The Puerto Rican Rails Dude

Well, well, even a week after Zed Shaw’s infamous rant on the Ruby and Rails communities, all the talk generated by it continues. While it has died down enough, I feel that the repercussions of the rant are still going on. And I’ll be writing about one I’ve been disturbingly seen more and more in the past week.

One of Zed’s problems with Dave Thomas was due to his book, Programming Ruby (or “Pickaxe” – weird how some people know that book only by the Pickaxe name). Here’s one of the paragraphs he wrote about that particular book:

That’s right, take a look at the original topics and you’ll see that Dave’s book is nothing but a giant me-too book that seems to appeal to the average OOP coder of 2001. However, the average OOP coder using Java, C++, or C# wasn’t doing much meta-programming then, and what Dave presented was nothing more than a book that said, “Hey look, you can do all the stuff you’re doing now, and make NO money at it.”

He goes on to explain a lot of the shortcomings of the book, like the fact that the chapter about classes was demonstrated by designing some sort of Karaoke jukebox machine. I do agree with these specific points Zed makes. I remember reading the book for the first time, and just skimming the classes chapter because it felt awkward that these very important terms were being explained this way. It was also confusing as well. I just read the basics (how to create classes, inheritance, access control, etc.) and skipped most of it. Also, failing to include one of Ruby’s strong suits (which is the ease of meta-programming in this particular programming language) seems very weird.

However, this isn’t what I wanted to talk about. My issue is that after Zed pointed these things out in his rant, I’ve been seeing a lot of negative comments towards this book – something I hadn’t encountered before. In fact, I bought this book because everywhere I read on the Internet, it specified that this was the book to own on Ruby. In fact, let me show you a screenshot of the current reviews this book has on Amazon.com:

See anything interesting there? Yeah, there are only 52 reviews, but 43 of those reviewers – a whopping 82.6% – gave this book either four or five stars. To me, for a supposedly “shitty” book (as I’ve read a lot of people call this book in the past week), this is a rather high positive rating.

So, are these people honestly criticizing the book by their own will, or simply just want to “follow the leader” and bash Dave’s book without thinking for themselves? I find it extremely odd that a lot of criticism is aimed towards this book in such a short period of time – the past week since Zed’s rant. Were people scared or something of giving their honest take on this book for fear that the Ruby zealots would find out where he/she lived and crucify them or something? Something just doesn’t seem right.

Now, while I’ve expressed before how much I enjoy the work of Dave Thomas and others who publish the Pragmatic Programmers books, do know that I’m not biased in any way, shape or form. I’m probably one of the most objective guys you’ll ever meet. If I think something sucks, I’ll gladly say so, as long as I can back my words up. But honestly, I don’t think the Pickaxe book sucks. Far from it. This book has helped many (myself included) find out about the awesome features of Ruby (well, most of them, at least). To ignore the fact that this book helped jump-start Ruby usage in the in the Western Hemisphere is really a disservice to it.

I think Programming Ruby is an awesome book and an excellent reference to the Ruby language. As a book to learn the language, I wouldn’t count on it to be the “be all to end all” Ruby book. To those who think there are programming books like this, you’re totally wrong. In my short programming career, I’ve come to know that one book for a specific language or technology isn’t enough. Obviously, that doesn’t mean you should buy all the books on a particular topic. But more than one, at most two or three books, should help you greatly. No two authors think alike, and there’s always some stuff missing from one book that another has, and vice versa.

So to all those newly-minted Pickaxe “haters”, I suggest you take an honest look at yourself and think if you’re really hating this book because you didn’t like it. If you didn’t like it from the get-go, good for you. I hope you find another book that helps you learn Ruby. But leave the rest of us who learned a great deal from this book alone. We don’t need a Zed-wannabe running around this joint.

UPDATE: After I posted this, I immediately E-Mailed David Heinemeier Hansson (yes, the creator of Rails), because I was curious on how he learned the Ruby language. I was surprised he responded in mere minutes, and his response was even more surprising:

I learned Ruby in large parts from the original Pickaxe and thought it was a great book. No, it didn't cover everything. And I picked up some metaprogramming tricks form The Ruby Way, especially Chapter 5, as well. But I was very happy to have it at the time.

Seems like David agrees with my thoughts on this subject. I bet that most of the big names in the Ruby world also learned the language from Dave’s book. I thank David for that quick and honest response.

Like I mentioned a while back in this blog, I really like the books released by The Pragmatic Programmers. They’re pretty easy to read and teach a lot in the process. Although Programming Ruby (known to many on as the ‘Pickaxe’ book) is getting some flak after Zed Shaw’s take on it (which I agree on, although I’ll give my own take on it in the near future), you can’t deny their books are among the best in their specific topics.

I’m a big fan of the Rails Recipes book (and no, it’s not because they actually put a pilón, a typical Puerto Rican food prep utensil, on the cover). When you learn a specific technology, be it a programming language, database engine, framework, or anything else, you’ll want to find some real-world usage to apply what you’ve learned. Books like Rails Recipes give you an entire list of specific things you probably thought of doing for your own application, but didn’t know how to exactly do it. It’s a great tool when you’re developing your own stuff.

Of course, thanks to the speed of Rails, that book is a bit outdated. So I was really psyched that a new version, named Advanced Rails Recipes, was going to be released in March 2008. So psyched, that I actually went over to their site and purchased the beta PDF (and, of course, I’ll be getting the paper book wne it’s released). I couldn’t wait to read it and apply some recipes to my current projects.

First, I would like to discuss the current system that’s used for beta PDF’s in the Pragmatic Programmers site. The site is relatively easy to use, and really well done. Once I created an account and purchased the PDF and paper book, I received an E-Mail stating that my PDF was ready to be downloaded. That was it. Withing a few minutes, I already had my PDF, with “This book was prepared exclusively for Dennis Martinez” in the cover – neato! A couple of days passed by, and I received an E-Mail stating that the beta PDF was updated. All I needed to do was to go into my account on their site, and a link immediately appeared in the main page, which took me directly where to regenerate the PDF (which apparently is run by gerbils – no joke!). A few minutes passed, and the gerbils sent an E-Mail with a link to my updated PDF. Okay, so they might not really be gerbils who work on the PDF’s, but I really like it when companies go with these informal jokes while still be very helpful. Kudos to these guys for making a very user-friendly site.

This book is filled with recipes from various Rails users all around the globe who have contributed their work to this book. One thing that took me by surprise is the fact that I expected this book to be pretty much incomplete, with tons of errors all around, but it really wasn’t like that at all. I found one or two spelling errors, and the recipes I tried worked straight out of the box. That’s a great thing, because this is why I went ahead and purchased the book three months in advance. I wanted to read up on the techniques used by people with much more real-world Rails experience than I.

In particular, I wanted to read up on the chapter about Capistrano and deployments. This, to me, is one of the most touchy issues in Rails nowadays. It’s something that everyone seems to do differently with mixed results. The recipes included here aren’t de-facto deployment strategies, but nice techniques, such as how to generate config files on the fly when deploying to a remote server, how to safeguard your database passwords so that they’re not in your source code repository, and many other techniques.

Sure, most of these recipes are readily available on the Internet, if you search hard enough. But when you’re developing a major web application, the last thing you want to do is to fire up Google and spend the next 15+ minutes searching and trying out pieces of code to see if they work for you. This book avoids all that. Not only does it show you how to do things, it also explains why and when you should use them.

Major kudos to the author, Mike Clark, for making such an awesome book. Even though it’s still in beta, I really recommend this book to anyone who wants to expand their Rails knowledge with bits and pieces of information from major Rails players. Once the ‘full release’ of the book is out in March, I’ll make sure to do a more informative review. For now, don’t be afraid to get this book because it’s tagged as ‘Beta’. That’s far from the truth, it seems.

Over the past week and a half, I’ve been busting my ass trying to learn the ropes of Behavior-Driven Development with RSpec. I had been interested in learning either Test-Driven Development or Behavior-Driven Development for quite a while, so I just jumped on one of the two and test the waters out for a while.

I’ve been reading a lot on TDD and BDD for the past couple of months, yet never took the time to implement it to my current projects. The main reason is because I thought it would take me too much time to learn and would bog me down too much, when all I wanted was a workable application running as soon as possible. But, an opportunity presented itself (more on that in the future), so I chose the development technique that seemed better for me: BDD.

Reading about RSpec through other blogs on the Internet, it just seemed so intuitive to use that for testing, I kinda thought it was too good to be true. So after installing RSpec and getting it running on my application, I realized it was true after all. It’s ridiculously easy to write your own tests with RSpec, and it’s almost plain English (just like good ol’ Ruby). It includes built-in stubs and mocks to simulate parts of code, so you can even drop fixtures if you want (although admittedly, a lot of the shortcomings fixtures had were fixed with Rails 2.0). In fact, here’s a simple controller test:

describe ProductsController do

  before(:each) do
    @product = mock_model(Product, :to_param => "1")
    Product.stub!(:find).and_return(@product)
  end

  it "should call the action successfully" do
    get :show, :id => "1"
    response.should be_success
  end

  it "should find the product" do
    Product.should_receive(:find).with("1").and_return(@product)
    get :show, :id => "1"
  end

  it "should assign the found product for the view" do
    get :show, :id => "1"
    assigns[:product].should eql(@product)
  end

end

This test actually works for testing the show action in a controller. I wrote this just to show how readable the code is. I never knew testing would be so readable! And like I said, once you wrap your head around mocks and stubs, it’s ridiculously easy to implement. As a great bonus, RSpec can also integrate with rcov, a tool that actually checks what parts of your code are tested. It’s a great tool for making sure you achieve 100% coverage of your code.

Unfortunately, I didn’t actually do any BDD this time around, as I already had a good chunk of my application up and running already. But now that I have discovered the joys of RSpec, all my upcoming products will be fully done using BDD techniques: I won’t write a line of code until I write the behavioural test before.

A couple of ceveats, though. Hey, nothing’s perfect! First off, if you use different types of plugins for your application, you’ll find some parts of it difficult to test in RSpec. Most plugins use the Ruby’s own Test::Unit testing framework to test their own functionality, but parts in your code where you use these plugins can get you the first time around. Also, if you’re like me, and already have written code for your app, you’ll start finding different ways to fix your code to conform to RSpec testing. This isn’t necessarily a bad thing, as this is most likely a warning sign that you need to refactor that piece of your code anyway. But it’s takes a bit of your time away, so be prepared for this.

In short, RSpec is so easy to use, I don’t see why people shouldn’t use it in their own projects. I suggest every Rails developer should look into RSpec (or any other testing method, really) and stop being lazy. I know I was, but never more!

RSpec References

Ryan Bates from Railscasts has a great screencast that can serve as an intro to testing controllers. And the best part is that it’s free!

* Testing Controllers – http://railscasts.com/episodes/71

Geoffrey Grosenbach over at PeepCode has an excellent three-part series on RSpec, with about three hours of RSpec goodness! No, they’re not free, but the $9.00 per episode is truly worth it. This was my secret weapon in learning RSpec in less than two weeks – The best 27 bucks I’ve spent in my software development career!

* RSpec Basics – http://peepcode.com/products/rspec-basics
* RSpec Mocks and Models – http://peepcode.com/products/rspec-mocks-and-models
* RSpec Controllers and Tools – http://peepcode.com/products/rspec-controllers-and-tools

Before I begin, I just want to wish everyone a Happy New Year! May 2008 bring happiness, peace and prosperity to all.

I’ve been keeping myself very busy lately with RSpec and Behavior-Driven Development, basically learning the ropes and how all the pieces fit together. For now, I’m totally enjoying it. But more on that in a future post.

Now, I know that everyone who read Zed Shaw’s rant towards most of the Ruby and Rails communities will have an opinion on this. But I’ll give my own thoughts on it. All I hope is that someone doesn’t read this and think “Who the hell is this guy to give an opinion?” I might not be a ’somebody’ in the Ruby or Rails communities at the moment, but I would really like to be part of those communities sometimes in the near future.

Upon first glance, Zed’s rant seems like a completely immature piece, just looking to damage the reputations of certainl people and companies. But if you read closely, ignoring the unprofessional language scattered throughout the text, there’s a whole lot of valid thoughts and reasoning to this entire rant.

Most of his attacks are aimed at two people: Kevin Clark and Dave Thomas. Kevin Clark has been a pretty big part of the Rails community, regularly contributing code and such, and Dave Thomas of course is the author of possibly the most well-known Ruby and Rails books in the market. His story on Kevin is that they possibly never got along and clashed multiple times, while the story on Dave is that supposedly Zed had a fix for a pretty serious bug in Ruby, yet Dave and others ‘threatened’ Zed to not release it. For what reason, it’s not clear in the rant. Now, about these allegations, I don’t know whether they’re true or not (there’s always two sides to a story). But from my own views, I think any environment has these types of problems all the time. My own workplace can be used as personal experience on these manners. There’s always someone who wants to be smarter and better than you, and for some unknown reason they go out of their way to make sure they come out looking better than you can. It’s stupid, but it’s just human nature, I guess. I bet almost any other open-source community is the same.

He also goes on to write about Thoughtworks, a software consulting company that jumped on the Rails bandwagon a while back. Zed’s beef with them is the fact that they charge a shitload of cash while providing not-so-great work in return. Isn’t this is the case with almost all software consultancy places? I can name a few off the top of my head here in Puerto Rico. In fact, all you need to do is go to a website one of these ‘expert software consultants’ made, look at how the site is built, and anyone with an eye for software development standard practices can name a dozen things they would change immediately. This is no surprise here.

I do commend Zed for not making this a 100% negative jab towards Ruby and Rails. He included some people who he knows and have helped him out or were unlike those who he vilified before. Like I said, all communities have their share of bad apples, mostly people who want to be most widely known at your own expense. So the fact that he names some people who were cool to him shows that.

I’ve read a lot of other blogs where people are dismissing Zed for the way he expressed his views, that he burned his bridges and what not. But that’s what he apparently wanted. He said he’s not happy being part of the Rails community, so he wants out. This is his way of getting out. Now, I may not agree with the way he did this at all. You never know when you need to cross a bridge you burned in the past, after all. But being in the position he was, seeing and knowing a lot of things that went down, he has a valid opinion, and he’s simply entitled to it.

I think Zed partly wrote this rant not to bash everyone associated to Ruby or Rails, but to try and help out, in his own way. He knows a lot more about the community than most of us will probably ever know. So hopefully some good comes out of him exposing some dirty details on how the Ruby and Rails world is run. In the end, it simply boils down to this: It’s one man’s opinion. No matter how important he is (or rather, was) to the Rails world, one man isn’t enough to kill it, in my opinion.

Zed, if you miraculously read this, best of luck to you in the future, buddy. Hope the rant was worth it!