72

CodingHorror writes:

So if you want to become a great programmer, start by becoming a great typist. [...] When you're a fast, efficient typist, you spend less time between thinking that thought and expressing it in code. [...] I believe in practicing the fundamentals, and typing skills are as fundamental as it gets for programmers.

ASNA writes:

Having spent tons of classroom time with lots of students for the last 14 years or so, one of my general observations is that many programmers aren’t as good at typing as they should be. [...] I’ll get arguments here, but for my money, better typists make better programmers!

Dodgy Coder writes:

Learning to touch type is a quick and effective way to give your productivity a boost as a programmer.

Is there empiric evidence that programmers who learn to type faster will tend to be better programmers? Do those programmers perform better on the job that employers pay programmers to do?

Christian
  • 33,271
  • 15
  • 112
  • 266
  • Comments are not for extended discussion; this conversation has been [moved to chat](http://chat.stackexchange.com/rooms/66971/discussion-on-question-by-christian-do-programmers-who-learn-to-type-faster-beco). – Sklivvz Oct 12 '17 at 07:10
  • This seems to be a misunderstanding of the claims. They aren't saying that you will be better relative to other people as long as you can type fast. They are saying that you will be better relative to yourself. If you take the exact same person that is typing 50 words per minute and enable them to type 100 words per minute, it seems trivial to say that the person is now "better" at the part of their job that involves typing. – bwarner Oct 13 '17 at 17:18
  • 5
    somewhat related to this [https://skeptics.stackexchange.com/questions/17224/do-professional-software-developers-write-an-average-of-10-lines-of-code-per-day](do professional software developers write an average of 10 lines of code per day) showing that the amount of written code lines per day is small and not necessarily requires typing faster. – Rsf Dec 11 '17 at 16:37
  • 2
    Basic typing skills are necessary, but once one gets beyond hunt and peck, the speed at which they type impacts good development, about as much as the windshield wipers affect the speed of a Ferrari. Doesn't matter how fast you can type... if you can't abstract your thoughts into good code, you won't produce much of anything. – tj1000 Dec 12 '17 at 01:42
  • 1
    I can't address the issue of speed but I do feel touch typing is a very important skill for a programmer--a touch typist doesn't **think** about what they are typing--less disruption to thinking about the code. I know I find it a hindrance typing on a keyboard where I have to think about anything. – Loren Pechtel Dec 12 '17 at 02:07
  • 2
    @LorenPechtel: I am German, but much prefer US keyboards out of habit. Needless to say, I don't always get them, e.g. when sitting down at a coworker's place. Yes, always stumbling over Y vs. Z and having to think twice over "-{}[]|\" etc. is a bit annoying. But -- as Rsf pointed out -- the number of LOC is not really impacted by this, as you spend most of the time *thinking*, not typing. I'd be interested in how much not being a good typist impacts the amount of *comments* and *documentation* written, but I seldom see people typing down code at speed for any length of time. – DevSolar Dec 12 '17 at 08:53
  • @DevSolar My point is that anything you have to think about while typing is a distraction, albeit small. The fewer distractions the better. – Loren Pechtel Dec 13 '17 at 04:03
  • 2
    I think the first quote could be better expressed, what I think that CodingHorror wanted to say is that if you're typing is too slow, it will disrupt the flow of your thought so you'll be even slower than your current typing rate. Once you are fast enough to follow the flow of your though, you don't need to be more than that. The speed of your flow of though will vary greatly depending on the type of software and your own skill (real time -> slower, simple forms/database -> greater). – Walfrat Dec 14 '17 at 13:23
  • To be honest? it seems really hard to have a study about how fast we learn to type "fast". You'll have to define fast, and most of us probably don't know our current typing rate, even less when we were starting developping. – Walfrat Dec 14 '17 at 13:25
  • 1
    The fact that most people don't know their typing speed doesn't mean that a study can't give them a test to measure it. – Christian Dec 14 '17 at 13:26
  • 1
    It's going to be really hard to nail down a causative relationship here as it is impossible to program much without getting better at typing. My personal experience had included 0 typing training, I started programming without any ability at all at typing, and naturally became as proficient as anyone who types all the time every day for years on end. – Jonathon Dec 27 '17 at 14:37
  • @Oddthinking: 1) I'd like to know *what* flags were raised. 2) The answer received no comment before it got deleted "for already earning flags". That implies multiple flags, yet only one downvote. I feel this is rather heavy-handed both by the flag-raiser and yourself. (What are comments FOR, then? I can't even post questions where they would belong - under my (deleted) answer.) 3) I *did* look into it, and apparently this subject has *not* been studied by any "appropriate experts". I focussed on why this may be, should I have made it more clear that there seems to be no statistics on this? – DevSolar Jan 05 '18 at 13:07
  • @DevSolar: Within 6 minutes, you had a "Not an answer" and a "Very low quality" flag. – Oddthinking Jan 05 '18 at 13:09
  • Suggestion for folks trying to answer this question: Studies about programming productivity based on keyboard familiarity/ergonomics might be helpful. A major problem with this topic is that it'll be hard to control for a lot of complex factors when comparing programmers by their keyboard proficiency. However, we might consider programmers who're artificially disadvantaged at typing to be good proxies for a hypothetical version of themself with weaker typing proficiency. – Nat Jan 05 '18 at 13:13
  • @Nat : Science is a thing. It's not really hard to run a controlled study where one set of people get typing training. Given the big economic value of improving programming quality, many institutions should have an incentive to run such a study. There's a lot of research run that's a lot more complicated. – Christian Jan 05 '18 at 13:34
  • @Christian: No such study exist (that I could find), which seems to indicate that sending programmers on courses about languages, techniques, and tools (which is done all the time) is considered more productive by those who pay the bills. Perhaps you could narrow down the question in one critical regard: Are you asking about *programmers*, i.e. pros and possibly undergraduates, or *students* who have a pipe dream of *becoming* a programmer once they don't have to look for the space bar anymore? – DevSolar Jan 05 '18 at 13:38
  • @Christian Seems that you'd then be testing the same programmers at different times after dedicated training initiatives, typically conducted in commercial contexts in which reporting'll tend to be positively biased. I mean, such a study would also be great, but it seem less plausible than other approaches. – Nat Jan 05 '18 at 13:38
  • @DevSolar : You provided no evidence for the claim that there are no studies for this. You haven't even provided any evidence that there are no published studies (organisations like Google might have inhouse data because they have an interest in knowing whether it's valuable to give their programmers typing courses). But even if you would prove such a thing it would be like answering a question about an alternative medicine intervention with "There's on research on this but the treatment worked personally for me." Such answers don't fit our standards. – Christian Jan 05 '18 at 13:40
  • @Christian: "You provided no evidence for the claim that there are no studies for this." -- I laughed. – DevSolar Jan 05 '18 at 13:41
  • @DevSolar : That might very well be true that the people who pay the bills hold those opinions but humans often hold mistaken beliefs without any good evidence for it and this question is about actually getting evidence for whether a claim is true. – Christian Jan 05 '18 at 13:41
  • @Christian: 1) Please narrow down the question. Students, or programmers? 2) Perhaps you could come up with a "claim" that is not just opinion itself? – DevSolar Jan 05 '18 at 13:42
  • @DevSolar : In cases where it's hard to find any evidence that matches a question, why would it be a good idea to narrow down the question? I would welcome a study that a CS professor does on his studies such as I would welcome a study that Google does with their programmers. – Christian Jan 05 '18 at 13:45
  • @Christian: When you're looking at students, you are looking at a high percentage of non-typists. When looking at professionals, junior professionals, or even undergraduates, you are looking at people who, in the vast majority, can already type at least halfway fluently. The difference lies in what a lack of studies would imply. In the former case the need for training -- formal or informal through raw experience -- is *self-evident*. In the latter case the argument can well be made that such improvement would be dwarved by the effectiveness of learning language, tools, and environment. – DevSolar Jan 05 '18 at 13:49
  • @Christian: Besides, it would narrow down the search parameters / keywords when *looking* for these studies, and avoid another answer getting shot down for talking about the "wrong" target group. And yes, it's a challenge on the notability of your claim, as it seems your sources are mainly aiming at non-(touch)-typists, not at the *speed* of your typing *per se* (as the enveloping question does). – DevSolar Jan 05 '18 at 13:53
  • @DevSolar : The point of a question is to search for evidence that's helpful for evaluating the claim. The prime reason answers get deleted here is that they just provide personal opinion. – Christian Jan 05 '18 at 14:09
  • Let us [continue this discussion in chat](http://chat.stackexchange.com/rooms/71240/discussion-between-devsolar-and-christian). – DevSolar Jan 05 '18 at 14:53

1 Answers1

1

According to Cognitive Consequences of Programming: Augmentations to Basic Instruction Journal of Educational Computing Research , volume 2, pages 75-93 (1986):

Teachers we surveyed recommended that we help students become better typists. Lack of typing ability was perceived as one of the largest obstacles to success in programming.

So at least at a very introductory level, yes.

It is a somewhat obsolete truth. Now little kids grow up surrounded by keyboards, but until recent decades it was common for adults to have no typing experience.

In Software Engineering: Proceedings of the Third Symposium on Computer and Information Sciences - 1969 : Jean Sammet of IBM stated:

on-line system requires that the programmer become an adequate (and preferably expert) typist. He spends a great deal of time typing—a task which previously was delegated and relegated to keypunch operators.

DavePhD
  • 103,432
  • 24
  • 436
  • 464
  • 10
    The quotes demonstrate that belief in this claim was widespread, not that it was true. – BobTheAverage Jan 03 '18 at 21:04
  • @BobTheAverage It's an expert opinion reference based answer. Hopefully someone will find a factual, actual study but nobody has so far after 5 years. https://en.wikipedia.org/wiki/Jean_E._Sammet – DavePhD Jan 05 '18 at 12:32
  • 2
    @DavePhD: The former was a query of junior highschool teachers -- qualified opinion on students perhaps, expert opinion on programmers no. The second is about how programmers now have to type themselves instead of writing down instructions by pencil and handing the paper to *the keypunch operator*. I don't think either really applies to the question. – DevSolar Jan 05 '18 at 13:36
  • @DavePhD : I don't think the "lack of typing ability" that a student in 1986 who never typed anything before his studies had is comparable to what Coding Horror or ASNA mean when they say bad at typing. As a result, I think the information doesn't really help to answer the core question of whether typing improvement is right now beneficial. – Christian Jan 05 '18 at 14:17
  • +1. Teachers who observe this seem to be providing qualitative, subjective feedback. It's certainly not conclusive, but it's still useful information. – Nat Jan 05 '18 at 14:52
  • Also, people who become great typists sooner are probably better programmers simply because they're more clever/hard-working, not because they can type faster. It's like saying that having a messy table is a sign that you're a great scientist just because [insert a famous scientist's name] had a messy table. In other words, `correlation != causality`. –  Jan 07 '18 at 11:31
  • "It is a somewhat obsolete truth. Now little kids grow up surrounded by keyboards..." This doesn't somehow cause them to absorb the skill of typing. Typing must be deliberately learned. – Kyralessa Jun 11 '23 at 12:01
  • @Kyralessa I took typing as a class in high school, on manual typewriters. We were graded on speed and number of errors. We had vinyl typing music records and were supposed to type to the beat. My kids learned in much less deliberate way. – DavePhD Jun 12 '23 at 11:32