[ art / civ / cult / cyb / diy / drg / feels / layer / lit / λ / q / r / sci / sec / tech / w / zzz ] archive provided by lainchan.jp

lainchan archive - /λ/ - 18797

File: 1474220453419.png (397.44 KB, 300x223, 1327272753031.png)


Would anyone be willing to update the gentooman's library? It's a great resource but a lot of the books are outdated by 3 or 4 editions. Even 5 sometimes.

I have a fuarrrkton of books but it's not realistic to do a project like that alone. If there is interest we could start a new version of it.


protip: editions don't usually matter with the content of textbooks, unless you're taking some class that requires the right exercizes.


Gentooman library is too bloated. It's unrealistic for one person to read all the books in there, so forget about updating it.
Ideally, books for each field must be recommended by a few people with great (20+ years) experience in a field, so that person interested would just read 5 best books about the field. But to do it, it would take a big effort finding really experience users that would recomend books.
Recommending K&R, SICP and TAOCP is no-brainer. But who really reads TAOCP (entirely)? Basically no one.


File: 1474325811913.png (109.76 KB, 200x113, 1316705397_99.jpg)

is mostly correct with this.
The library doesn't really need to be kept up to date. Its true that alot of this material is outdated, but most of if not all of the material is still relevent. this goes double for textbooks were the difference between most editions is some typos, rephased paragraphs, and in the case of Java textbooks, Java.FX instead of Java.Swing.

Realistically the library only exists for the sake of information hoarding. Everyone knows that nobody is actually going to stop fapping and read a damn e-book. Even with my personal IRL library, with few exceptsions have I ever read a refence book cover-to-cover. Theoretical books are dry, Research Papers usually inconsequential, and nobody actually reads a programming referance book, they just look at the pretty code snipets. Unless you are alergic to stackexchange, you are not going to be flipping through these books for help in anything.


That's the whole point of it. You're not supposed to read everything, but whatever you want to learn will be there if you need it. I used my books just like i would use google, it's a great thing to have.


What kind of book gets three new editions in 5 years? The only ones a I can think of are references fro programming languages.
Protip: You don't need those.

The library isn't supposed to cover the average person needs, but most of everyone's needs (simply because we can in this time and age).

Although, I agree that it could use some curation. TAOCP is a bad example, but there are a couple of books that are actually bad and having them in might mislead people into thinking they're decent.


>What kind of book gets three new editions in 5 years?
Some books were added in already outdated i guess. But i agree with you, it's too soon.


I'm familiar with SICP but could you expand the title for K&R and TAOCP? I'd like to look into those texts


Never mind. I answered my own request.

K&R is The C Programming Language

TAOCP is The Art of Computer Programming


>Although, I agree that it could use some curation. TAOCP is a bad example, but there are a couple of books that are actually bad and having them in might mislead people into thinking they're decent.

Are you claiming that TAOCP is a bad text?


nobody would read all that, you're supposed to take the books that are good for you.
On the other hand, I share your feeling, the gentooman library pretty much is like kicking dead whales down the beach.
It's filled with soykafty pdf editions on mostly useless books, and sometimes not even the books themselves but some pdf pointing you to some other page where you can either pay money or download some windows virus.
The listing is not even comprehensible, and sometimes the books are not complete. And most of them are for java/javascript/php kids.
A lot of those books are pretty much useless too.
The library is huge, yet it has so little of value. It'd be 9001x better if it were just a few titles (say a few hundred) complete, with a good edition (not, say, fuarrrking Tiger), and be it a fuarrrking good book.
There are soykafloads of javascript books coming out every year, and java books, and as >>18826 pointed out, books that need constant revisions and therefore are not really relevant in the long term and would be more sensible to look in the web when you need them.
Yet there are a lot of books that are great reads, and aren't even listed in the gentooman library.


We should ask around renowed universities what courses they have for freshmen in CS and then what books they recommend for those courses.
If a book comes up often, it should go into the gentoomenv2 library into /basics/
If its a book like Cormen which is used everywhere in intrduction to algorithm class, yet is pretty difficult, we could include an easier version (like algortihms for dummies.pdf or headfirstAlgorithms.pdf) into /prebasics/.

In /meta/ we could put all the classics that are not about a particular concept or technique but about the field of CS in itself: The tao of programming, The mythical man month, Code complete...

Then /algortihms/ all books about algorithms that go beyond the basics. Could include Encyclopedias of algorithms, research papers on rarely used but brilliant algorithms...

Of course following this principle there would be /os/. While there should be 1-3 books about OSes in /basic/ in /os/ there would be the more advanced concepts or books about exotic OSes like Lisp Machines.

Same for /db/: Database books beyond basic level or nonSQL database systems or even distributed databases.

I think you get the structure I have in mind.
Perhaps we could structure the folder structure in the torrent hierarchically, making it easier for users to navigate it. /basics/ and /meta/ would be toplevel with /basics/ holding both the book files about basic concepts as well as the folders like /db/ /os/...
The user, even with no CS experience could then navigate down the tree, reading along ever book (assuming infinite time) and be ready for the next "level" once he has read at least one book at the current level.


It would be nice if everything could be converted to .pdf and .epub. I have a few I've converted from .chm to .epub


You know this is a very bright idea. I usually teach myself concepts by searching for "Concept + Syllabus"; you find so many universities have post their resources for their classes, and leave them up. What is really nice, is when they have quizzes and lecture notes.



I believe this is from the gentoomans library


There's just too much stuff there. There's no way to tell what's worth reading and what's soykaf, what kind of reading order should be followed for which books, which of the books are introductory, intermediate or advanced, etc.

It needs more structure and careful curation.


No offense lain, but why not see for yourself? I have been using the library since it's creation, and never have I had an issue with it. It seems like you would like someone to hold your hand--and to that all I have to say is that not everyone on this board is "learning to program".

All you have to do is check if they have the book you need, and if they don't no problem. Now let's say you don't know the book you need, and you want to know if this book is worth reading. A simple search engine will show you not only reviews of the book (which are often done by professionals), but also summaries and table of contents.

It sounds like you want to:
and frankly for something free that's too much to ask. Why not say what you are looking for and perhaps me or another lain could help you find a good book?


It's not that I want someone to hold my hand. It's that I want a more closely curated library, instead of one where every book available gets thrown into the pile. Asking everyone looking at the library to search for reviews of every book to see if it's soykaf, instead of centralised curation, looks like duplication of effort to me. But if that's the way everyone else prefers to do it, then it's fine. I already have a backlog a mile long, so the problem doesn't actually affect me.


You are going about this the wrong way if you really want to learn.
Having a curated list of books is having someone give you a program to follow.

Think about all these "soykaf" books and why they were made. Would the authors really have put all that energy into making something that was useless? It might be that you simply do not see their true value.

If you spend all your time wondering what direction you want to go in then you will go nowhere.


echo what this lain said.
You aren't thinking about it with the right mentality. Most of those books have something to offer. In fact, I find myself looking at books that would be deemed outdated all the time. It's a library, not a learn how to programming resource. A library can contain books on learning to program... but it doesn't need to only cater to you. If you were to spend time with books to figure out if they work for you or not, you would be much closer to your goal. I've been programming for the last 16 or so years, and I've read some books people have praised, and found them to be soykaf for myself.
When you enter a library, they don't put all the "good" books at the front, nor are all the new books that sit at the front "good", instead you go to a library with the intention of finding a specific book, or heading to a section to discover a book. There is no reason this shouldn't be the same.

Your attitude is the same one you see on programming boards by beginners asking " What type of WM do I need to program" "What text editor should I use to program" "What IDE should I use to program" "which resource is the best to start programming". Because you seem young, I insist that if you fix this bad habit today, you will be a much more complete user in the future.


I agree; I was very excited when I first downloaded the gentooman library, but I found it useless. A tiny library with only the most excellent resources would be far better.


I don't see why you'd need a static list of books if sites like http://gen.lib.rus.ec/ exist.
You can just look at up-to-date blog posts if you want to know what the current best read is and go download that.
If you were to compile a new Gentoomen library, chances are, it'd be outdated in another 5 years.


What constitutes an excellent resource to you though may not be an excellent resource to me. I'm not sure you actually know how to program.


>what constitutes an excellent resource to yo though may not be an excellent resource to me
Well, that's why the theoretically superior tiny library would contain only the most excellent resources as opposed to the most excellent resource (on any given relevant topic).

>I'm not sure you actually know how to program



Why not just have a gentooman officially endorsed list and description of books on the readme?


Sturgeon's Law.



I kinda like this Idea, I downloaded a copy of the library a while ago. It would be nice to have a wiki sorta thing that provided a brief description of the content in each book, maybe that way you could search in the library for keywords like 'language design'.