Mehul Kamdar

Mehul Kamdar

Data scientist, developer, writer, bibliophile.

On Singapore, Indian education, and current computing hardware

It is in part the difference between the more intense and exacting Sinic culture of East Asia and the less intense and less demanding values of Hindu culture of South and Southeast Asia that accounts for the differences in industrial progress between Eastern and Southern Asia.

Lee Kuan Yew, Foundation lecture at Cambridge on 8th November, 1971

  1. The True Story of Lee Kuan Yew’s Singapore.

    “The PAP casts Operation Coldstore as a harsh, but necessary action that crippled a violent Communist Party of Malaya. In contrast, the opposition identifies Coldstore as the moment political pluralism was wiped out on the island—Singapore’s original sin.”

  2. On the topic of Singapore, Lee Kuan Yew’s Foundation lecture at Cambridge in 1971 (pdf) is a must-read that practically every Singaporean knows very well. But in case you haven’t, maybe now is as good a time as any. A few excerpts, keeping in mind that this is from 1971 in the early years of the Singaporean development project:

    Why can [East Asia] not organise themselves, maintain effective administrations, ensure political stability and provide proper sanitation, clean potable ater, reliable electric supply? … Why have they not widened the base ad raised the levels of education of their people, and trained them in industrial techniques and technology? Why have they not made better use of machinery and equipment they have bought, on soft-loans, from the West, and made the economic and social progress which would have made them equal to the West?

    The first contradiction South and Southeast Asian societies face is how to revamp their value systems and culture patterns, to meet the needs of an industrial society. It is not possible to move from the agricultural economies of Asia, equivalent to those of the 15th and 16th century Western Europe, into the “technetronic” era the Americans have created and named, without jettisoning parts of the value systems and culture patterns of the past. Some of them inhibit instead of encourage punctuality, work discipline, the desire to increase production norms, the acquisition of scientific knowledge and engineering techniques.

    Rapid acquisition of knowledge in the sciences, industrial know-how, management expertise, marketing techniques, and higher manipulative skills are only possible if the people are intense in the pursuit of these goals. They do not go with a relaxed culture, in which fatalism is a tranquiliser for anxiety over failure.

    Meanwhile, in the less deveoped countries of South and Southeast Asia, disillusionment and near despair has set in because of their failure to make the grade. The despondency is all the greater, because of earlier beliefs that growth was the natural and effortless result of the transference of capital equipment, and work techniques, from the developed to the less developed. This has not happened because there never was this easy ride towards the industrial society.

  3. This podcast episode with Karthik Muralidharan on The Seen and the Unseen with Amit Varma (AV) is great. It’s a long podcast (over 3 hours). I wish there were a transcript but if you have the free time, it’s well worth tuning in. Karthik Muralidharan (KM) is an economic professor at UCSD who’s written extensively on development and the Indian education system. A preview of some of the concepts explored:

    ~ 2 hr 3 minute mark: This was the most important thing for me that is missed by AV and by most educated Indian that I know. KM notes that perhaps there is another podcast to be done on Indian federalism. That the Indian democratic system is so federal and each ruling party is obsessed with curbing state and local power is the most urgent problem facing us today. And similary around the 2 hr 30 minute mark: KM notes that the entire culture in Indian government is one of lack of trust. Now with that said, some highlights from the rest -

    ~37 minute mark: AV makes the currently very commonly believed idea (especially among the Mumbai crowd) that India has suffered in the past by considering the ‘profit motive’ to be something bad. And argues that to him “profit is a driver of benevolence” since you can only make a profit by providing value to someone else’s life. KM correctly responds that “in a world where you have the level of abject poverty that we have, the market solution is to essentially leave those people out completely.” Or, as I like to put it, corporations (as proxies for the ‘profit motive’) are built to make money, and there is no easier way to make money than to find rich people and give them what they want (which is invariably more money and power - not in absolute terms, but in relative terms).

    ~40 minute mark: KM notes that “what makes India an outlier in human history is that we are the only country that had universal adult franchise and democracy at such a low level of per capita income.” And “where democracy has failed in India is not the idea of democracy but that the level of aggregation is so high, that you’re expressing your preferences over so many dimensions of what you care about as a citizen in one vote. So decentralizing is a way to unbundle the vote and allow your local vote to focus much more on issues of service delivery and build democratic accountability.”

    ~57 minute mark: KM notes that the education policy in the first couple of decades post-independence can be characterized by neglect of primary education. In the past 30 years, primary education enrollment (and other input-based measures) increased dramatically but the translation of that spending into outcomes has been remarkably poor - the two main reasons being i) weak governance, and ii) sub-optimal pedagogy. Why is all this spending not translating into better outcomes? The budget has four main line items - i) teacher salaries, ii) school infrastructure, iii) mid-day meals & school grants, iv) teacher training & quality upgradation investments. None of these four categories appear to be having much impact on learning outcomes.

    ~1 hr 17 minute mark: Two robust interventions at the public schooling level that have been found to be effective: i) governance (e.g., basic monitoring to reduce teacher absence, some performance based payment structure for teachers), ii) locally hired contract teachers (often someone local in a village with only a 10th-12th grade education & no teacher training) is at least as effective as trained, highly paid teachers at the primary school level.

    ~1 hr 25 minute mark: KM notes the common issue that the Indian education system is primarily about sorting rather than learning. However, he doesn’t seem to discuss how or why this is different in other education systems around the world - practically every education system around the world will be criticized on the same basis. But he does note some caution in simply dismissing objective testing at a state or national level, which is good. He described only a couple of reasons for this, but there are many more.

    ~1 hr 50 minute mark: KM gives his views on the New Education Policy. He generally likes it (partly because he has had input in it) for its move away from just inputs (as was the case with the Right to Education Act 2009) to outcomes based measures. However, he doesn’t like that there isn’t enough discussion of cost-effectiveness.

    ~1 hr 56 minute mark: I thought this was key. AV continues making similar (understandable) mistakes - this time that a democratic system ends up implying the focus is more on inputs (short-termism). Most educated Indians I have known seem to think this & end up drawing one of two conclusions - i) we need a Singaporean state (but the tragedy is that our size doesn’t allow it), or ii) this is the price of democracy, and we have to mostly unfortunately accept it unless we want to devolve into an authoritarian state. KM correctly points out that this is not a useful framing of the issue. The question is more of decentralization under a democratic system. And I think this is clearly one of the issues India faces today - the Chinese party can in fact be quite decentralized in its operations, while if anything, the Indian ruling parties keep trying to move towards a centralized machinery (something, unfortunately, our constitution has always allowed).

  4. The can’t-miss, hard-to-beat Ben Thompson @ Stratechery writes about Nvidia’s acqusition of ARM. This acquisition is big. Pay attention. Couple of excerpts:

    That neutrality is gone under Nvidia ownership, at least in theory: now Nvidia has early access to ARM designs, and the ability to push changes in the ARM ISA; to put it another way, Nvidia is now a supplier for many of the companies it competes with, which is a particular problem given Nvidia’s reputation for both pushing up prices and being difficult to partner with. Here again Apple works as an analogy: the iPhone maker is notorious for holding the line on margins, prioritizing its own interests, and being litigious about intellectual property; Nvidia has the same sort of reputation. So does Intel, for that matter; the common characteristic is being vertically integrated.

    Nvidia is not setting out to be a partner, someone that gets along with everyone in exchange for a couple of pennies in licensing fees. Quite the opposite: Huang wants to own it all.

  5. Jane Street, the hedge fund, launched a new podcast called Signals & Threads about hardware engineering. This episode on Programmable Hardware is great - discusses FPGAs, tools for designing FPGAs and the current state-of-the-art.