Yoda Timers 3: Speed

post by alkjash · 2018-03-13T18:00:00.861Z · score: 40 (15 votes) · LW · GW · 9 comments

Contents

  Day 22: Speed
    1. Mistakes Are Fatal
    2. Speed Limits are in the Mind
    3. Speed is Easier than Strength
  Daily Challenge
None
9 comments

This is part 22 of 30 in the Hammertime Sequence. Click here for the intro.

At some point around the end of high school, being fast became unfashionable. When did this happen?

Why do we channel so much more energy into doing more difficult things, instead of doing simple things faster? How much faster could you do your job? Two times faster? Five times?

Instead of I want to be stronger, say I want to be faster.

If you pay attention to speed, you might just find a way to do a whole week’s worth of work in five minutes.

Day 22: Speed

Here are three exercises to help you find that rush of mind working on overdrive:

  1. Typeracer. Play this game for five minutes. How much faster did you get at typing? What did you learn?
  2. Go on YouTube and watch a talk at 2x speed for five minutes. If you’re having trouble following, turn on subtitles. Notice that this is possible.
  3. The Arithmetic Game. Play three rounds of this with standard settings. How much faster did you get?

Here are three principles I’ve extracted from using Yoda Timers to do everything faster.

1. Mistakes Are Fatal

When I first played Typeracer, I started out at a measly 70 wpm and worked my way up to around 90 by just trying harder. Eventually, I hit a plateau because I was still constantly making typos and backspacing. Each mistake cost the time of four or five characters. The backspace key was my Achilles heel.

That’s why I forced myself to slow down and get everything right the first time. At first, this lowered my wpm, but with a bit of work, my fingers felt more nimble and intentional. I cut down the number of typos I made by a factor of 4 or so – it turns out there’s a handful of sequences of keys I constantly get wrong or out of order. My wpm skyrocketed to 120.

In real life, mistakes are even costlier. Getting sick is way costlier than having good hygiene. In programming, it’s common knowledge that testing and bug-fixing takes at least three times as long as writing code in the first place. In math, months of paper-writing can go down the drain when you finally notice a severe and unfix-able logical misstep. At the Olympics, every single mistake will cost you the medal.

If you want to be faster, you must have zero tolerance for even the slightest errors, and slowing down (at first) to practice perfectionism will be worth it. Get it right the first time.

2. Speed Limits are in the Mind

Everyone has a rough idea of how long things have to take. Solving a hard research problem always takes at least a month, right? Writing a paper should take at least an hour, right?

When I first started playing the Arithmetic Game for middle school MathCounts training, my high score was close to 20. After a few months of dedicated training, my record hit 90, making it onto the leader-board of the time.

For any given task, do not assume you’re doing it anywhere close to your real speed limit. It used to take me at least four hours to write a blog post this length. This one clocks out in just under forty minutes.

3. Speed is Easier than Strength

In intellectual work, it’s much much easier to get twice as fast than twice as good. It’s much easier to multiply twice as quickly than to learn to solve harder problems. It’s much easier to type twice as much content than to write twice as well.

Human beings are supremely good at training rote tasks to maximum efficiency. Take advantage of that. Learn to read twice as fast, write twice as fast, talk twice as fast, walk twice as fast, watch videos twice as fast. I’ve been watching videos at 2x speed for as long as I can remember, and I can’t even tolerate regular speed anymore. Once you habituate to going faster, you reap all this free energy that was just lying around while you were waiting.

Speed is underrated. Short training sessions focused on speed will create lasting impacts on your productivity.

Daily Challenge

Share your proudest speed record. Fast is fashionable again!

9 comments

Comments sorted by top scores.

comment by Qiaochu_Yuan · 2018-03-13T20:51:46.387Z · score: 42 (9 votes) · LW · GW

I like this.

Share your proudest speed record. Fast is fashionable again!

Some miscellaneous ones, not in order of proudness, but I'm just curious how many examples I can produce.

  1. I type at around 140-150 wpm and have for most of my life. I got up to 160 wpm on typeracer, which would have been enough to get on the leaderboard except that I didn't create an account. typeracer is somewhat easier than other similar typing tests I've done in the past because you can read the short segment you're about to type in advance. 10 seconds is a long time.
  2. I credit my typing speed to a combination of playing piano, spending a lot of time on MSN Messenger as a kid, and playing (edit: keyboard, 4-finger) Stepmania, where my all-time speed record is being able to AA a song that was mostly 16ths played at 260 bpm (so 1280 notes per minute). Stepmania's an amazing game for practicing sheer speed because you can speed up and slow down any song by multiples of 10%, so you can always play at your speed edge, or slow down as much as necessary to aim for perfection.
  3. I once read Harry Potter and the Chamber of Secrets 7 times in a single day. I don't remember why I did this.
  4. I watched the videos for Andrew Ng's Coursera course on machine learning at speeds of between 2.5x and 3.5x (thanks to the godsend that is Video Speed Controller).
  5. Arithmetic Game scores: 27, 28, 33 (Edit: more scores: 27, 42, 41, 30, 41, 35). I'm quite confident I could get much faster now and that I would have been much faster in middle school. My performance is mostly being slowed down by a combination of old mental arithmetic TAPs firing slowly and some TAPs around double checking my work firing and getting in the way. I noticed several tricks specific to this format, taking advantage of the fact that correct answers are automatically submitted while incorrect answers are ignored and only penalized by the wasted time: sometimes it's more time-efficient to guess the correct answer via various estimation and mod-10 tricks than to compute it (e.g. estimating the first digit and trying out a few candidates for the second digit), and I did that a few times, especially on division problems. (Edit: Mental arithmetic TAPs are speeding up, mostly I'm getting slowed down by tricky divisions.)
  6. The last final I took in college was for differential equations. I already knew all of the material and never went to class; I just needed it to graduate. I was given 3 hours to finish, and I finished in about 40 minutes. Also I showed up in a suit and tie, and with champagne, because I was told that this was traditional for graduating seniors, but I was finishing a semester early and also taking a class that most seniors would have taken already, so I was the only person in a suit and tie in a gym full of hundreds of people. I popped the champagne at about 45 minutes in and the entire gym heard it because everyone else was completely silent. Then I quietly drank a glass, and left.
comment by alkjash · 2018-03-14T04:46:29.455Z · score: 12 (2 votes) · LW · GW

Video Speed Controller is much better than the jank chrome extension I was using, thanks! Anime might just be barely watchable at 3x.

comment by steven0461 · 2018-03-15T18:23:12.606Z · score: 9 (2 votes) · LW · GW

I've sometimes found it helpful to subvocalize the word "go" as often as possible.

comment by tcheasdfjkl · 2018-08-23T20:48:52.359Z · score: 3 (2 votes) · LW · GW

I'm pretty bad at doing things quickly, but I think my roadblocks are less mechanical skill and more anxiety, executive function, and most of all perfectionism. I know you mention perfectionism as useful for increasing speed, and I think for mechanical skills that's true, but in a lot of tasks I am hugely slowed down at decision points (which are many) by a felt need to make the best possible decision every time, even when the decisions are fairly trivial and/or multiple paths forward are valid and/or perfection is not actually helpful. In this sense I most need to cut down on perfectionism for increasing speed.

Timers are indeed a useful tool for this, though. If I decide beforehand that it's more important to do the task quickly than to do it perfectly, having a timer helps me satisfice more readily (usually a good thing for me, given my baseline) because I'm more aware of how much time I'm spending.

Then, of course, I just need to decide wisely when speed is more important than making good decisions and when it isn't...

comment by norswap · 2018-03-15T17:16:46.641Z · score: 3 (1 votes) · LW · GW

It's been a long time since I thought along those lines, but it makes a lot of sense, thanks for writing this.

(If I really want to be my usual grudging self, I must add I think you probably gain very little from fast typing - thinking is by far the bottleneck and you can think as you type.)

comment by Qiaochu_Yuan · 2018-03-15T18:37:15.662Z · score: 12 (2 votes) · LW · GW
I think you probably gain very little from fast typing

Calibration exercise: what do you think the average typing speed is?

Rest of comment is in rot13: Vg'f nobhg sbegl jcz. V zragvbarq nobir gung V glcr nobhg n uhaqerq naq svsgl jcz; gung znxrf n uhtr qvssrerapr gb ubj yvxryl V nz gb glcr va trareny. Vg gerzraqbhfyl ybjref gur znetvany pbfg bs zr glcvat Snprobbx fgnghfrf, YJ pbzzragf, rgp. V nyfb guvax dhvgr dhvpxyl naq V orarsvg n ybg sebz orvat noyr gb glcr nyzbfg nf dhvpxyl nf V guvax, r.t. jura glcvat sbe gur checbfrf bs guvaxvat gb zlfrys va Jbexsybjl. Glcvat ng sbegl jcz jbhyq nofbyhgryl pevccyr zr.

comment by norswap · 2018-03-16T09:18:30.728Z · score: 3 (1 votes) · LW · GW

I guessed 35 based on typing a bit on type-racer to get a feel of the speed. I seem to be able to sustain 90, although on the long run you'd have to see how mistakes affect that average. I get your point but for me, writing is fast spurts of writing punctuated by long thinking pauses. In my comments and status, I probably spend 10x ballpark more time thinking than typing. It's even worse for some blog post and waay worse for academic articles (only for the first rough draft, not factoring editing).

comment by Qiaochu_Yuan · 2018-03-16T17:18:19.309Z · score: 6 (1 votes) · LW · GW

That's fair. I'd like to avoid typical minding here, though; my experience is that I have way more thoughts than I have time to write them up (both cached and on the fly), so even at my typing speed I still find that typing speed is my bottleneck.

comment by tcheasdfjkl · 2018-08-24T01:26:33.216Z · score: 1 (1 votes) · LW · GW

My scores on the challenges here:

Typeracer: 113, 108, 111, 101

(I'd done this before - not for training, just for fun. I expect my speed is pretty good largely because of piano. I do wonder how fast it might get if I trained for speed on purpose - though I am not sure it would actually matter very much to my life?)

Arithmetic game: 12, 7, 16, 5

(this depends primarily on how many division problems there are especially how many times I need to divide by 7. I realized I don't actually remember any actual methods for division, so all I can do is decompose the dividend into more easily divisible numbers somehow. I guess it would plausibly be useful for me to relearn how to actually do division, though it doesn't actually come up often.)

I don't think there are any things that I'm very fast at, so nothing to share for the daily challenge.