindful meditation such as Vipassana to better see the reality of the world and its systems. He’s right, but it’s going to take a lot of de-conditioning to get there: individuals who take such an approach in a world rooted in individualism and simplistic systems will quickly suffer. As David Cain says in one of the blog posts on his blog Raptitude, “Your Lifestyle Has Already Been Designed” for you. It’s a mutation in thinking that by its nature will quickly die out. What is unknown is how we will get a whole community to change its thinking en masse.


In response over text to a friend that shared this:

I think we have to factor in exponential growth and then some. For example the Bombay metro should look to 100x, not 10x today’s local train capacity.

That seems to be how the Chinese think. What they do defies imagination literally. If one were to imagine how many daily Beijing Shanghai high speed trains existed I’d have said 3 or 4. Never 35.

We are thinking of one ‘bullet’ train between Bombay and Ahmedabad five years from now. Why not plan for one every 20 min like a Bombay-Poona bus from day one?

Public infrastructure is almost always a case of build it and they will come. Instead we build private infra unscrupulously and with abandon, then pressurising politically some bare bones public infra to be built, the design bad scope of which is beholden to and limited by what exists. This is effectively then privatisation of public infra. Case in point is all of the excess real estate in Poona built right up to the literal edge of the national Poona Bangalore highway.

Speaking of China, I’d read this on a blog post:

“That’s the distance from Hong Kong to Beijing, and if you’re on a train that cruises at 306km/h, you can leave at 8:05AM and arrive one minute past five in the afternoon. The train has a number and a Wikipedia entry: G80”

Packed, uncomfortable train rides to undergrad college had had a few of us thinking about the impact of near-universal high speed trains in India: it wouldn’t just mean shorter travel times, but changes beyond that as well. There would be no sleeper cars. Railway food and support infra would be therefore vastly more efficient. Shorter times = more predictable arrival times => you could book hotels + trains together like you do with flights today. Housing changes altogether. Many more people could live in, say Aurangabad and Ahmedabad and even Belgav and commute to Bomay in a hour or so. Pressure on flights reduces. It’s crazy.

Blog that links to songs I like

Just started publishing Radio Station In My Head, where I’ll do exactly what the post title says, along with some basic info about the song. The mobile WordPress app makes this super-easy. And the idea is to mostly leave out songs that are mainstream. Follow it if you’re the sort that likes Bollywood songs from the early 60s to the mid 80s.