Hacker News new | past | comments | ask | show | jobs | submit login
Reddit's favorite programming books, from 3.5B comments (redditfavorites.com)
124 points by mgdo on Dec 2, 2017 | hide | past | favorite | 32 comments

I've read most of those books, and while I wouldn't discourage anyone from reading them, I don't think it's a great reading list either. You'll have to dig through a bit of cruft to get to the gems.

Partly, it's that programming books have such low lifespan that the ones that still get recommended after a decade are often deeply assimilated into programmer culture already. For example, when I read Clean Code and Pragmatic Programmer last year, I didn't find any of it to be interesting or remotely surprising - but I have been working at "agile" shops for a while, and perhaps those ideas were more revelatory when the books first came out.

Here's a more fun reading list, based off my own tastes. Maybe some of these are on the list, but I found it hard to scroll past a certain point as well:

- Expert C Programming: Deep C Secrets

- The Nature of Computation

- The Little Schemer

- Programming Pearls

The Little Schemer is indeed a fun read, and it's relatively short and digestible (compared to SICP which I have on my reading list but is daunting to me). If you are used to languages with imperative loops it will give you a nice perspective on recursion. Slightly related but I found myself using Reduce/Fold pattern a lot more in my everyday languages after reading it.

Maybe following a tutorial in Erlang/Elixir might have the same effect, in addition to being more actual and practical, but also maybe less fun.

The only “programming” books I ever found useful we’re essentially computer science books that cover the basis of computational theory and computing concepts.

Anything that is more than a foundation is likely to be out of date by the time it’s published.

This especially holds true for modern (FoTM) languages and concepts.

So for the most part I stay away from practical books execept for the few ones that are set in stone.

Beyond that “philosophy” books can also be useful even if they are outdated simply because it makes you think and exposes you to other points of view and concepts that you can draw on.

Looking for recommendations on a "best" one or two books to read on SQL / PostgreSQL.

Have any of you read Mastering PostgreSQL in Application Development[+]? I'm aware it's a fairly recent title, but I'm in need of something like that and am wondering if it's worth the money. The free sample chapter looks promising, I'll give it that.

I'm working my way through the official PostgreSQL docs, and they're really good, but I find myself struggling a bit (a lot). I'm fluent in a handful of programming languages (dabbled in many), and appreciate abstractions like HOFs and monads, but I'm finding it difficult to express myself in SQL, and that tells me some underlying concepts haven't quite clicked.

[+] https://masteringpostgresql.com/

I don't know about books, but I've found the website use the index luke to be helpful with sql in general. I've also heard good things about joe celko's books.

I'm sure the comments that were analyzed span many years: 2 Perl books and "Javascript: The Good Parts".

The Art of Computer Programming, SICP and Introduction to algorithms are books that are commonly recommended but very few people have finished reading them.

"Javascript: The Good Parts" is often shown as an example of just how bad JavaScript is because it's almost like a pocket edition.

Solid list, but didn’t see any of Scott Meyers C++ books there, which is pretty surprising. Unlike other C++ books he tends to discuss the “why” along with the “what”. Nice to see Reddit knows about SICP though!

> Nice to see Reddit knows about SICP though!

...truly a tome from the abyss of obscurity.

Seems super common and overhyped for the HN or /prog/ community, sure. Though, at 5 jobs over 3 years I haven't worked with a single person who's ever heard of SICP (or really any programming book for that matter). That includes security & big Data work at a Fortune 500.

Quite few professional programmers (in the grand scheme of things) actually read programming books unless they're forced to.

> Quite few professionals (in the grand scheme of things) actually read programming books unless they're forced to.

FTFY. I agree, it's a shame but at the same time I understand.

Why should a barber, mechanic or whatever be expected to read books about programming? Your fix makes the sentence easier to read, but only with a trade-off

I can't tell if you're joking or not. If you read HN ten years ago, you would hear about SICP and Lisp constantly. Makes sense when you consider who made this site.

I am, I know and I know.

needs more Aho books

also no physics or maths ?

oh and Loomis is everywhere in the drawing list, pretty cool

Nice to see SICP growing.

Outside of the standard literature, those are some of the worst book covers I've ever seen. I refuse to look into this any further, as I don't trust reddit's taste anyways.

Literally judging books by their covers?

I now go out of my way to do so, in most shopping. I figure if their marketing team is tasteless, there's not much hope for the rest of the business's operations. (In the sense that bad marketing implies bad product; good marketing can go either way)

I mean just compare SICP and TAOCP covers to.. head first *; the latter look like self-help trash in b&n.

Now O'Reilly, they've got covers you can really believe in

> I figure if their marketing team is tasteless, there's not much hope for the rest of the business's operations.

This is the most ridiculous thing I've read all day. I can't even think of a comparison to further illustrate the stupidness of this sentence.

Its the primary public facing function of a consumer-facing business. Its really a lot more difficult to argue that they shouldn't be judged on for it, when its specifically the main thing a business is trying to be judged on. The only immediate argument I can think of is that

A) mass marketing is, and should be, frowned upon as a field, due to their use of extremely abusive strategies

But that doesn't mean you shouldn't appreciate those who advertise well, and that all of marketing should be frowned upon

B) marketing doesn't imply anything about the business

But in a lot of cases, it does, because they're either in a market where they should (most consumer products), or they're in a market where there's no need to (free software, extremely cheap retailers. business to business, non-finalized products, middlemen like amazon, monoprice, etc). And if they're in the former, and they pull off a shitshow, I don't see how its reasonable to not judge them for it.

There's a reason a large chunk of silicon valley puts such a massive strain on look&feel, UX, and all the shitty css frameworks, despite tech having the history of not giving the slightest shit.

They've moved into a market where marketing does play a major role, and their success at it should be indicative of other successes. The backend stuff might not have to care (but, it turns out, they still do; just a different fashion), but any final product landing on your phone as an "app" certainly does care, and they should be judged for trying, and failing.

Looking at the Author, Reviews and the Table of Contents is probably a better heuristic.

That said, for things that aren't books, a marketing department that is complete dogshit might point to an opportunity. Berkshire Hathaway doesn't have great marketing. It doesn't need to.

If you go on Alibaba and find a merchant that has a great prospectus and a broad range of products, that's most likely a reseller. That means they take their share of the pie, and if you try to get something manufactured, you'll be working through a proxy. Better to find their manufacturer instead.

Non-comsumer facing, retailers, middlemen, manufacturers etc don't follow the same rules of marketing; they still have to do it, but in a different fashion. IE alibaba and amazon are "advertising" based on how cheap they are; it would be a bad show for them to put out extremely nice, expensive ads for their e-shop. Its absurd to see ads for a car parts supplier.

Marketing exists in the context of its market, and must be judged in that light. Sometimes cheap, shitty marketing is the best thing to do.

And to be clear, books to programmers is one of those fields where marketing matters; it just be because I'm more involved here than any other community, but its one of the heaviest trend-following communities I'm aware of, and seems to have been that way since the 80's.

Also toc and reviews is a second level filter; author may be first-level, mostly dependent on whether you actually know who he is; cover is almost always the first point of contact; physical or e-store, you're most likely looking at a list of covers.

It would be absurd not to realize this, and not to try taking advantage of it. And you should be judged for both, if you don't do it well.

And it would be absurd if you, as a consumer, are not aware, and do not take advantage of this situation (by acknowledging, and applying, a filter on covers!)

I try to imagine what kind of a marketing team came up with the cover of SICP. How did they plan the reaction of the target audience, A/B tested the cover art on focus groups, etc. Entertaining!

I like to imagine they didn't; Sussman had an idea about the book he liked, someone drew up a decent image (or even found it somewhere), and they slapped it on. My favorite kind of marketing is the ones where they don't have to do much work to get it good; or at least don't show it.

Indeed. I couldn't imagine it being otherwise. What I wrote was ironic, of course.

> Now O'Reilly, they've got covers you can really believe in

I agree that the Head First books look gross, but there are a lot of bad O'Reilly books (on the inside) even if the animal looks cool.

Sure, but the relationship I'm claiming is one way:

bad marketing implies bad product

Good marketing implies nothing

Its just a first-level filter

> Sure, but the relationship I'm claiming is one way:

> bad marketing implies bad product

That definitely doesn’t hold up to any scrutiny.

“Reddit’s favorite books that very few have actually read”

Many of these covers are from the 90s, so they obviously do not use the latest cover design trends. I've seen lots of lists recommending books for programmers and many of the ones listed here are routinely among them. They are classics in the field.

Most of the ones from the 90's are the ones I was referring to as the standard fare (recommended every, single, time: TAOCP, SICP, intro to alg, js good parts, etc), and their covers are a hell of a lot better than the newer books..

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact