Live data from Hacker News

Ask HN: How to RTFM?

news.ycombinator.com

41–47 of 47 posts

Re: Ask HN: How to RTFM?

#41
post #28

Just so you know, manuals aren't read like story books, from beginning to end. They're arranged in a tree structure with the root/trunk giving a rough Overview or Introduction on whatever the manual is about, the branches explaining various features in broad terms and ordered by their importance, and the leaves explaining the most arcane details you might eventually need to know about. What I do is first take note of…

> Just so you know, manuals aren't read like story books, from beginning to end.

Yes they can be! And if you're working a great deal with some technology, it behooves you to read the manual from cover to cover. Try it!

I've read a number of programming language standards from cover to cover, after which my facility greatly improved.

After I read the GNU Make manual from cover to cover, I was able to understand complex Makefiles like never before, never mind generate them from scratch.

At university I read a Motorola 68000 architecture spec from cover to cover (while writing an emulator). My ability to code MC68K assembly shot through the roof.

I haven't read the entire POSIX standard cover-to-cover (not even the much smaller classic 1990 version), but sections of it, like the complete description of the Shell Command Language. My ability to write robust shell scripts shot through the roof.

I've read the complete manuals for GNU Flex and Bison, plus the POSIX descriptions of these and other materials. Complete whiz here.

Highly recommend cover-to-cover. Problems is, standards have gotten a lot thicker. Everything takes more time. You have to choose wisely and prioritize.

> branches explaining various features in broad terms and ordered by their importance, and the leaves explaining the most arcane details you might eventually need to know about.

That simply isn't how most technical manuals work. Once you get past the intros and overviews, the arcane details are spread throughout all the descriptions like raisins in a pudding. Generally, the description of a specific feature or construct is completely given in one lump, with all the arcane details. If there are four paragraphs, the first two may be general, the second two full of arcane details. If you don't read the arcane details, you could be screwed. Oftentimes, you're reading this stuff because of the arcane details, not because you don't get the gist of it.

Re: Ask HN: How to RTFM?

#42
> Whenever I have tried to go through the manual, the kind of language used has diminished my interest and ended up in my giving it up.

I think you have to change your attitude and get used to reading that language, and learn how to write it, too.

It is the language of engineering specifications.

I recommend reading at least one standard (e.g. programming language) in its entirety.

Things refer to other things; at some point, you need the transitive closure of the whole darn thing in your head to properly understand any part.

Re: Ask HN: How to RTFM?

#43

>if someone actually reads the manual Uh, I thought everyone does. >I don't see how manuals can be useful in the real sense I'm...not sure how you can say this. Maybe you mean, reading a manual cover-to-cover. I don't think "RTFM" means that. Don't read the whole thing if it's boring! But I find once I'm well into using/learning some piece of software, and want to know everything, reading the entire manual is fascina…

> > >if someone actually reads the manual > Uh, I thought everyone does. ... have ... have you ever met another programmer?

Haha... well, in person, probably not. I've been programming since the 80s, and most of the 'voices' I've heard have been writers on programming, besides more recently online on here, SE etc. Not programming professionally, just doing whatever I wanted. I got a copy of Stroustrup 20 years ago, but never read most of it. Have a few other massive C++ books, and a few smaller ones, that I consulted constantly though.

Recently I got into LaTeX, read almost every book on LaTeX + TeX available (a couple of dozen), all very good. Got into Prolog, (about a dozen books), all very good... All the books on AWK are great. Same with Lisp. Sure, Stack Exchange is a wonderful resource, hard to imagine doing without it. It's like having a teacher - read the books, then consult the teacher for questions that you can't immediately find answers for in the book. Some things are done better online: bash has a few great websites, and w3schools is great for css/js/html.

Re: Ask HN: How to RTFM?

#44
I would like to start by declaring that very, very few software manuals are can be read cover-to-cover and offer great benefit. There are some exceptions to this--the one that comes to mind the most is Ableton Live's manual.

Clearly, the most common use case for a manual is when a user has encountered a situation where they don't know how to use a product as it should be used. Here are two examples of different scenarios where RTFM as ended in different ways:

* Keras: I'm currently optimizing the architecture of an artificial neural network at work, and I needed a little more detail on how quickly the network was training. The Keras documentation quickly guided me to the Callback section, where I quickly learned what callbacks were, and then found the CSVLogger callback. Not only did the documentation link directly to the source, but I thought the documentation gave a good high level overview of what the function did so I could see that clearly this was the function that was going to solve my problem.

* Vim: I use vim quite frequently at work as my main editor, and some of my co-workers also use it. Recently, I had asked a coworker how he configures vim for his projects, and he told me about Vim sessions. I asked him for more details about how vim sessions and tabs worked, and he referred me to the vim documentation. While the explanations were clear, I've always held the opinion that the vim documentation is a bit of pain to navigate, and suffers greatly due to the lack of example vimrc snippets. Maybe others think differently than I, but I've learned much more vim through blogs and vim-tutor than I have through vim's own documentation.

The biggest piece of advice I can offer is that regardless of the quality of the documentation there is always some guarantee that the official documentation is not the only documentation out there. It is quite rare that I ever truly grasp an answer or new piece of information on the first read; I often need examples or further reading for me to incorporate that information into my everyday life.

All that being said, start with official documentation. I can't say for certain, but in my experience it has always been the most authoritative and the highest quality writing on the software itself. Seek out blog posts and unofficial technical writing for further examples or different writing styles that might be easier to grasp.

Re: Ask HN: How to RTFM?

#45
I would suggest taking a look at the book How to Read a Book by Mortimer Adler. The title of the book does not do it justice.

It covers analytical and syntopical reading. But it also addresses how to read different types of texts

Re: Ask HN: How to RTFM?

#46

Earlier quoted context omitted.

> > >if someone actually reads the manual > Uh, I thought everyone does. ... have ... have you ever met another programmer?

The old ones did this. Want to learn C? Read K&R. Want to learn Perl? Read the Camel book. Want to learn C++? Read Stroustrup. That was the way we learned . These days? Not so much. Stroustrup is now big enough to use in self-defense. I'm not sure there ever was an equivalent book for Java. (There is one for Go, however.) React? Angular? TypeScript? Do they have authoritative books? If they do, is that the canonical…

FWIW Rust has "The Rust Book" as their primary long-form documentation, separate from their reference documentation. It's a common starting point for people new to the language. I read it in chunks as I was getting started.

I'm not sure how it compares to those works of the past (perhaps less detailed?), but I found it much more approachable than most languages' documentation and more in-depth than most languages' tutorials.

The old ways may not be entirely lost.

Re: Ask HN: How to RTFM?

#47

Earlier quoted context omitted.

The old ones did this. Want to learn C? Read K&R. Want to learn Perl? Read the Camel book. Want to learn C++? Read Stroustrup. That was the way we learned . These days? Not so much. Stroustrup is now big enough to use in self-defense. I'm not sure there ever was an equivalent book for Java. (There is one for Go, however.) React? Angular? TypeScript? Do they have authoritative books? If they do, is that the canonical…

FWIW Rust has "The Rust Book" as their primary long-form documentation, separate from their reference documentation. It's a common starting point for people new to the language. I read it in chunks as I was getting started. I'm not sure how it compares to those works of the past (perhaps less detailed?), but I found it much more approachable than most languages' documentation and more in-depth than most languages' tu…

All three of the books your parent mentions were explicit influences on TRPL. So yes, it lives on :)
Post reply on HN