Why Simple Isn’t by Steve Marshall Friday, February 21, 14

Steve Marshall Web Developer Photograph by David Thompson Friday, February 21, 14 Alright?/Hello Edinburgh! - Steve Marshall - Web developer/wannabe pirate - Care about making the simplest, best solution to problems - Make awesome web software: military-grade messaging, Spacelog, work for Yahoo on Yahoo! entertainment, news, maps - Warning: professional swearing enthusiast - Talk brought to you by the number 8, and the words shit, bollocks, and fuck. Glad to be speaking; love Edinburgh; great line-up. Thank Norm for his help and support. “Ominous” by David Thompson

@SteveMarshall Friday, February 21, 14 - “Chase me on Twitter!”—Tom Coates You probably don’t want to follow me: quoting lyrics, swearing, etc.

Friday, February 21, 14 What am I talking about? - Specialism is a gift; left unchecked, though, it can lead us into false beliefs - Encourage embracing simplicity in your projects - Build the right thing, at the right time - How to think, how to do “Your Road Ahead” by Trey Ratcliff NEXT: Challenges in our minds Keeping things simple is hard, particularly when it’s not as cool as building an all-singing, all-dancing, thing. Further, learning the right thing is made harder by aspects of our psyche that we can’t necessarily recognise. - Cognitive biases - 5 I think are most relevant

Friday, February 21, 14 Bandwagon effect Quad biking! - Follow person in front: go off course - More people go off course, more likely to follow This is the bandwagon effect. - Follow the road, not the person in front Oxford Dictionary of Psychology (2003) “the probability of any individual adopting [an idea] increasing with the proportion who have already done so” People believe in something => more people “hop on the bandwagon” Happens: - regardless of evidence - consciously and unconsciously - If consciously: “information cascades” form, ignore personal information signals in favour of following others. Why bad? - Robert Heinlein, Glory Road “Wisdom is not additive; its maximum is that of the wisest man in a given group.” “Ogden Sunset” by Paul Duke

Friday, February 21, 14 Denial - Hanging paintings using no more nails - Dad comes ‘round. Asks why the fuck you didn’t use nails, like any sane person. Three options: - Simple denial: deny nails are up to hanging paintings (deny reality altogether) - Minimisation: Admit nails work for hanging paintings, but aren’t as good as ‘no more nails’ (admit the fact, but deny the seriousness) - Projection: Admit nails work, and are better, but claim you couldn’t get any (admit the fact and seriousness, but deny responsibility) Why? - Defence mechanism: truth is difficult to accept, insisting that it’s not true, despite evidence. - Over-run projects, stick-in-the-muds - Closed to other options “The Wall Was Framed” by Leonski

Friday, February 21, 14 Confirmation bias Search for “No More Nails hanging paintings” - Biased search for information - Search for evidence consistent with current beliefs; pass over when it isn’t. Search for “hanging paintings”, but only pay attention to those praising about No More Nails - Unbiased search, so full result set - Biased interpretation - Even if unbiased evidence is presented, will rate confirming evidence more highly Search for “hanging paintings”, read all, only paremember those praising about No More Nails - Biased memory - Even if search and interpretation was unbiased, may be selectively recalled to support their beliefs and reinforce their expectations. This is confirmation bias. - Favour evidence that supports existing beliefs, regardless of legitimacy - Ignore alternatives, and build patterns around current beliefs - Contributes to overconfidence and, thus, poor decision making - Leads to attitude polarisation and, thus, to backfire effect “Hardware Store Nails” by Patrick Hoesly

Friday, February 21, 14 Backfire effect Uni Michigan/Georgia State U study: - Create fake articles about polarising issues (eg. WMDs found in Iraq) confirming a widespread misconception - Subject reads fake, then reads a true article correcting the first - Those who agree with the the first disbelieve the second, and vice versa - All as expected so far… - People who agreed with the first (fake) article were more certain after the second (true) article, not less! This is the backfire effect. Happens because negative stimuli require attention/response: ancestors who failed to address these failed to keep breathing. - Don’t care about a subject: new info assimilated and viewpoint changes accordingly - Care deeply: new info supporting strengthens views, new info opposing strengthens views - Induces polarisation of views IMG_1935 by Devin

Friday, February 21, 14 Authority bias Examples: Hitler, IDs/classes, … - Someone claims something is true of an ambiguous subject. - That person is seen as an authority. - Therefore, that thing is true. Fallacy: the person is not infallible; just because they say something is true, it doesn’t mean it is or isn’t. Essentially: Don’t trust anything I say; it could all be bollocks. NEXT: How to think So, if we can’t trust our own minds, how can we be sure we’re doing the right thing? Let’s look at how to go about solving problems…

Friday, February 21, 14 Innovation as presented by the media: “Yes, of course! November 5th, 1955… That was the day I invented time travel. I remember it vividly: I was standing on the edge of my toilet hanging a clock. The porcelain was wet, I slipped, hit my head on the edge of the sink, and when I came to, I had a revelation! A vision! A picture in my head! A picture of this… This is what makes time travel possible: the flux capacitor!”—Dr. Emmett Brown, “Back to the Future” - Cute narrative device, trope accepted as truth Newton: - Apple that fell on his head probably didn’t exist. - Popularised in essay by Voltaire, long after Newton’s death - LIE: Actually a story Newton told his niece in a letter he sent from prison - FUCKING PHYSICIST, DON’T BE SO INSULTING - 20 years grappling with theory of gravitation - Didn’t discover gravity—was already discovered—developed consistent model of it Dyson: - Looks, from outside, like Dyson took the world by storm - Started working on improving vacuum cleaners in late ‘70s - Made over 5000 prototypes - Founded company in 1993, after failing to sell idea to Hoover and others - Didn’t want to cannibalise precious bag sales - Highest market cap as of 2010 of all vacuum cleaner manufacturers Fuck “innovation” - No ‘eureka’ moments - True innovation comes from the relentless, painstaking quest for perfection in one’s endeavours. - Based on wealth of experience, constant learning/testing, and play NEXT: Pursuit of perfection through scientific method

Friday, February 21, 14 Innovation as presented by the media: “Yes, of course! November 5th, 1955… That was the day I invented time travel. I remember it vividly: I was standing on the edge of my toilet hanging a clock. The porcelain was wet, I slipped, hit my head on the edge of the sink, and when I came to, I had a revelation! A vision! A picture in my head! A picture of this… This is what makes time travel possible: the flux capacitor!”—Dr. Emmett Brown, “Back to the Future” - Cute narrative device, trope accepted as truth Newton: - Apple that fell on his head probably didn’t exist. - Popularised in essay by Voltaire, long after Newton’s death - LIE: Actually a story Newton told his niece in a letter he sent from prison - FUCKING PHYSICIST, DON’T BE SO INSULTING - 20 years grappling with theory of gravitation - Didn’t discover gravity—was already discovered—developed consistent model of it Dyson: - Looks, from outside, like Dyson took the world by storm - Started working on improving vacuum cleaners in late ‘70s - Made over 5000 prototypes - Founded company in 1993, after failing to sell idea to Hoover and others - Didn’t want to cannibalise precious bag sales - Highest market cap as of 2010 of all vacuum cleaner manufacturers Fuck “innovation” - No ‘eureka’ moments - True innovation comes from the relentless, painstaking quest for perfection in one’s endeavours. - Based on wealth of experience, constant learning/testing, and play NEXT: Pursuit of perfection through scientific method

Friday, February 21, 14 “Truth is sought for its own sake. And those who are engaged upon the quest for anything for its own sake are not interested in other things. Finding the truth is difficult, and the road to it is rough.” —Abū ʿAlī al-Ḥasan ibn al-Ḥasan ibn al-Haytham’s (Alhazen) “Critique of Ptolemy” 1. Consider the problem; try to make sense of it. 2. Form a hypothesis. Try to state an explanation, and note it somewhere. 3. Deduce a prediction from that hypothesis. If #2 is true, what follows? 4. Test. Look for the opposite of each consequence in order to to disprove 2. Example - Consider: what shape is the earth? - Hypothesis: round - Prediction: if round, we can travel in same direction until we reach origin - Test: travel in same direction, trying to find an edge Rather than seeking truth, we’re seeking a solution to a problem. If we build a complex, over-wrought solution, building it will take lots of effort, and testing whether it solves the problem will be more and more complex. Build the simplest solution. If that doesn’t fix the problem, consider why not. Is it the wrong approach? Is it the right direction, but needs more? “Blinded by Science” by Florence Ivy

Friday, February 21, 14 Solve only what you need to solve, when you need to solve it. If you try to fix something too soon, you’ll end up wasting effort if you find that you don’t need that problem fixing. Premature optimisation - Donald Knuth (father of algorithm analysis, Art of Computer Programming, etc.), Structured Programming with go to Statements, 1974 “We should forget about small efficiencies, say about 97% of the time: premature optimization is the root of all evil” Don’t know the full extent of the problem until you get there, so can’t solve it yet. Examples: - Driving down the road, if I turn for the 3rd (medium short right) turn now, I’ll plow into the fence. Need to be aware that I will need to turn later, but not act on it until necessary. - Jared Spool’s $300m button: remove registration requirement from site, increase revenue by $300m (http://www.uie.com/articles/three_hund_million_button/)

Oscar Wilde 1854–1900 Friday, February 21, 14 Know your limits “I am no longer young enough to know everything”—Oscar Wilde - No-one can know everything, even in small domains - Admit your limits: test ideas frequently, ask others (but ask to see their working) - Takes more balls to say “I don’t know” than to blindly plow on - Mark of a professional - Doesn’t have to be a committee: take ownership of final product, but ask advice of others Oscar Wilde portrait by Napoleon Sarony NEXT: How to build

John Donne 1572–1631 Friday, February 21, 14 “No man is an Iland, intire of it selfe; every man is a peece of the Continent, a part of the maine […] Any mans death diminishes me, because I am involved in Mankinde; And therefore never send to know for whom the bell tolls; It tolls for thee.”—John Donne, Devotions Upon Emergent Occasions: Meditations 17 (1624) If you’re not using the entire web stack—CSS, JavaScript, and HTML—you’re not using the web to its fullest. Use all three in harmony, don’t ignore—or bastardise—one of them. If you’re not doing this—and claim one part or other of the web stack isn’t up to scratch—you’re the problem, not the web stack. “Portrait of John Donne”

Friday, February 21, 14 GMail is my absolute favourite example of this. They launched a JS-only UI in April 2004, then took nearly a year to release a completely separate HTML version. If they can search billions of pages in fractions of a second, why can’t they load my inbox without JavaScript?! It’s HyperText Transfer Protocol, so transfer some hypertext! HTML: fundamental building block: not JS, not CSS. If your system relies on things HTML can’t do, you’re not building for the web. Shows lack of respect for users and misunderstanding of the medium. Design with HTML in mind first, then layer on functionality. Makes everyone’s life better. More robust.

Changsha, China Ark Hotel Friday, February 21, 14 - Build structure of the site, and core functionality, server-side and with HTML. Usable (but not pretty) without styling or interactive stuff. - Build more and more pages. - Can start adding styling and interactivity after the first story’s done! - Some pieces just drop in, fully formed, your framework provides these! - Only building things when they need to: not building the top floor before the entrance. - Layer on CSS to get the building looking good. - Finally—finally—add interactive elements to the structure. - Analogy doesn’t always work: easier for us to swap out our foundations with no-one noticing than on a hotel. “Ark Hotel Construction time lapse building 15 storeys in 2 days”

Friday, February 21, 14 - Build structure of the site, and core functionality, server-side and with HTML. Usable (but not pretty) without styling or interactive stuff. - Build more and more pages. - Can start adding styling and interactivity after the first story’s done! - Some pieces just drop in, fully formed, your framework provides these! - Only building things when they need to: not building the top floor before the entrance. - Layer on CSS to get the building looking good. - Finally—finally—add interactive elements to the structure. - Analogy doesn’t always work: easier for us to swap out our foundations with no-one noticing than on a hotel. “Ark Hotel Construction time lapse building 15 storeys in 2 days”

Friday, February 21, 14 Do you really need a DB? - RJ’s IRCCloud session! - IRCCloud.com, web-based IRC client by James and RJ, former last.fm developer and CTO/founder, respectively - If you want to use it, request a beta invite mentioning highland fling - Humility: asking peers - Not averse to unorthodox solutions - Only cared about scalable solution, not own pride - Your blog: Write once, read many. - If your site was just HTML, you’re less likely to be fireballed, slashdotted, etc. “Timeless Books” by Lin Pernille

@SteveMarshall Friday, February 21, 14 MAYBE MERGE THIS AND LAST SLIDE? - Build foundations and structure first - Take the next step, not the thousandth - Don’t blindly trust authorities, make your own judgements - Base your decisions on evidence, not opinion, but if it must be opinion, at least make it well-reasoned “If you aren’t breaking a sweat, your workout isn’t hard enough.”—? KICKER - Newton was never imprisoned. But you believed me, didn’t you?