24 comments

[ 2.4 ms ] story [ 63.8 ms ] thread
To me, this is stating the obvious.

The best way to learn is by doing, and the best way to get better is to (happily) make mistakes and learn.

Nobody/nothing is perfect, and often (perceived/contextual) perfection is derived from iteration. And the hardest part about iteration is often making the start.

Jeff likes to give his posts misleading titles.

Iteration (which is what he's referring to when he says "quantity") is an essential part of quality software. But "quantity over quality" suggests perpetually producing lots of crap without regard to outcomes.

Still, his call to prefer working, shipping code over paralysis-through-analysis is very good.

Actually, the preference for Iteration is really just another form of "Evolution is Smarter Than You." Or, if you will: You'll never learn as much about [X] by just thinking about it, as opposed to actually doing [X] a whole bunch! (While thinking about it.)
An interesting counterpoint (kind of anyway): http://ravimohan.blogspot.com/2007/04/learning-from-sudoku-s...
But that is Doing & Thinking vs. Doing & not enough thinking.

Our options are:

    1) Too much thinking, not enough doing
    2) Balanced thinking & doing
    3) Too much doing, not enough thinking
As often is the case, the middle path is best.
Yes I agree, its not a terribly interesting conclusion, but a balanced approach wins (at least for interesting problems). I guess there is often a risk that there is too much thought/talk versus action more so then the reverse (not to say that the reverse doesn't happen, it is just rarer).
(comment deleted)
the "quality" group had sat theorizing about perfection, and in the end had little more to show for their efforts than grandiose theories and a pile of dead clay.

Is it narcissism when authors believe their readers are so stupid that they'll drink everything they write ? Don't we deserve at least some sort of credible example, even if it's obviously made up and exagerated to fit the author's narrowed perception of reality ?

See, by Atwood's very sound reasoning, we'll help him improve quality by not reading his blog and by supporting somebody who's actually a writer worth reading. But his article doesn't cover the fact that hackers are subject to inertia like everybody else.
Misleading title.

More accurate: Quantity Causes Quality.

Quantity implies practice, which leads to quality.
But it doesn't imply practice. Look at how productive the average fanfiction writer is.

It's a misleading statement written as-is.

Sometimes quality doesn't cause anything at all. If you practice chess, for example, online, you'll find many player who really are no good despite playing literally thousands of games. I think there's no substitute for trying to do something well.
"Sometimes quality.." I guess you mean quantity
Practice does not make perfect. Perfect practice makes perfect. If you keep repeating the same mistakes over and over again, you gain nothing.
When learning to play a new piece of music on an instrument, one of the recommendations I heard was this:

Focus first on playing all the notes in the right order. Don't worry about the timing at all, even if it takes several seconds to move from one note to the next. Once you can play all the notes, then you start to work on the timing, playing gradually faster and faster until you have both the melody and the rhythm mastered.

The idea is that once you have memorized the correct movements, it's much easier to work on the timing. If you start playing too fast and make mistakes, you'll start to memorize the mistakes.

My tae kwon do master states the same:

"Slow, well done, steady."

[as opposed to aerobics shake-it-all with music "gymnastics"]

As I've always said, the difference between the expert programmer and the novice isn't that the expert is smarter, he just did the same stupid mistakes so many times, he knows his way out.
This only works if you learn with each iteration. At times I don't. Learning is an active process and needs time and effort dedicated to it. Otherwise you're running in place.
This illustrates why RSS is a failure. Almost all of the people with the most subscribers are the ones who churn out mindless crap on a regular schedule. So much for the theory that RSS would give those who publish great stuff intermittently an equal shot at the limelight.
Is there a service which monitors high-quality, infrequently updating sites and easily collates them for regular perusing?

If not, would you like to see me make one? :)

I don't see how reguarly posted crap takes anything away from occasional quality. Of course the people who post regularly get the most subscribers - they produce the most stuff! But I subscribe to about 200 feeds, most (150+) of which don't publish even once a week. Without RSS I would never bother to follow them.