Girotto * about my computer studies

* about my computer studies

This blog will go through some changes. Why?

Well, the simple reason is: there are many important things demanding my attention, resulting in less time spent studying HTML5 and its related technologies. But I’ve been studying many other things, and I believe this blog is very important to me, as a learning log and a public ‘portfolio’, and to others, who might benefit from something written in here.

So, what will I write about now? Basically, every interesting thing that comes up to me in this semester of studies. That involves HCI, Software Engineering, QA, design patterns, SOA and specially compilers. This last one will likely be the one that I talk about the most, since I’m currently attending CSE440 - Compiler Construction here at ASU and am currently developing a portuguese pseudo-code interpreter at my university in Brazil as my capstone project. This project uses ANTLR at its core, and the development is done with Java.

I have interesting topics in mind, and will be writing about them soon. :-)

Yet another conference - Microsoft HTML5 Devcamp

I haven’t even written about the conference in SF yet, and am already attending another one. This time, it’s the Microsoft HTML5 Devcamp, here in the sunny and hot Phoenix.

It hasn’t started yet, but so far so good: free breakfast! :-)

P.S.: I’ll write about both conferences soon. But it’s important to note that they are a bit different: while the one in San Francisco was (mainly) for people who already knew (at least some) HTML5, presenting it’s multiple features, this one seems to be more introductory, presenting an introduction to HTML5 and CSS3, and some basic technologies.

HTML5 Developer Conference - San Francisco

So far, I’ve watched two talks: Doug Crockford’s and Giorgio Sardo’s (one of the guys behing cuttherope.ie).

Doug’s wasn’t technical. It was entitled “What Would Crockford do”, and he basically talked about Yahoo’s situation, and what he would do if he was the companies next CEO.

Sardo’s was technical, but he only showed a couple of code snippets. He will have a more in-depth talk later on the day, in which he promised to share more code and answer many questions.

Later on (not today, though), I’ll talk more about each talk. I’m taking notes. :)

Books I read: Javascript: The Good Parts

You know a book is good when you keep reading it even when you’re on finals week. That’s what happened with Crockford’s Javascript: The Good Parts. His book is great.

This is the book’s table of contents:

1. Good Parts
2. Grammar
3. Objects
4. Functions
5. Inheritance
6. Arrays
7. Regular Expressions
8. Methods
9. Style
10. Beautiful Features
11. Awful Parts (appendix A)
12. Bad Parts (appendix B)
13. JSLint (appendix)
14. Syntax Diagrams (appendix)
15. JSON (appendix)

First of all, a little about me. I have a good background with object oriented programming (basically with Java), and have been ‘programming’ with Javascript for a while. With this background, I’ve been able to read and understand basically everything he wrote in the book. My lack of experience with real Javascript programming, however, keeps me from seeing the big picture. So, this is certainly a book that I’ll read again later on my Javascript road.

Of the things that stood out to me, was all the talk about Javascript’s prototypal inheritance. This is a concept that I need to exercice more, given my background with ‘classical’ inheritance, as seen in Java.

Chapter 4 (functions) was pretty interesting. Javascript relies heavily in functions, so it’s important to understand some things about them. He talks about four invocation patterns (Method, Function, Constructor and Apply - all affecting the this pointer), the fact that Javascript does not have block scope (only function, meaning that if you declare a variable inside an if clause, for example, it’ll still be accessible from outside of the if), and the very interesting Closure. What’s this? Inner functions (function within a function - just like Inception’s dream within a dream) always have access to the parameters and variables of the functions that they’re defined in. Even after method execution. This can be used along with constructors to create private variables, something that many people think is not possible in Javascript.

I also liked Chapter 9 - Style. Nope, it has nothing to do with CSS, but with coding style. The chapter itself is pretty short, but I consider it to be directly connected with Chapter 10 and Appendix A and B. He gives some advices on how to to make your code speak. There are some pretty useful things, and some that many may consider a little too radical. He tells us not to use incremental or decremental operators (++ and —). While this may cause confusion in some circunstances, I honestly see no harm in writing index++; instead of index = index + 1;.

Lastly, the chapter about JSLint is also pretty sweet. He basically provides us with a tool that checks many things that he wrote there. But here is were a subtle problem lies. By using JSLint (which IS a great tool), his recommendations stop being recommendations and start becoming rules. You have to follow them, or your code will not do well on his tool. I brought up this theme a couple of posts ago, so just refer to this and that.

Anyway, this is an awesome book, that I will surely read again, later on my quest to be a front-end Jedi. I’m also pretty excited to see Crockford’s speech on the HTML5 Developer Conference later on this month. :-)

Now, I’ll hold Javascript back a little bit, and already have started reading Mastering CSS, Vol. 1, by Smashing Magazine. Seems a pretty good book, and wasn’t expensive at all. Although the title doesn’t explicitly state it, this book is really about CSS3’s new features. As usual, I’ll say a few words about it when I’m done with it.

Polyfills

Two posts ago, I talked about the cool website HTML5 Please. It basically is an abstraction over caniuse.com, providing you with simple a answer to if you should use a given feature or not.

But when you check the site by yourself, you’ll see that many APIs have the Use with Polyfill. Ok, sounds good. But…

What is a polyfill?

A Polyfill - a term coined by Remy Sharp -, is simply a way to fill the holes that some browsers leave on the new APIs, allowing you to use them as if nothing was wrong. Under the covers, they provide mechanisms compatible with those browsers - like flash, if available -, but expose it through the normal Javascript API that you’d expect to be available.

Here’s the article were Remy explains it.

And here’s a list of cross browser polyfills, brought to us by the awesome guys behind Modernizr.

* { box-sizing: border-box }

While skimming through Paul Irish’s blog, I found a very interesting article.

This is the situation: you set an elements width to, say, 300px. Then you set its padding to 50px. You save the file, reload the page, look at the result, curse it, and go back to the code to set the width to 250px. This happens because of how the box model works.

Basically, you can overcome this behavior with this simple line of code:

* { -moz-box-sizing: border-box; -webkit-box-sizing: border-box; box-sizing: border-box; }

Haven’t tested it yet, but surely will.

Another cool detail is that people are always worried about using the * operator alone, due to performance issues. I had read that before, and always avoided it. Well, he says that it shouldn’t be on your top concerns. And, I guess I agree.