[rickhanlonii]

That seems like a cop out to me. Textbooks are by far the best resource you have available for your class work and even though they take the most effort, they tell you exactly what you need to know enough to know what you don't know.

[laughinghan]

Sorry, what? Literally the first time I've heard anyone claim that textbooks are better than in-person one-on-one tutoring by an educator, which is what we're comparing to, right? (Getting help in-person from the TA.)

Like sure, certain exercises may be more ideal with a group of peers, or may be more ideal if the student does it while the instructor stays hands-off; but if you had to pick just one of all the possible ways to learn something, until I read this comment I thought it was quite universal that everyone would say yeah, one-on-one in-person tutoring with someone trained in the subject matter is the single best way to learn something. Which is why, in spite of being the most expensive option, tutoring is still a thriving industry.

[easytiger]

> Sorry, what? Literally the first time I've heard anyone claim that textbooks are better than in-person one-on-one tutoring by an educator, which is what we're comparing to, right? (Getting help in-person from the TA.)

A huge number of software developers are autodidacts. The implication is you don't know many people.

[laughinghan]

I know plenty of autodidacts, and arguably was one. I'm almost certain not a single one would say that if they could've afforded to be tutored one-on-one instead of reading on their own, they would've still chosen reading on their own instead. They learned from textbooks because the cost and inconvenience of tutoring was prohibitive, not because they're like "pffft, the tutor was just slowing me down, in the time spent with them I would've learned way more from a textbook".

[reitanqild]

OK, here is one counterpoint.

I actively do avoid asking questions even when it is free.

Not only in computing. I really did not do to well on piano lessons. My siblings who where just taught the basics (IIRC) and then left to their own ended up playing the piano (and other instruments) because it was fun.

Of course in computing this is reinforced from time to time by e.g.:

- by visiting stackoverflow and see how many (IIRC again) of the questions that has helped me most are closed as not constructive,

- or by a colleague who seems they will use any opportunity to talk about something being junior-level stuff and then go on to waste a lot of everyones time by rewriting the whole thing.

[therealdrag0]

Having gone through two degrees, I have rarely found getting help from others to be more efficient than learning on demand myself. I definitely deliberately avoided tutoring apart form the occasional sampling.

[humanrebar]

I've had exceptional teachers and tutors at times, but I always learned best from textbooks and articles. Information density (ideas per second) in text is much higher and I can scan ahead at my own speed. And with text, I can time-shift my learning. You only get personal tutoring at certain hours and only for a limited amount of time. You can't save your tutor on your hard drive and pull it up later.

Regular evaluation and correction from a human being is useful, though, so that I don't mistakenly believe I know something before I do. But there are subjects where self-evaluation (math quizzes, small software projects) can serve the same purpose. But it certainly speeds things up to have someone set a curriculum.

[Chris2048]

How many times have you heard the opposite? Is it discussed often? A good book is is way more convenient than a good tutor.

[laughinghan]

I can't recall any discussions directly comparing textbooks with tutoring, but I can recall many discussions that assume in-person one-on-one teaching to be the gold standard, and all other methods are merely cost-effective compromises. It's implied that textbooks, while not as effective as interacting directly with the textbook author, are an excellent compromise because the author's pedagogy can be widely dispersed for relatively low cost (compared to the author personally tutoring every reader).

While convenience could arguably be considered part of what makes something "better", I think it's also commonly not. Like, you might say a restaurant a half-mile away is "better" than one across the street because its food is tastier, even if the one across the street is so much more convenient that in practice, you eat at the less tasty place five times as often.

In this context it's not clear-cut, but even when re-reading great-grandparent it still seems to me to not be considering convenience to be what makes textbooks better:

> Textbooks are by far the best resource you have available for your class work and even though they take the most effort, they tell you exactly what you need to know enough to know what you don't know.

If anything, great-grandparent seems to be emphasizing an aspect of textbooks that's less convenient than a tutor.

[lotharbot]

> "Textbooks are by far the best resource"

for people with certain styles of learning. I've always found lecture with some degree of interaction to be of much higher value than the textbook for me.

[rickhanlonii]

So I didn't mean that the textbook is the best resource for the concepts, just for the coursework. Side note, though, learning styles are a myth [1].

[1]: http://www.wired.com/2015/01/need-know-learning-styles-myth-...

[heinrich5991]

I can't find any references as to why learning styles are a myth in that article.

[rickhanlonii]

As mentioned in the article:

> surprisingly few studies of this format have produced supporting evidence for learning styles; far more evidence (such as this study) runs counter to the myth.

There's been a number of meta-analysis done (e.g. [1]), and they overwhelmingly conclude that there's just no support for learning styles. So, it's a myth until proven otherwise. See also Willingham's book based on his research [2], and his FAQ here which links to more research [3].

[1]: https://www.indwes.edu/cli/research/meta-analysis-summary---...

[2]: http://www.amazon.com/Why-Dont-Students-Like-School/dp/04705...

[3]: http://www.danielwillingham.com/learning-styles-faq.html

[wonnage]

For you, maybe - the commenter below me preferred lecture and I didn't learn anything until I did the project and trolled the mailing lists asking (and eventually answering) questions. Some people do better with the human interaction.

[rickhanlonii]

I don't think I was clear enough. I'm not saying that textbooks are the best resource for the concepts, but that they're the best resource for the course work.

The comment I replied to said that people can be excused because they "don't know enough to know what you don't know well enough". But the textbook is exactly that information. Most CS courses plan the lectures, notes, homework, tests, etc off the textbook. So the textbook tells you enough to know what you don't know well enough so you can look it up on mailing lists or pick it up in lecture.

[Cthulhu_]

Depends on the textbook tbf - a lot of textbooks are really, really bad. Same with basically all sources of information. They pretty much all make assumptions about knowledge level, mainly because they're written by people that (think they) know a lot about the subject. In a lot of cases (the PHP tutorials online are a poignant example), they don't or spread misinformation (yeah just use mysql_query ololooo)