WALP (We Are Lazy People)

laziness

Yesterday, one of my colleagues reminded me that I once said that I’m going to do the minimum to maintain project X. Project X is not its real name but maybe I should back up a few steps and start with providing some background:
one of the things that happens at Netflix pretty often is that teams change, responsibilities shift and as a result projects are handed off between teams. Project X is such project, it was written long time ago (more than two years back which is basically forever in our industry) by a guy that now manages another team, and was handed off to a team that currently doesn’t exist, and when that team was disassembled I was given this project.

So yesterday, the guy who wrote project X reminded my that I said I’m planning to do the minimum to maintain it. I was surprised because one of the things that doesn’t fit me is to come up with such a statement, and then I remembered: WALP.

Some people will describe it like this, but I just rather say that I’ll do the maximum to be in a condition where I have to do the minimum in order to maintain an existing project. Sounds confusing ? well, not really…

Project X was written in Python and had two major components (class-files) which became monolithic: functions that perform multiple tasks (which goes against SRP) with hundreds lines of code, a few levels of nested loops – complicated to follow and understand, a nightmare to debug and horror to test.

So, in order to be able to “do the minimum to maintain it” one of the first things I did was refactoring these two classes: breaking the logic into smaller functions, minimizing the number of arguments, working according to DRY and SRP, renaming variables and functions, adding unit-tests and while I was “shaking the tree” a few bugs were discovered and fixed as well.

There wasn’t much “fun” in it, it’s not like writing a project from scratch which is really tempting to most engineers but usually is a bad idea. It’s tempting because you think “I’ll get it right this time” or “I can do much better than that” and let’s be honest: when you’re maintaining someone else’s code – it will never “belong” to you, that other guy will always get the credit for writing this project – even if the code changed so much that it doesn’t look anything like it originally did…

Back to the story, the refactoring paid off, and it wasn’t even in the long term:
* a few bugs were found and fixed
* bugs that were discovered later on, were easier to debug and fix
* I had to write a few additional features, all were implemented very quickly, easily, without any pain and were easy to test

So yes, I AM lazy, and I’m willing to work really hard to sustain it…

WALP (We Are Lazy People)

How I wrote FeedMe

Breastfeeding

A few days ago I released the first version of FeedMe, a small Android app I wrote over the last few weekends. I promised my wife to write it after our first daughter was born and we found ourselves keeping tracks of her feedings using notes that were all over the house. After a few nights that we woke up, the baby was crying, and we couldn’t find the notes anywhere around (had to go to the living room, kitchen etc) I’ve had it and decided to build this side project.

By the time I got to it, our daughter was already a few months old, and she slept through the night and there was no real need, but I promised my wife that for the second one – I’ll make sure to build it in time.

And I did!

Our second daughter was born a couple of days ago, only a few days after I released the app.

2014-01-03-13.06.45-225x300

Needless to say that the app is free, and I intend to keep it so, and not only that: I plan to open source it and I’ll appreciate if other hackers will decide to contribute, add features, improve the code, or do whatever they want with it.

I will release the code together with other Bitbucket and GitHub projects, as soon as I fix the code a bit and make it presentable (I did some quick & dirty hacking in order to make it work cause I was short in time and I don’t have experience developing apps).

I’d love to get your feedback, but please don’t send feature-requests because I got plenty features & ideas I want to implement but I lack the time to do so…

How I wrote FeedMe

Grails is Groovy!

Grails

Grails is a framework that uses the advantages of Groovy, a dynamic type language which is almost a superset of Java, and the principle of “convention over configuration” in order to supply a very easy to use and quick to develop web-applications. It relies on the JVM and Spring as well as some other technologies that are out of scope for this post (Hibernate, templates etc).

Grails is based on MVC design pattern, in case you’re not familiar with it yet – you should, but until you do I’ll give you the highlights:
MVC stands for Model-View-Controller, which separates concerns and thus eliminates dependencies between the data (model), display (view) and business logic (controller). The controller is the entry point to the app, the HTTP request is dispatched to the relevant controller which drive the data from the Model, and might process it using one or more of the Service classes and will choose the proper View for display. The View doesn’t have to be strictly “display” it could also be JSON/XML or any other representation of the data.

In the rest of this post we’ll create the helloworld example and run it – with one major added value: I’ll help you avoid the pitfalls that I ran into. Ready ?

If you already have JDK installed – good for you, if not, you’ll have to install it first (not the JRE!)
It took me less than 10 minutes to download, install and run my first “Hello World!”
All you have to do is:
1. extract the zip file into the location you want it to be
2. add the environment variable GRAILS_HOME – make it point to that folder
3. update the PATH variable to PATH="$PATH:$GRAILS_HOME/bin"

After we’re done with the installation, navigate to the location where you want the project to be created (command prompt if you’re a windows user) and type:
grails create-app hello
after a few seconds, a new directory will be created by the surprising name of…
yes you guessed right: “hello” (convention over configuration anyone?)
Lets get in by typing cd hello and create our first controller. Can you guess the command ?
grails create-controller hello
if you missed this one – don’t be upset, you’ll get the next one.
One of the lines that were printed after the last command should have been:
Created file grails-app/controllers/hello/HelloController.groovy
Open a text-editor and edit HelloController.groovy:

package hello
class HelloController {

    def index() {
        render "Hello World!"
    }
}

That’s it! now, can you guess how to run our app ? that’s right:
grails run-app
it starts running, looking good, it even prints:
Server running. Browse to http://localhost:8080/hello
which seems awesome, but before we get to copy the URL and paste it into the browser…

*** java.lang.instrument ASSERTION FAILED ***: "!errorOutstanding" with message
transform method call failed at ../../../src/share/instrument/JPLISAgent.c line:
844
Exception in thread "main"
| Error Forked Grails VM exited with error

Hey! that’s not a nice thing to do!
Don’t worry guys, I hear you, here’s how we can fix it:
go to: hello/grails-app/conf
and edit the file BuildConfig.groovy – comment out the following section using /* … */ like this:

/*
grails.project.fork = [
// configure settings for compilation JVM, note that if you alter the Groovy version forked compilation is required
// compile: [maxMemory: 256, minMemory: 64, debug: false, maxPerm: 256, daemon:true],
// configure settings for the test-app JVM, uses the daemon by default
test: [maxMemory: 768, minMemory: 64, debug: false, maxPerm: 256, daemon:true],
// configure settings for the run-app JVM
run: [maxMemory: 768, minMemory: 64, debug: false, maxPerm: 256, forkReserve:false],
// configure settings for the run-war JVM
war: [maxMemory: 768, minMemory: 64, debug: false, maxPerm: 256, forkReserve:false],
// configure settings for the Console UI JVM
console: [maxMemory: 768, minMemory: 64, debug: false, maxPerm: 256]
]
*/

Now for most of the readers it should work: if you’ll grails run-app paste the following URL into your browser:
http://localhost:8080/hello
you should see the link to the controller that prints “Hello World!

But… if you’re like me… all the wrong things WILL happen: another issue I had was that port 8080 was already in use. To work around that, you can run your app on a different port:
grails -Dserver.port=8090 run-app

That’s it, now you should definitely see in your browser the following page:

sc-300x279

and once you’ll click the controller-link, you’ll receive a sweet “Hello World!” in Grails.

Update: there is actually an open bug for the issue described above, and a fix should be deployed on the next release (Grails 2.3.3)

Grails is Groovy!

A simple calculator in Ruby

Today I ran (again) into the following question:

  • Write a function that doesn’t use eval that calculates input strings with operators +,-,*,/ eg. “5+5*6+4/2″ should output 37

The first time I ran into this question I designed a solution, in Java, which had a SimpleCalculator class, and enam Operation which supported the four basic arithmetic operations: +-*/ each of which had an apply method etc. Very object oriented.

When I read this question today, I figured that it would be a nice exercise to do in Ruby – a few minutes later the result spread, elegantly, over less than 20 lines of code (and I bet that a professional Rubiest can do it in less)!

def is_number? expr
  return false if expr.nil?
  expr = "#{expr}"              # we need this condition in case the expr is a number
  expr.match /^(\d+|\d+\.\d+)$/ # since match() is defined only for strings
end

def calc(expr)  
  return expr.to_i if is_number? expr
  expr.gsub!(" ","") # clean the string from whitespaces
  # pay attention to the order: + and - should come before * and /
  # can you figure out why ?
  arr = expr.split /\+/
  return arr.inject(0){|x,y| calc(x) + calc(y) } if arr.size > 1
  arr = expr.split /\-/  
  return arr.inject(0){|x,y| calc(x) - calc(y) } if arr.size > 1
  arr = expr.split /\*/
  return arr.inject(1){|x,y| calc(x) * calc(y) } if arr.size > 1
  arr = expr.split /\//
  return arr.inject   {|x,y| calc(x) / calc(y) } if arr.size > 1
end

puts calc("5+5* 6+4/2.0")
#output 37

Do you have a better/shorter/more elegant solution ?
Please post it in the comments section!

A simple calculator in Ruby

Inheritance overflow…

Been fiddling with Java lately when I ran into the following cute quiz about inheritance.

Reading some articles about what’s Good/Bad in Java – I ran into the following code from an article called why Java sucks?:

class Boo{}
class Buzz extends Boo{}

class Play{

    static void foo(Boo b){
        System.out.println("in foo(Boo b)");
    }

    static void foo(Buzz b){
        System.out.println("in foo(Buzz b)");
    }

    public static void main(String...args){
        Boo b = new Buzz();
        foo(b);
    }
}
  • What would you think will be printed ?
  • Do you know why ?

Answer: variable b is declared as a reference to object of type Boo during compilation, and that’s why foo(Boo b) is chosen over the “more intuitive” option.

The reason that it works this way is that Java (as well as many other OO languages) is single dispatch. There are other languages like Common Lisp that supports “multiple dispatch”: the decision which version of a method will be called is based both on the class of the object as well as the run-time types of the arguments!

By the way, “multiple dispatch” behavior can be achieved in “single dispatch” Languages by using the Visitor pattern (described in “design patterns” book).

I started a list of my own about the things I like/don’t-like in Java and here’s the outcome:

Plus Minus
write once – run anywhere still slower than C++
Many good libraries (both java/javax as well as 3rd parties) no multiple inheritence
Garbage collection horrible character escaping when using regex
boxing can’t pass methods as a parameter
inner classes + anonymous classes handling primitives differently (why not treat them as objects?)
reduces boilerplate code every version (generics, try with resources, filters) still too verbose (open a file…)
If inner class wants to use a variable of the parent – it must be declared final
checked exceptions can be a pain with interfaces
Inheritance overflow…

The Free Electron Legend

free-electron

A few days ago I read the single most productive engineer that you’re ever going to meet – Free Electron which glorifies the ultimate hacker, for example, there’s a story about this computer geek which re-wrote the whole DAO layer in one week – something which took two other engineers approximately six months…

Personally, I don’t like the glorification that is done around coding “rock-stars” and programming geniuses. I rather have a collaborative environment, such that you won’t be afraid to make mistakes, but unfortunately in most of the organizations I ran into it’s more important to cover your butt and be a politician than actually be a good engineer (which includes to be able to make mistakes too!).

Anyways. it reminded me of another story, one that happened back in the 80’s. I heard this one from Mayer Goldbreg, a professor that taught a compilation course I took during my studies for a B.Sc.degree. On that course we built all our projects using a weird language called  Scheme which is a functional programming language that has a syntax that doesn’t look very appealing to the majority of object oriented programmers:

(let loop ((n 1))
  (if (> n 10)
      '()
      (cons n
            (loop (+ n 1)))))

===> (1 2 3 4 5 6 7 8 9 10)

It took some time getting used to the weird syntax with all the nested brackets (oh god – just thinking about all the times I counted opening/closing brackets…)

But I’m diverting… so back to the story: Mayer had a friend which, not surprisingly, was such a free electron. He worked for a big company that used some low-level programming language (let’s call it X) to maintain their platform. As you can imagine, maintenance was hell, not to mention adding new features or finding bugs. But that friend came up with this great idea: he wanted to write a compiler from Scheme to X – after his project will be completed life will be sweet – everything will become much easier!

This guy had just one problem, his boss was not so enthusiastic about that idea… After a few arguments his boss told him specifically that he should forget about it – but like any good free electron he didn’t…

It didn’t take him long to build that compiler, sure enough, he had a few bugs – but he fixed them and sooner than he himself expected – he would produce so many lines of generated (and unreadable) code – and it was working great: whenever he got a new assignment – it would practically take him minutes to code it in Scheme and then he would run the compiler and take another 1 hour of coffee break. He was producing the same amount of work of the entire division!

Soon enough, his manager heard the talks and came to his cubical furious. He yelled at him and then fired him. But if you think that this is the end of the story – you’re wrong.

It didn’t take long for the higher management to find out that suddenly all the tasks were taking much longer – and after a few weeks the story got all the way up – this manager call, but now he was practically begging him to come back and maintain his compiler…
He didn’t.

UPDATE: I just ran into the following session and I couldn’t agree more with these guys!

The Free Electron Legend

Setting up Spring web-project on IntelliJ using Maven

Till yesterday I used eclipse when I wanted to work with the combination of Spring-web/Maven projects. To be more exact I was using STS which is a version of eclipse called Juno (if I’m not mistaken – but if I do – please correct me) that comes with the relevant plugins already installed (Spring and Maven’s jars for example) and it wears an ugly green skin. It also comes with vFabric server which is a web-container (like Tomcat) that was made by VMWare, which bought Spring in order to leverage Spring to push their products.

Now, if there’s an IDE that I hate – that’s Eclipse. It’s just so slow and buggy… You can spend a couple of hours on something that doesn’t work just to find out that if you create the project from scratch – everything’s peachy, and no, clean/re-built/restart/computer-restart wouldn’t have fixed it…

If IntelliJ wasn’t an option I would definitely choose Netbeans which is the fastest Java IDE I’ve experienced, but intelliJ is a tool you fall in love with

So yesterday I decided that I’ve had it, and that I’m migrating my Spring projects to IntelliJ. In case you have an existing Maven project – it’s really easy to import and start working, but then I tried setting up a project from scratch, and after fighting with it for a couple of days I finally found out how to do it (by the way, I read a few other posts in regards which didn’t help…). I know, it should be easier and that’s a couple of black-points to JetBrains, but the effort was worthy!

In order that you won’t have to go through the same painful experience which I did, here’s a short guide on how to set up a Spring Web environment on IntelliJ. So without  a further adu:

1. create a new project from scratch

i1

2. select “Maven module” as a type and click “Next” after filling out the project name and path

i2

3. check the option “create from archetype” and choose “maven-archetype-webapp” from under “org.apache.maven.archetypes”. If you don’t have that option go to step 4, otherwise you can skip it and go directly to step 5.

i3

4. In case we didn’t have the archetype, we’ll click on “add archetype” and add it manually as follows:

GroupId: org.apache.maven.archetypes
ArtifactId: maven-archetype-webapp
version: RELEASE

i4

5. before you click on “Finish” check if the parameter M2_HOME is defined, if not, you can define it as an environment variable, or alternatively you can set it up right here:

i5

6. once you click “Finish” a pom.xml file will be generated and some, other stuff will run on your screen, you should also see the following option to enable auto-import for Maven projects – click it!

i6

7. goto “edit configurations”, click on the “+” button and choose the local version of Tomcat

i8

i7

8. Name your server, check the option to “Build artifacts” and click on the “…” right next to it. Choose the .war file and continue

i9

9. go to the “deployment” tab and click the “+” button to add an Artifact. Again, choose the war file

i11

i10

10. That’s it – we’re done! If you’ll expend your project you should be able to see something like the following screenshot.

i12

Click the green “Play” button and after Tomcat starts, you should be able to view “Hello World” jsp version on localhost

i13

Setting up Spring web-project on IntelliJ using Maven

Front-end development

I got to start by saying that I am not a front-end developer.
That said, today, when there are so many available web-stacks, it becomes inevitable to handle at least a bit of UI.

On this post, I’ll share some info that’ll help you on your way to become a front-end developer, so let’s begin:

Learn CSS!

Learning HTML is not very complicated but if you know nothing about HTML you should probably start there… I assume that you have some familiarity with HTML and that’s why we started here. CSS is a very good practice of separating styling from content. There are many web-sites that will teach you CSS and it’s very easy to Google something you’re looking for, but, if you really know little or nothing at all in regards, I recommend a course called “30 CSS Selectors You MUST memrise” in a beautiful website: memrise – a wonderful free site (there’s also a mobile app) that’ll teach you anything from languages and history to art & literature and much more. In order to dynamically change CSS on the page (and see the result without switching to the editor back and forth and refreshing the page) I highly recommend using LiveCSSEditor – a plugin to chrome which I can’t live without!

Another two plugins to Chrome I find very useful are: MeasureIt! and ColorZilla

Javascript & jQuery

Codecademy is another awesome web-site for self-education, here you can study the basics (and more) of both Javascript and jQuery. This is critical since there’s no other way to manipulate DOM elements after creation, and in fact, whenever you hear the buzz-word HTML5 it refers to Javascript!

In order to debug JS/jQuery on-page I used to use a Firefox plugin called Firebug (which is awesome!), but today I use Chrome and the same developer-toolbox is available in Chrome by pressing F12. Two relevant tabs are “sources” – which allow you to add breakpoints, inspect objects etc, and “console” which enable you to view errors in the code that ran as well as to write a Javascript/jQuery expression and see what it evaluates to. Another good resource for JS is mozilla.

Performance

Google’s PageSpeed add-on to Chrome and Firefox’s YSlow! are two excellent extensions to test the performance of your web-page. Both supply a list of recommendations which, in case you decide to implement, will improve the loading speed of the page. Two websites that offer the same analysis are the famous Pingdom and webpagetest. TMO, you should use more than one tool to have a good coverage. That said, you shouldn’t get too obsessive with the performance grades.

General

Firefox still has two plugins that I really like:

  1. Tamper Data is a great tool that helps viewing the parameters that were submitted/received, usually via. forms (HTTP calls). If you want to dive deeper into lower levels of protocols like TCP/IP you’re welcome to installWireShark (previously called: Ethereal), but as a web-developer you’ll probably find yourself using only the filters HTTP/HTTPS anyway.
  2. I use Poster mainly to debug REST calls which are usually less of a “UI thing” and considered more as “back-end”. But, if you use AJAX to fetch information from the server you might find this plugin helpful.

If you’re interested in SEO perspective, there are toolbars like seoquake and sites like http://www.seositecheckup.com/  and http://www.hubspot.com that will provide a detailed analysis of your website.

keep learning!

You can do this by subscribing to weekly newsletters like Javascript weekly, online courses, reading blogs (you can search for relevant information via. redit and Hacker News), following pros on Twitter and asking questions on stackoverflow (make sure to answer some questions as well – give back to the community!)

Do you have a plugin/extension/web-site you found useful ?

Front-end development

JAX Conference 2013

IMG_20130604_134149

A few days ago I went with a friend to JAX conference 2013, a conference in which experts & leaders from the industry share their thoughts and experience with other Java community members.

We started with a session about testing HTML5 in all browsers using Java by Kevin Nilson. He got me hooked right from the beginning with two nice Javascript riddles:

The first:

var t = ['a','b','c',]; // pay attention to the last comma
console.log(t.length);

Is that a legal expression? (yes)
What’s the length? (hint: browser dependent…)

The second:

var abcData = ['a','b','c'];
var xyzData = ['X','Y','Z'];

function xyz(){
  var msg = '';
  for(i=0;i<3;i++){
      msg += xyzData[i];
  }
}

function aXYZbXYZcXYZ(){
  var msg = '';
  for(i=0;i<3;i++){
      msg += abcData[i];
      var tmp = xyz();
      msg += tmp
  }
} 

What’s going to be printed ?
My guess was “aXYZbXYZcXYZ” – but it was wrong…
If you want to see the answer – you can watch the video of the session (in the link above).

After the warm-up, Kevin discussed the limitations of using selenium to test mobile and the advantages of using tools like TestSwarm and QUnit.

Getting Started with WebSocket using Java by Arun Gupta introduced a light full-duplex bi-directional communication protocol between server and client. “Upgrading” HTTP protocol to WebSocket is done using HTTP-headers:

upgrade: websocket
connection: upgrade

for which the server supposed to return HTTP code:

101 - switching protocols

Full-duplex means that once communication has been established, both sides are peers in the manner that both can send information to each other simultaneously (unlike HTTP in which only the client triggers the request and the server can only respond). Another advantage of using Websocket is the speed: the thin frame that is used by the protocol allows sending bigger chunks of data more quickly than using HTTP which carries the overhead of sending headers in every request/response. Arun benchmarked the performance of WebSocket vs. REST and the difference were (as expected) huge, for example: sending 5,000 messages of 1KB each using REST took 55 seconds and only 1.2 seconds using WebSocket.

Another interesting session was: Apache TomEE, Java EE 6 Web Profile on Tomcat by David Blevins. TomEE (pronounced: “Tommy”) is a project that aims to combine plain vanilla Tomcat with some of the more common JavaEE components, such as EJB, JPA, JMS and more (they call it “Web-profile” and it contains about half of the “full profile”). Recently they passed the TCK (Technology Compatibility Kit) of the Web-profile, David says, without increasing Java memory!

IMG_20130604_143229
In The Road to Lambda (will be released in Java 8 – around sprint 2014) Mike Duigou showed the power of using lambdas which makes your code simpler (reduces the boilerplate), and as a bonus you get lazy evaluation and support for parallelism. Like Generics (introduced at Java 5) helps us abstract over types, the goal of Lambda is to help us better abstract over behavior (passing “behavior” = method, between objects using vars).

I got to admit that even though I like functional programming I hate the use of anonymous classes in Java, to me, it always felt like some kind of ugly hybrid between OOP and functional programming:
Lambda (closure) ties up code to data, and the equivalent to it in OO is… well, Objects.
So I was pretty skeptical about the whole idea but I’m glad to say that I was wrong!

The session started by using the fact that other languages like C++ and C# already introduced closures (bad reasoning TMO), but then Mike gave a couple of good motives, one of which was the transformation of the following code:

for(Shape s: shapes){
    if(s.getColor() == RED)
        s.setColor(BLUE);
}

to:

shapes.forEach(s->{
    if(s.getColor() == RED)
        s.setColor(BLUE);
});

At first glance it looks like there’s no much difference besides the different syntax, but what really matters here is what goes under the hood: the former piece of code translates into an iterator that traverse the shapes in a serial manner, while the latter enables the library to set its own implementation which could, for instance, support parallelism.

Another example which I really liked is

int sumOfWeights = shapes.stream
				.filter(s->s.getColor == BLUE)
				.map(s->s.getWeight())
				.sum();

or even a bit simpler:

int sumOfWeights = shapes.stream
				.filter(s->s.getColor == BLUE)
				.map(Shape::getWeight)
				.sum();

but what really impressed me was how easy it is to modify the code to use parallel processing:

int sumOfWeights = shapes.parallel()
				.filter(s->s.getColor == BLUE)
				.map(Shape::getWeight)
				.sum();

IMG_20130604_153015

Since this post is becoming too long I won’t write about other sessions, that said, I cannot not mention (and post links to) other great sessions such as Designing a Beautiful REST+JSON API by Les Hazlewood, The Spring 3.1, 3.2 and 4.0 Update by Josh Long and The Java EE 7 Platform: Boosting Productivity and Embracing HTML 5 by Arun Gupta.

See you in #JAXConf next year!

JAX Conference 2013

20 Tips for becoming a better programmer

1. There should be only one single exit point to each method (use if-else whenever needed).

2. When using if-else, make sure to place the shorter code snippet in the if:

if (cond) {
   
}
else {


...
...
.
.
.
}

3. Do not throw exceptions if you can avoid it, it makes your code much slower, if you feel like throwing something and then catching it – go play ball with your dog. If you don’t have a dog get one – they’re awesome!

4. Do not try to do many things on the same line – when you’ll get an error – it will be harder to debug, example how to NOT write your code:

String result = hasInformation()? getState() : (hasMoreInformation() ? getOtherState() : getState());

5. Look for code pieces that look the same and if you find any – refactor!

6. Meaningful names are a must. If you’re not sure, meditate on it for another minute. Still not sure? ask your colleagues for their opinion.

I’m still shocked everytime I find out that the following is not common knowledge:
7. Whenever you can use HashMap instead of List/Queue – use it!

And on the same note:
8. If you can use caching instead of I/O (usually DB) – use caching

9. If a nice and simple regex can do the job – use it!

10. Do not use regex for parsing (for example: HTML/XML/json) – there are special tools for that in every language, for example in Ruby, Java etc.

11. Print to Log. You should have at least two levels of logging: DEBUG and ERROR. The latter should be the default. Nice tip: you can send your self a text when a critical error occurs, by sending an email to your_mobile@your_carrier – look here for more details.

12. Use stackoverflow – not only for asking questions: Take a few minutes, every day, and try to answer questions – you’ll be surprised how much you’ll learn from it!

13. A rule of thumb: if your method is over 50 lines – split it. If your class is over 500 lines – split it. If you think you can’t split it – you’re doing something wrong. This recommendation is a “language dependent” and shouldn’t be taken “as is”: while 50-line method feels natural in Java/C# – it will be considered very long in more expressive languages such as Ruby & Python.

14. Writing a code which is “self explanatory” is great but also very hard, if you’re not sure how “obvious” your code is – use code-comments.

15. When writing code comments always assume that the reader doesn’t know what you’re trying to do. Be patient & explain. No one is going to *bug* you because your comment was too long…

16. You want to improve? read books, blogs, technical online newspapers join relevant groups on Linkedin, update yourself with the latest technologies, go to conferences, got the point ?

17. Practice makes perfect: solve code-challenges, join hackathons, go to meetups etc

18. Experiment with different IDEs until you find the one that “feels right”, study it carefully and make sure you know the major features (including plugins). Tune-up the keyboard shortcuts – it will make your workflow smoother.

19. Whenever you find a bug, before you fix it, write a unit-test that captures it. Make sure it does.

and my favorite:

20. Don’t get lazy – RTFM

We’ll finish with two quotes:

“Debugging is twice as hard as writing the code in the first place. Therefore, if you write the code as cleverly as possible, you are, by definition, not smart enough to debug it.”
– Brian Kernighan

“Always code as if the guy who ends up maintaining your code will be a violent psychopath who knows where you live.”
– Rick Osborne

Update:
Many thanks for Jeff Grigg, Tom Burton and Keith M. O’Reilly for their feedback which helped me improve this post – cheers!

20 Tips for becoming a better programmer