This article reminds me of an essay by Edsger Dijkstra, EWD1036 [0], comparing digital vs analog processes, and how they lead us to think of discrete bugs vs continuous errors.
(I recall it as arguing against the naive tendency of considering machines as continuous devices).
Interestingly, Dijkstra also uses a musician as example, for the alternative view:
if the violinist slightly misplaces his finger, he plays slightly out of tune
> if the violinist slightly misplaces his finger, he plays slightly out of tune
This depends on whether the slightly misplaced finger ends up on the correct string, correct? I suppose to a violinist, however, a finger that was no longer on the correct string would not be slightly misplaced.
I try to attribute behavior to extrinsic properties whenever possible. And then consider it intrinsic if proven otherwise.
Most people seem to do it the opposite way, which probably leads to a lot of classification errors. Fixing an intrinsic problem is so hard that it's considered a waste of time by most, whereas fixing an extrinsic problem is within the realm of possibility.
I figure it's better to waste time trying to do the impossible if the alternative is to leave a lot of low hanging fruit untouched.
Oddly, I've been pondering the intrinsic/extrinsic divide recently. I was seeing the same thing you are referring to, I believe. Easy example is traffic. People think it is something intrinsic to traffic that has them upset. That is clearly false, if I can go through the same traffic and not be upset.
So, firstly, is that essentially the same thing? And second, any good reading on this idea?
More close to extrinsic/intrinsic is the traffic problem itself.
Is traffic created by individuals, or is traffic created by a larger systematic effect?
If you have the first perspective, then you're likely to be upset at people and at the traffic. If you take the second perspective, you're likely to understand its true nature and possible ways of fixing it.
In my experience, the systematic perspective is almost always the correct one. Individuals are merely operating their best within their system. Exactly as in the original post with her fingers and the piano. Brilliant analogy.
We need more systems thinking, beyond just intrinsic/extrinsic understanding.
I don't know any good reading on the psychological effect you speak of, but good reading to start on systems thinking is Deming and go from there: http://en.wikipedia.org/wiki/W._Edwards_Deming
Interesting view, but I wonder if it applies on so large a field the author seem to believe. Piano may be a good example: some wrong notes can be bugs but if the kid do not enjoy playing music and take it as a chore, no amount of bug fixing will help.
Same for math, no-one said everyone must love them, because it is not that lovable, and to be good at it math lover have a head start. So before getting to the learning disability bug fixer you might want to make your kids fall in love with math or grammar or whatever.
That's how I remember my love affair with math: one day a math teacher did trigger love for math in me. It was a mix of telling me I was good at it, showing me how powerful and wonderful a machine it is, and so on.
You do have a point, but I wonder what's behind that 'love in math' you're talking about. I tend to think that curiosity is the only basic character trait needed to fall in love with something as basic as math, grammar, music or physics. However, the way this kind of love is triggered doesn't necessarily correspond to how it is taught in lower classes. At least when I went to primary school in the 90ies, 'math' was really just numerical algorithms (which are IMO not at the root of what math is, it's just one sub field). What can it be applied to in our daily lives? (considering someone doesn't take a higher math education) Knowing the price total in supermarkets, that's about it.
Couldn't we include symbolic math at a much earlier age? Children enjoy the building block properties of Legos starting at age 7 I'd say. How about we give them basic problem solving tools in math at age 10? I think we need to bring the potential feeling of achievement with mathematical tools to a much lower age. That could be geometry or color theory or basic physics (calculations with distance, velocity and time as an example).
I can only speak for myself but for me, the fact that math did not have any daily life application was one of the greatest reason to love it.
I despised physics even if I was fairly good at it. I considered physics was taking pure math virgin to a mud field and raped her there with "daily life" problems.
I understand what you mean, I think I'm wired in a similar way. However I'm skeptical whether you can spead that kind of 'love' to a wider audience without coming from the applied side of things. I believe the power to describe some daily live phenomenon in a mathematical model and then use mathematical tools to predict behavior could be a very eye opening experience for many children - and I think we could design exercises that follow such a path, even with very basic algebraic tools.
Simultaneously we could also come from the other end - describing a new world in a structured language (programming) - in order to awaken desires to learn new expressions. I remember, when I was a kid I wanted to program an 'Asteroids' clone, but I couldn't get the flying behavior quite right. I was excited when I learned about vector geometry - finally I had the tool I need to model that.
I see also what you mean, but this might apply more to programming or physics than to math. Being completely "useless" is a feature, not a bug, for math (and poetry, music, metaphysics, astronomy). Pure speculation or useless braingames is a thing many humans enjoy, see Rubix-cube or Sudoku.
The premise still applies in this case. If a student consistently has poor performance in piano and you can attribute that to lack of motivation, then that would be the bug (contrast with the notion that the student is stupid or bad at piano). If you can fix that one bug, you stand to gain a whole range of performance improvements.
When we say bug, I guess we just mean "avoidable issue" of a relatively modest importance, but that can have big impact, right?
Then, if you learn singing or violin, and you are "tone-deaf" or near tone-deaf, then it is not a bug, it is an inability to love and learn music, and no amount of patching will fix it for real, even if it could hide the problem under the carpet for a while, in the case of piano learning. (Because a "tone-deaf" could play piano relatively well, but probably won't be able to play violin.)
I would love to be convinced that "math-deaf" do not exists or is only very rare, but it simply do not match with my experience. To take an example: When I was young, and still now: I find the "open" and "closed" segment symbolised by ]----[ and [----] to be a very enlighting and powerful piece of mathematics. But when I tried to explain it to people who are not in love with math, I just get blank stares: they would understand it, be able to use it, but would never have tears in their eyes when trying to explain it to someone else. I think they just don't see the beauty of it.
Not seeing the beauty in math is not to be considered a defect or a bug. I'd say on the contrary: seeing beauty in math is borderline psycho and an anomaly, just like seeing music graphically or being haunted with stories (Balzac was). But it helps a lot for math studies, just as loving music will help learning music. And math studies are important for most scientific or engineering careers.
To wrap this up, it means that nature is not fair: some kids are gifted, some are not. In fact I do not see any reason for nature to be fair, and many events show enough it isn't (See Lisbon earthquake). The "bug theory" of the OP is trying to hide this fact, as is the "learning disability" movement.
But the basic point is correct. Generally, most people would rather get something right than get it wrong, even if they don't care very much. Calling them stupid doesn't help them get it right next time. Finding the source of error does.
This goes back at least as far as W. Edwards Deming in the 1930s.
If I had to summarize this blog post in one phrase it would be: work the problem not the person. The act of solving the problem changes the person.
The major exception to this is intent. When a person isn't interested in solving the problem this technique will be less effective. And since all pre-college education is compulsory there will always be a significant percentage of students in need of an attitude adjustment.
Then the attitude is the problem, and you need to work that. Fix the system to change the attitude.
As an aside, no one commenting on this post really understands it.
She's saying: look, it's not a person problem, it's a systematic problem. If you look at everything this way, it's a better paradigm; it opens doors, it allows you to grow, instead of shutting down based on incorrect assumptions and false pretenses.
In fact, I'll go broader: few people on HN or in the business community at large really understands that people problems are really systems problems.
Systems thinking. That's what she's talking about. And she's absolutely right, it's the most powerful paradigm shift available, and more people need to have it.
In the matter of talking about bugs, I try to use terminology influenced by the SEI.
"It's crashed" --> "It has failed".
"It's behaving buggily" --> "There is a fault / it is faulting".
"It's a bug" --> "It's a defect".
The one to me that really matters is "defects". It's a matter of taste, but I find that the term "bug" implies mysterious agency. "Defect" correctly describes that ... there's a defect. Something is wrong and it can be fixed.
Fault, error, failure, hazard. Technically speaking, nothing actually matters until it becomes a hazard (if you're a cynic, this means "costs you money"). I think that this is one way of thinking that engineers could benefit a lot from, in that framing issues within these boundaries gives you a lot more perspective.
(I recall it as arguing against the naive tendency of considering machines as continuous devices).
Interestingly, Dijkstra also uses a musician as example, for the alternative view:
if the violinist slightly misplaces his finger, he plays slightly out of tune
[0] http://www.cs.utexas.edu/~EWD/transcriptions/EWD10xx/EWD1036...