Skip to content

Comment on Slow and Steady is Bullshitparent

Comments

Yes they would, however that would be adding a healthy dose of "slow and steady" to the equation, wouldn't it?

Not really. Slow and steady would be splitting the entire learning material over a longer period. This would be learning everything as fast as possible (cramming), then taking a couple weeks break, then cramming again. It is indeed much more effective.

(the general case is much less explored, and the article provides no evidence either way, unfortunately)

You assume that all learning takes place over the cramming phase, then all practices is about keeping the knowledge fresh.

It is most likely that the cramming phase is an effective though limited way to break the novice threshold, though any knowledge gained here will be by necessity shallow. A later phase of slow and deliberate practices allows this initial mental structures to be settled in a pattern of competence an maybe provide the seedlings of insight that may be cultivated into true mastery of the subject.

I don't think cramming then taking a couple of weeks break and then cramming again is an effective strategy. It only works with tasks that have a slow rate of decay.

Good for making progress on a side project and yet horrible for learning.

I think slow and steady combined with a few sprints each year is optimal. I do 1 hour/day most of the year and then 40+ hours/week for two non-consecutive months.

Well, I guess we're mostly arguing definitions here. If you do 1 hour/day, but fit as much knowledge as you can into that hour, are you cramming for one hour, or taking it slow and steady?

In practical terms, spaced repetition is with no doubt the best method for remembering things, and you could either see it as slow and steady, or repetetive cramming. For understanding things, I don't know of a good method; but it seems like understanding 'keeps better' as long as you remember the relevant facts.

AboutSource Built by g1lg1l

Hackerly is an independent reader for Hacker News, built on the public HN API. Not affiliated with Y Combinator.