Occasional Reading Post #8

(I’ve given up on calling these weekly reading posts because they’re clearly not weekly)

Random Reads

Floating Utopias. Somehow these libertarian community projects never quite pan out.
My condolences, you’re now the maintainer of a popular open source project
. I relate to this a lot.
The Really Big One. Your regular reminder that the planet earth has almost infinite means to destroy us at its disposal. Yet another reason not to relocate to San Francisco.
The Other Side Is Not Dumb – good piece on the importance of understanding your political opponents
Creativity: A Crime of Passion – “Be creative, but not TOO creative”. On the societal disincentives for creativity
3 reasons the American Revolution was a mistake. Yeah, um, I don’t agree with this piece at all. I mean I don’t think much of the USA, but I also don’t think much of the UK right now.
How I Cracked a Keylogger and Ended Up in Someone’s Inbox
– short piece on reverse engineering a key logger.
The Principia Misanthropica – Ribbonfarm’s… rather idiosyncratic brief history of everything. Amusing read, tells you more about Ribbonfarm’s philosophical biases than actual history but that’s OK.
The BBC problem – on the problems that the BBC’s “impartiality” requirement creates, and the resulting false balance. Specifically in the light of brexit but true more generally
Alberta, Canada is the proud owner of the largest man-made pyramid on the planet – file this one under weird artifacts of industrial civilization
AAVE is not SE with mistakes – a discussion of the grammatical features of AAVE (African American Vernacular English) and the fact that it’s a distinct dialect in its own right and not just speaking badly.
Putin’s Puppet – on Putin’s funding of the campaigns of various western politicians, and Trump’s relationship with Russia
Mental Models I Find Repeatedly Useful – a bunch of useful concepts. I’m not sure I would have labelled most of them mental models. I was familiar with most of them, but there were a few useful ones I hadn’t encountered (e.g. Best Alternative to Negotiated Agreement

Selected Reads

Just one this time: Taming the stream roller: How to communicate compassionately with non native English speakers. I could have really used this advice before spending a week in Mexico then speaking at a Russian conference. I’ll try to remember it at Europython.

Books

No new wish list items recently.

I’ve been rereading What the best college teachers do as part of an attempt to improve my training courses more.

How to navigate goal directed learning

As I’ve written about before (here and here), I’m a big fan of goal-directed learning (this doesn’t appear to be a standard term, but I think it’s a reasonably descriptive one).

The core idea of goal directed learning is:

Pick a goal
Learn the smallest amount required to achieve this goal
Stop once you have achieved the goal

The question I’d like to ask is this: How do we do step 1? What is a good goal for learning?

There’s a wide variety of goals with different strengths, but I think there are at least two different types of goal that are different enough that it’s worth considering separately.

Roughly, these are:

Goals which will expand your capabilities
Goals which will reinforce and flesh out your current capabilities

You might call these far and near goals.

Obviously this is actually a spectrum and I’ve introduced a false binary in order to make a point, but in this case I think that binary is useful for thinking about how to choose goals.

The benefit of the first type is reasonably clear: You’re learning to do more, which is inherently a good thing.

But it’s important to not neglect the second type: If you pick a goal, then pick another goal that is unrelated to your previous one once that is finished, you will end up learning much more slowly in the long run.

Because the objective of learning, goal-directed or not, isn’t just to fill your head with a grab bag of facts: You want to build a solid foundation of knowledge which both provides you with capabilities but also improves your ability to acquire new capabilities. Fragile, disconnected, knowledge does the former but not the latter.

What sort of goals help you build this knowledge?

There are probably a variety that do, but I think the *best* goals for this are ones that:

Are just beyond your current capabilities.
Combine two (ideally not much more than two) existing capabilities that you haven’t used before.

The first is important because it means that your work acts as a test of whether your capabilities are really what you think they are, which is vital for cementing your understanding of it. Generally you don’t understand something fully until you’re confused about the next thing that uses it.

The second is important because it lets you put your knowledge in context. Again, this improves your understanding of knowledge by making you examine its boundaries and seeing it in a new light from the other side.

What’s the right amount of time to spend on each type of goal?

Well, I don’t know. It depends on what you’re trying to do: If you’re just trying to get things done, you might not need any far goals – if everything you need to do is a near mode goal then we might as well just do that. We won’t learn that much, but everything we learn will be useful.

Conversely, if you don’t know anything about the problem domain then *everything* is a far goal, so you don’t really have the option of choosing a near one.

But when you’re starting to get familiar with the problem domain but are still trying to learn, you’ll need a mix of the two. In that case, I think the following rules are probably a good heuristic for goal selection:

Spend more time on near goals than far goals
Try to make sure you regularly do far goals
After any far goal, do some near goals related to it

This will keep you moving and effective, force you to expand your capabilities, but will mostly ensure you have a solid and applicable knowledge base.

Weekly Reading Post #7

A relatively small week due to being in Mexico for almost all of it:

Disturbances #9: Cryolite to Cryoconite – Jay does a tinyletter about dust (and thus, environmental collapse). This is the latest in the series.
Where Geek Girls Come From – An article on how having an adult make you realise that you’re “allowed” to be a geek can be essential when society tells you otherwise. TBH I basically recommend subscribing to Siderea’s blog, because I suspect > 60% of what she writes will end up here in random or selected.
Press embargoes – a threat from the shadows – An article on how the terms of some journals can be a major problem for open data. A bit inside baseball for me.
Tire Pressure – Stop Guessing And Read The Science – An article on the implications of tire pressure for cyclists. I’m not really a serious enough cyclist to benefit from this.
The Brexit debate is far from over: there will have to be a further vote – I don’t really remember reading this. It’s just where the ping happened to fall on a bunch of Brexit related reading.
Why Google Stores Billions of Lines of Code in a Single Repository – I’m a big fan of the monorepo approach. My limited experience of this at Google was that it worked pretty well, though being able to do local checkouts of parts of the repo was crucial to that for me.

Books

I read Danah Boyd’s “It’s Complicated” on the trip back from Mexico. It’s about teenage experiences of social media (and life more generally). Can recommend. it’s not very long and is quite enlightening. It also reinforces pretty heavily that the way teenagers get treated in the USA kinda sucks (being a teenager elsewhere does too, though differently).

Three thought experiments on majority voting

I present to you three examples in which we employ majority voting between two options (that is, we ask a population “Would you like A or B?” and we choose the option that the largest number of people preferred).

You can and should infer the obvious context for this, but I am not going to comment further on it in this post.

Tea and Cake or Death

Suppose 51% of the population vote for a bill that will result in the death of the remaining 49% of the population.

Questions

Should the country go along with this?
If they decide to, do the remaining 49% have a democratic obligation to accept that?
Does the answer change if it’s 90% and 10%? 99% and 1%?
What if the remaining 49% are merely financially ruined? Moderately inconvenienced?

Pizza or Barbecue

A group of nine friends regularly meet for dinner. Of these, five of them really like pizza and four of them really like barbecue. As good citizens of a democracy, they put this to a vote. Unsurprisingly, this results in them always having pizza.

Questions

Is this fair?
Suppose only four of them really like pizza, and the ninth person changes their mind regularly and thus always gets to decide where they go for dinner. Is that fair?
Suppose the friends in question are tired of putting it to the vote each time and some of them push for a vote to go to a regular meeting place. They put the question “Should we have Pizza or Barbecue for all future group dinners?” to a vote. Pizza wins. Is that fair? Does that answer change if we have the previous 4/4/1 split?

Which president?

Fair warning: This one is by far the most complicated of the three thought experiments.

Our student mathematical society decides to elect a president. There are three candidates, Alex, Kim and Pat. We’ve read all this confusing stuff about voting theory and we can’t really decide what we like except that majority rule is clearly the best for two candidates, so we decide to reduce this to the solved problem. I pick two candidates, we vote between them and the majority winner stays in. We then vote again between them and the third remaining candidate, and the winner of that becomes president.

Note that the student body is roughly equally split between the following three preferences:

Alex, Kim, Pat
Kim, Pat, Alex
Pat, Alex, Kim

As a result, the majority of people prefer Alex to Kim, the majority of people prefer Kim to Pat, and the majority of people prefer Pat to Alex.

Note that:

If I put Alex and Kim together in the first round, then Pat is president because Alex beats Kim then Pat beats Alex.
If I put Alex and Pat together in the first round, then Kim is president because Pat beats Alex, then faces Kim and the majority prefer Kim to Pat
If I put Kim and Pat together in the first round, then Alex is president because Kim beats Pat in the first round and then Alex beats Kim

Questions

How strong is the resulting president’s mandate?
Does the answer to the previous question depend on which round they were in?
Does the answer to the first question depend on how we chose the rounds?
Does the answer change if instead of explicitly voting for three candidates as part of a single batch we instead have a system where a new candidate always runs against the incumbent?

Concluding Statement

Majority vote between two options is often held up as some pinacle of uncontroversial democracy where at least in this case we know what the right answer is, even though voting is complicated in general.

I hope I have convinced you that is not the case.

Weekly Reading Post #6 (Bi-Weekly Edition)

There was very little to report last week so I ended up skipping posting this for a week.

Random Links

Derivatives of Parsing Expression Grammars – This takes the work on parsing with derivatives and applies it to parsing expression grammars as an alternative to packrat parsing. I can’t tell if this is a good idea or not. I was interested in it for generating values given a parsing expression grammar, but PEGs have certain properties that make that hard.
Disgust made us human – about the history of civilization and the role of purity based morality in it. I’m not totally convinced by its thesis but it’s an interesting read.
Singular Value Decomposition Part 1: Perspectives on Linear Algebra – An interesting two part piece for understanding singular value decomposition.
Shuffleboard At McMurdo – a fun piece about how the US Antarctic base is terrible. Idlewords in relatively non-infuriating mode.
When the End of Human Civilization Is Your Day Job – some of the sociological aspects of being a climate scientist and how we’re all fucked.
Handling sub-process hierarchies in Python on Linux, OS X and Windows
Wikipedia article on Titanium Biocompatibility – honestly I was just randomly wikisurfing until I got a nonfiction point.
Long Names are Long – a reasonably decent piece on identifier names and preferring shorter ones.
Quotes From Feminists That Will Make You Rethink Trusting Men’s Rights Activists – it didn’t make me rethink trusting MRAs, but that’s mostly because I already don’t. It provides some interesting context.
Wikipedia on the Broyden–Fletcher–Goldfarb–Shanno algorithm – an optimization algorithm. I was reading about this because of the adversarial examples link below.

Selected Links

Explaining and Harnessing Adversarial Examples – a neat paper about the fragility of neural network models when given things that are just slightly off the typical distribution.
E-Prime language – a linguistic style in which one avoids statements of the form “X is Y” in preference for “X does Z”. This was sent to me in response to my pointing out that arguments about the former tend to be very frustrating and largely go away if you focus on the latter.
How to Grow a Weetabix – an interesting breakdown of the effect of a possible exit from the EU on the British farming industry and landscape, along with a lot of interesting related information about British farming subsidies and their ecological focus.

Books

I’ve been reading Thinking, Fast and Slow this week. It’s come very highly recommended, but to be honest I’m not very convinced by it. Part of the problem is that I distrust a lot of the underlying research.

Elements of Information Theory arrived as a gift from Zack M. Davis. Thanks, Zack!