Reading List, August 2017

1. Simon Penner, Another point of view, in Status 451, 3 August 2017. [Online]: https://status451.com/2017/08/04/another-point-of-view/

   This is not a post for people who have already made up their minds. If, however, you are the type who can wrestle with an alternate point of view, or even acknowledge that privilege is not necessarily universal and is fluid within social and political contexts, do read this post about the challenges faced by a Canadian Midwesterner in Politically Correct Liberal California Bay Area.

   The reality is that software engineering jobs are not magical privileges, not lifetime keys to the lands of luxury and riches. They are hard, important work taking smart, skilled, and highly practiced people a lifetime to master. Most people here will have struggles, because hard things make you struggle. Focusing on a very small subset of those struggles is a very curious definition of “fairness”. Respectfully: Check your goddamn privilege.

2. Alex Mayyasi, The invention of Pad Thai, in Priceonomics, 28 October 2015. [Online]: https://priceonomics.com/the-invention-of-pad-thai/

   Answering the question, how long does it take to create a cuisine.

   Not long: less than fifty years, judging by past experience.

   This is a fascinating tale of how the Pad Thai was created by executive decree, in an effort to make Thailand more ‘cultured’ and to prevent colonisation.

3. Rachel Kolb and Dakota McCoy, Gene-editing tool raises questions about what is ‘disease’, in San Francisco Chronicle, 3 August 2017. [Online]: http://www.sfchronicle.com/opinion/openforum/article/Gene-editing-tool-raises-questions-about-what-is-11732894.php

   What are the ethics of gene manipulation. Where do we draw the line between preventing disease and interfering with the natural course of life? In this op-ed, the authors point out that while most people would consider being deaf as something that should be cured if possible, there is a rich culture that exists between deaf people, such as sophisticated sign languages. Curing deafness at the embryonic stage could easily destroy this culture.

   This article makes the list just as a reminder that we engineers and scientists need to consider social, cultural and ethical implications of our decisions, not just the scientific impact.

4. Ken White, Lawsplainer: How federal grand juries work, part one, in Popehat, 7 August 2017. [Online]: https://www.popehat.com/2017/08/07/lawsplainer-how-federal-grand-juries-work-part-one/

   Ken White’s a former federal prosecutor who has now moved to private practice. He’s also very funny, and as evidence, here’s an article explaining how grand juries work, in all their detail. Recommended for sounding more intelligent during your next conversation.

5. Simon Penner, Judging things by their side-effects, in Status 451, 6 August 2017. [Online]: https://status451.com/2017/08/06/judging-things-by-their-side-effects/

   Again, I would not recommend this article if you’ve already made up your mind. Simon talks about the infamous Google memo. His point, which is something to be considered is that the criticism around the memo wasn’t that the memo was just plain wrong, instead it focussed on the effect – that it created a hostile work environment. In this article, Simon argues that by focussing on the effect, Google and others are implicitly acknowledging the contents of the memo as true, otherwise they could simply say, ‘That’s wrong’, and move on.

   If we adopt a social norm that the side effects of speech are more important than its truth value, another problem arises: we can no longer trust the truth value of anything by default, ever.

   In 2003, the Bush administration knew damn well that there were no WMDs in Iraq. However, they really truly sincerely believed, according to whatever weird messed up metric they were using, that not bringing war to Iraq would cause massive harm and danger to many people, so that nevertheless we should lie about there being WMDs in Iraq.

   Simon also points out that the reaction to the memo – dropping a banhammer – is harmful because it implicitly assumes that there is a single actor, and that firing one engineer will prevent such memos from being written and circulated in the future.

   Or, to use an analogy the left might like: “If we blow up terrorists, there will be no terrorists to blow us up” is an insane and ridiculous statement because, among other things, it assumes that blowing up terrorists will not radicalize any new terrorists.

6. Scott Aaronson, The Kolmogorov option, in Shtetl-Optimized, 8 August 2017. [Online]: http://www.scottaaronson.com/blog/?p=3376

   Talks about the Soviet mathematician Kolmogorov, who founded a significant portion of probability theory as we know it today, and his choices to not pick fights that he could not win against the Soviet Union.

   In fact it’s often because they fear you might be right that the authorities see no choice but to make an example of you, lest the heresy spread more widely. One corollary is that the more reasonably and cogently you make your case, the more you force the authorities’ hand.

   To an Inquisitor, “good heretic” doesn’t parse any better than “round square,” and the very utterance of such a phrase is an invitation to mockery. If the Inquisition had had Twitter, its favorite sentence would be “I can’t even.”

   Does this mean that, like Winston Smith, the iconoclast simply must accept that 2+2=5, and that a boot will stamp on a human face forever? No, not at all. Instead the iconoclast can choose what I think of as the Kolmogorov option. This is where you build up fortresses of truth in places the ideological authorities don’t particularly understand or care about, like pure math, or butterfly taxonomy, or irregular verbs. You avoid a direct assault on any beliefs your culture considers necessary for it to operate.

7. Scott Aaronson, What I believe II (ft. Sarah Constantin and Stacey Jeffery), in Shtetl-Optimized, 15 August 2017. [Online]: http://www.scottaaronson.com/blog/?p=3389

   A follow-up to ‘The Kolmogorov Option’ that addresses the elephant in the room.

8. Scott Alexander, Contra-Grant on exaggerated differences, in Slate Star Codex, 7 August 2017. [Online]: http://slatestarcodex.com/2017/08/07/contra-grant-on-exaggerated-differences/

   Alexander takes on a LinkedIn post by Adam Grant. Do read till the very end, including the comments section in which Alexander and Grant continue the debate.

   Why am I sharing this? Because it is refreshing to see in today’s day and age, two people having a debate based purely on facts and their interpretations of data without resorting to ad-homenim attacks.

9. Fredrik deBoer, Why selection bias is the most powerful force in education, in the ANOVA, 29 March 2017. [Online]: https://fredrikdeboer.com/2017/03/29/why-selection-bias-is-the-most-powerful-force-in-education/

   Fredrick points out that the outcomes of most schools are decided simply by the students they accept. So, schools that select bright, motivated students end up with students that perform better. The most amazing data in this comes from Fredrik’s thesis, where the performance of students post college goes up by a fixed amount (essentially changing mostly the intercept of the graph, not the slope). This is a fascinating reading on how we need to be careful about selection bias whenever discussing education and student performance.

10. Al Williams, You know you can do that with a 555, in HackADay, 10 August 2017. [Online]: http://hackaday.com/2017/08/10/you-know-you-can-do-that-with-a-555/

    I had a conversation with a friend recently, in which I pointed out that the task he wanted to accomplish could be done with a 555. This hack goes even further, making logic gates and op-amps with a 555. Fascinating reading.

11. Steven Dufresne, Decoding Enigma using a neural network, in HackADay, 11 August 2017. [Online]: http://hackaday.com/2017/08/11/decoding-enigma-using-a-neural-network/

    Using an LSTM neural network for encoding and decoding the Enigma, which leaves me asking, Why?

    What next, neural networks for LQR? Neural networks for displaying text on a screen? Neural networks for encoding and decoding key-presses on a keyboard? Neural networks functioning as an adder?

12. Eric Berger, SpaceX is launching a supercomputer to the International Space Station, in Ars Technica, 11 August 2017. [Online]: https://arstechnica.com/science/2017/08/spacex-is-launching-a-supercomputer-to-the-international-space-station/

    The headline’s quite a misdirection. Basically, HP-Enterprise will be installing a 1 TFLOP/s computer in the international space station. Not a supercomputer by any modern standards, but it is the most powerful computer sent to space.

    The trouble with space is with all the high-energy radiation. This radiation can get inside electronics and flip bits in all sorts of crazy ways, which is why the space shuttle used core memory, a technology from the 1960s. This study is meant to see how ordinary computers fare in the radiation-rich environment of space.

13. Lee Rowland, Controversy is curriculum, in Inside Sources, 25 April 2017. [Online]: https://www.aclu.org/blog/speak-freely/we-all-need-defend-speech-we-hate

    Lee Rowland is a senior staff attorney at the ACLU. In this post, she argues that free speech, and the ability to confront and counter opinions that we disagree with is a skill that colleges are failing to teach their students. She rightly points out that hate speech, and speech that affects vulnerable students should be stopped, but that recent events show that colleges are failing students, be it the use of violence, outrage, or threats to shut down speech that is protected by the first amendment, but with which students disagree.

    These incidents have not shut down a single bad idea. To the contrary, they’ve given their opponents’ ideas credence by adding the power of martyrdom. When you choose censorship as your substantive argument, you lose the debate. Because none of us are the wiser about the better world those protesting students want to see instead of telling us, they silenced others. In curricular terms: They didn’t do the assignment.

    When students leave the nest of higher education, they will immediately be thrust into a rough-and-tumble world filled with things many of us don’t want to see: racism, sexism, ableism, cruelty. But we all know these things won’t go away if we close our eyes. We need a next generation of students trained to take the deep breath, open their eyes, and change that world with their words and ideas.

14. Jonathan Haidt, Why so many Americans don’t want social justice and don’t trust science, in Boyarsky Lecture in Law, Medicine and Ethics, Duke University, 3 July 2013. [Online]: https://youtu.be/b86dzTFJbkc

    Technically, not a book or an article, but useful nevertheless. Given the sheer divergence of political discourse in the USA and the massive echo chambers that people live in, this, in my opinion, is a useful starting point to understand the other side.

    To push further, this video also gives me some insights into why we Indians love talking about politics and listening to the other side, and why and how this has changed recently. (Spoiler alert: I blame the American import of ‘us-vs-them’ ideology that is discussed in this video.)

15. Dan Lyons, New tech start-up bubble, in Open Source Leadership Summit, February 2017. Published by Coding Tech, 27 July 2017. [Online]: https://youtu.be/G7vrCpWbmDw

    Again, a video, but this one is totally hilarious, and short, at just 20 minutes. The story of a 50+ year old tech reporter inside a startup dominated by millenials.

    Also made me realise that I’m essentially a 50 year old man at heart.

16. Denise Cummins, Why the STEM gender gap is overblown, in Making Sen$e, 17 April 2015. [Online]: http://www.pbs.org/newshour/making-sense/truth-women-stem-careers/

    The short answer, the gender gap varies across fields, with some fields having more men (notably computer science and engineering) and some fields having more women (social sciences and psychology). Maths, physical sciences, and biosciences have a near 50% gender ratio.

    I shall not say any more even to merely summarise the article. It is impossible to say more without the appearance of making a political statement these days. Even two years ago, this article came with a disclaimer from the editors just for people who choose to get angry about headlines without reading the entire article.

17. Nick Srnicek, We need to nationalise Google, Facebook and Amazon. Here’s why, in The Guardian, 30 August 2017. [Online]: https://www.theguardian.com/commentisfree/2017/aug/30/nationalise-google-facebook-amazon-data-monopoly-platform-public-interest

    Google, Facebook, and Amazon are almost monopolies that are increasingly seeking to inject themselves into all aspects of our lives. Srnicek argues that these companies should be treated the same way as utilities or railways and nationalised to serve the public interest.

    This article makes it to the list simply as an acknowledgement of a problem. There’s no way this can actually happen, there’s a huge counter-argument for government censorship and big-brother, along with the all-important question of which government gets to control the internet. No, we’re stuck with the status quo for now.

18. Matthew Prince, Why we terminated Daily Stormer, in Cloudflare blog, 16 August 2017. [Online]: https://blog.cloudflare.com/why-we-terminated-daily-stormer/

    This is an excellent article to read because it expresses the tension between the first amendment and the right of private citizens to have an opinion or a stance, especially if the citizen in question owns a company that handles about a tenth of the web traffic in the world.

    Here’s the story. After the Nazi demonstrations in Charlottesville, Google and GoDaddy decided to veto the Daily Stormer, an alt-right website that was ostensibly the brains behind the demonstration. The website went off to a Russian domain, then the dark web, apparently shit-talked how Cloudflare supported them. This made the founder of Cloudflare angry, and he pulled the plug on the website, opening it up to a DDoS attack that would eventually take it down.

    The tension here is that a few companies control almost all of the internet, effectively making it an oligopoly. I know that I’m simplifying the debate here, but the details are irrelevant to the point being made. The point is that these companies can engage in censorship, such censorship is legally justified because they are not a government but private entities. This, however, is dangerous territory filled with slippery slopes. Prince acknowledges all of these issues and the fact that his decision probably sets a very bad precedent.