Hacker News Reader

Comment by etep | Hacker News Reader

This: Git is so amazingly simple to use that APress, a single publisher, needs to have three different books on how to use it. It’s so simple that Atlassian and GitHub both felt a need to write their own online tutorials to try to clarify the main Git tutorial on the actual Git website.

View on HN

I think the author mistakenly assumes that the number of books on a topic is somewhat proportional to its complexity level. It's actually inverse - how many books do you have on advanced calculus vs. how to tie a tie?

People write books to make money. The easier a topic, the easier is to write a book on it. The more popular a topic, the more profitable is to write a book on it. Hence, you get most books on things that are cool and relatively simple.

So, going by the number of books and tutorials metric, java is a much more complicated programming language than erlang? Have you seen how many books and tutorials java needs?

Yes, Java is orders of magnitude more complex than erlang. What kind of comparison were you trying to make?

I knew that no matter what languages I picked for my examples, that someone on hn was going to respond that yes, in fact language B is much simpler than language A. I believe you understood the comparison I was trying to make and I'm not interested in a discussion about how simple erlang really is, it's besides the point..

No I do not understand the comparison. Java's syntax, virtual machine, libraries, everything is orders of magnitude more complex than the erlang stack. And indeed, proper introduction and reference of Java can fill a whole bookshelf, where as a couple of printed technical reports does the same for erlang.

Back to the original topic, git is also way more complex than a DVCS needs to be (compare e.g. with monotone, or just about any other DVCS). And any DVCS is way more complex than even the most difficult to use centralized repositories. So I think the OP's point is valid. There is at least corrolation going on here.

So you do believe that if there are more books and tutorials about a topic that means its more complex?

It's Bayesian evidence that the subject might be more complex, yes.

No, it's not. It's less complicated and more popular, hence has more books written about it.

No, its both more complicated and more popular, and both factors contribute to it having more books written about it.

You do know that there are many multiple SVN books out there in addition to the red book?

But you don't really need them. The red book tells you pretty much all there is to know (I say that based on memory from years ago, the last time I used SVN. I liked SVN. It was easy to use, easy to understand, and did everything we needed for our project (a half a dozen developers all working in one office). It was such a breath of fresh air coming from our previous VCS (Visual SourceSafe).

And you don't really _need_ a git book either. There is a lot of nice information on the git site and the man pages (even considering the man pages notoriety in not being consistent in terminology).

Which is why the nvie.com "git flow" (that everyone points to as the ideal branching model for git) is not actually on the git site?

And as for git man pages... I have been hacking on Linux and complex codebases for 20+ years at this point but have never been so befuddled as I have been by the git man pages.

Reply on news.ycombinator.com