Thoughts this week:

Career stategy

Thiel isn't decisive on the topic. Is the definite-optimist view is the dominant approach to candidacy in the grand marketplace of talent today?

Kumon

Kumon franchises are cheap. The branding and rep is good. Tutoring is a very attractive market in general and kumon makes it easier for the teachers. But is it ethical, I wonder? To me it's ethical if it delivers value to the students. A caveat is that it seemed cruel the kind of mind-numbing maths done by my classmates as a kid who attended Kumon.

A study with very small sample size) says 'that there may be a significant relationship between participation in the Kumon programme and development in computation skills (p = 0.053), but not with development in mathematical reasoning skills (p = 0.867)'

Going by that alone, it would seem short-term experimental studies don't explain the small-sample (2 out of 2) size correlation between Kumon attendance as kids and extreme adult success I see in my friends today. A stackexchange-esque Q&A suggests no further effects are suggested by other experimental Kumon literature

Political psychology

If human rights were reframed as entitlements, and laws that protect those contrasted with laws that protect discretions, those two having obvious tensions, I wonder what impact that would have on human rights law reform. Branding is influential.

Effective Altruism politics

Based on Sam Deere, Effective Altruism affiliate and former ALP staffer: tractability of political ideas for conversion to legislation comes down to: novelty, high impact, cost-effectiveness, budgetary considerations, interaction with other policies, popular, ideological and strategic/tactical considerations.

A hereustic for lobbying is to focus on Ministers, not departments, other politicians, yada yada...

Nara

I have heard of an exotic Namibian plant, crowned with thorns that grows in the desert called Nara that bears a fruit which when dried, tastes like chocolate but is highly nutritious. I'm skeptical. It sounds too good to be true. I wonder why it isn't a common treat around the world now. If the rumours are true, I would love some to grow in the Australian outback, biosecurity permitting.

Is data science a profitable industry?

Anyway, it's super easy to transmit machine learning how-to online, it's the most popular class at Stanford. Can I get a 'market efficiency'? It's not long till we automate the basic tasks since machine learning is an empirical field too, so it can be subject to it's own learning in polynomial time. I reckon people neglect this (and thus, the market efficiency is quickly gained) because there is little public education to self taught programmers about tihs kind of thing, outside of boring lectures

If you're a data science training provider - yes.

If you're employing data scientists in a data rich operating environment in 2016 - yes.

If you're anyone else...I doubt it.

I'm extremely skeptical about the data science boom. Just because a field is valuable doesn't mean companies around them are strategically places to capture that value in a market for their owners.

Data science currently operates as:

(1) Big data products - on marketplaces like Amazon Web Services where machine learning algorithms are available via the cloud.

(2) Product - offline data analysis automation tools

(3) Service - manually doing (1) or (2)

Amazon currently dominates (1). Microsoft is a close 2nd. Their delivery is on point. They've basically created a platform for people to trade their knowledge about data science as algorithms. Since there are a finite number of algorithms, that can be combinatorially generated and tagged for particular applications, the only real challenge is creating a system for triaging a user's needs into which particular algorithmic application.

If there is to be an big money to be made in this space by new players, it will be solving that problem.

The issue of algorithm generation for statistical analysis should not be confused with the sophisticated tasks of software and application development. The former requires little creativity, while the latter can utilise immense creativity. I say that as someone who's forte is data analysis, rather than software.

As for (2), there is likely to be high efficiency in a market between cloud based algorithms and algorithms implemented offline due to extreme low barriers to entry. Basically those first person in with a good method of translating those algorithms offline, surmounting potential legal hazards, and scaling up (no trivial tasks) will make a quick buck. Though, these are problems that I can define. If I can define them, the big names probably already have and are working on solutions for them. You're out of luck, garage entrepreneurs.

Now (3), the one lay people think of when they think data science, and the one aspiring data scientists entering Kaggle competitions and hopping onto data camp think of. This is a highly commodifiable area, subject to total automation, and outsourcing. There may be some money to be made here as the ubiquity of data rises and you find loyal, computer illiterate clients. However, you'll be picking up scraps the same way that web design as a profitable avenue for freelancing oddballs to make money worked and continues to work - by essentially ripping people off who don't know better, when absurdly user friendly DIY web design options are a reality or the guy down the road is a million times better than you, your client just doesn't know better. Sure, it may be profitable, but ethically it's questionable.

First principles

Elon Musk cites first principle thinking in physics as a key to identifying neglected market opportunities. Can someone give me an example of how it may work in that application?

Social skills

Simply reframing ''approach anxiety'' to the crude, macho 'bitch butterflies'' has done wonders to dampen the phenomenon. I wonder what if that formula could dampen other anxieties...

Concept learning

I wonder what it is like to have genius level verbal abstract reasoning: e.g. 2SD+for instance as reported by the usual neuropsychological tests. The Wechsler Adult Intelligence Scale (WAIS) 'Similarities' subtest measures verbal abstract reasoning or ‘concept learning' (see concept learning on Wikipedia and the Edutechwiki.. Subjects are asked to say how two seemingly dissimilar items might in fact be similar.

When an average person talks to someone at just 30 points of IQ less than average (IQ 70 - the cut off point for intellectual disability), that experience may be comparable to a genius (130) talking to an average person. When it comes to particular subscales that relate to ‘’understand’’ such as concept learning, this may not in fact match up with IQ. It’s conceivable there may be savants with incredible high concept formation with incredibly low IQ’s. This presents an additional layer of complexity to a hypothetical interaction between a low IQ concept savant and high IQ person concept lay-person that I can’t even simulate, mentally. So, I’m opening it to the floor for a fun thought experiment.