Little background about Scala. It is “Scala (pronounced /ˈskɑːlə, ˈskeɪlə/) is a multi-paradigm programming language designed to integrate features of object-oriented programming and functional programming. The name Scala stands for “scalable language“, signifying that it is designed to grow with the demands of its users.” –wikipedia

Even if you are not excited by the idea of functional programming which you might be after a few lines. Scala is a dynamically typed language written on top of JVM. This means you can use some very reliable well written Java libraries in its web framework — lift . Also you can potentially write applications for android. Although their is a project to get scripting languages to android, this could be an interesting option. Also since Scala is compiled, it should run faster than other scripting languages. (Either I am starting to like scala or I have had too much cheese.)

But more importantly talking about the functional programming . It is a paradigm which is closer to mathematical expression. Take for example How would you define second order derivative in an object oriented  language .Well till the idea of f(x) , functions in programming languages correlate well to their mathematical cousins. You send variables and get results. But what if you wish to do f square (x) . You call the function recursively and have a control structure to see how it is called . The control structure is an additional aspect we have added to the program not really related to the actual expression. Now see this little program and see how elegantly  can solve some problems .

def sigma(term, a, next, b)
     return term(a) + sigma(term, next(a), next, b)def term(x): return 1.0/(x * (x+2))
def term(x): return 1.0/(x * (x+2))
def next(x): return x + 4
sigma(term, 1, next, 1000)

This little code solves the following problem :

1/1*3 +1/5*7 …..

more importantly this is in directly correlated to its mathematical analog.

So how do I exactly  implement functional programming aspects in python …

Using lambda

f=lambda x: x*x

by doing this i have associated the object f with the expression x*x  so to do a second order equivalent would be simply

f(f(2))

which is a normal mathematical expression .

Closures: Nesting of functions

def cube(x): return x*x*x
def differentiate(f):
     return lambda x: (f(x+0.001) - f(x))/0.001

So now we come to our initial question , how to do second order differential .

d=differentiate(differentiate(cube))
d(3)

The beauty of the differentiate function is that it returns the expression(i.e the function) instead of the usual object so what we did here was same as f(f(x) in maths. Hence this forms a better analog to mathematical expression than usual programming paradigm

Another very interesting idea in python is that of decorators.

def addspam(fn):
    def new(*args):
        print "spam, spam, spam"
        return fn(*args)
    return new

@addspam
def useful(a, b):
      print a**2 + b**2
useful(3,4)

As you can see @addspam is decorator which recieves the function useful as one of the arguments . Infact decorators is a very clean way to implement things like checking for authentication in your program .

(Its really almost been years since my last blog post . I really didn’t do it all this while because I thought their are people who do it better than me . However this was an Idea that really excited me. I am just a how cool is that kind of blogger . So if you want the real thing just read my sources especially http://linuxgazette.net/109/pramode.html )

Apple, in India still goes and makes deals which tie consumers down to the same carriers. So carriers dictate their terms to the consumers. The price of 8 GB iphone is Rs. 16000 while the iPod touch is for 12,000. Difference is that, iPhone just has GSM chip built-in. Interestingly, the iPhone bought from Vodafone ties you to their two year plan. And Airtel is no better.

3G is another story. The carrier cannot start 3G service till one case gets resolved in court. That may take million years, if you have any idea of how the Indian courts operate. Resultantly, while carriers are nicely making money by selling super fast devices, there is no 3G network. MTNL and BSNL only offer 3G network in India but what do we know, your newly bought iPhone is tied to Airtel or Vodafone.

From Apples perspective in this situation, disconnecting oneself from carriers and being consumer friendly would be a nice move. But nice moves seem to miss Apple every now and then. Take the case of AppStore. Where are the details on selection processes? Transparency is missing completely.

Apple has sold many devices in India and the world over. Apple can still make more money, if they walk, this last mile.

• Compile One: Compile with gcc option -fprofile-arcs. This produces branching information for optimizing code further.
• Compile Second: Compile with option -fbranch-probabilities. This will use the branching information generated above to optimize the binary.

By following this simple step, the difference in benchmark for ffmpeg was 1.5 secs at 36.92 secs illustrated here. This is the kind of difference between “winner” and “loser” in a stereotypical benchmark.

By running some of the IO tests on whatever machine available, it was clear that Ubuntu Intrepid does not that perform badly in any of the tests as shown. Surprisingly, Ubuntu Hardy performs better than Intrepid.

Secondly looking at hardware specs was also a significant revelation.

Notice the processor used for Mac OS X has a Intel Core 2 Duo and the processor for Ubuntu tests has Intel Core 2. By further looking at Mac Mini Wikipedia page here, it is clear that 3rd generation Mac Mini hardware has been used for all Mac OS X tests while 2nd generation Mini for all Ubuntu related tests.

How much does this effect performance?
Quoting from comparison done between Core 2 and Core 2 Duo at tech.yahoo.com “Core 2 Duo computers outperform Core Duo computers on tasks across the board, from rendering graphics and video to spell-checking documents. The speed improvement varies widely, but on average it’s about 30 percent. That’s significant in an industry that thrives on tiny, incremental improvement.” That is in my opinion a huge difference in hardware for drawing an authoritative benchmark comparing two different operating systems.

Without questioning the intent for using two different hardware, looking at the benchmark again one can easily conclude that Linux performed better given Mac OS X considering Mac had the 30% processor advantage.

Here’s is how to find passwords for sites:

• Go to BugMeNot.com and type the URL of the website you need password for. As a sample try flickr.com
• Use the username/passwords available in the listing
• If the username does not work just use the dialog next to the details with yes/no

While BugMeNot is really convenient of surfing the web. The lack privacy is an aspect to look into.

• Virtualization is cheaper than having two desktops : Many people, I know run two devices because both are configured with different aims. One box being dedicated for playing ‘World of war craft’ and other being the workstation. Where ever you hear, I need Windows just for purpose XXX. Virtualize.
• Run a specific application: Let us assume for a second, I am Windows .Net developer and the application I am working on, is cross platform. Interestingly, I don’t know anything about setting these systems. I get virtualized installations with all these apps setup by my friend joe (the hacker). I happily upload my work and test in almost native environment.
• Virtualize test environment: Many developers have premium hardware which does not allow testing with constraint systems. Virtual system can be constrained with CPU usage, RAM or disk space to test border-line conditions
• Build network on desktop: Run networked application is the guest environments which interact with each other. Test at leisure, how two p2p applications behave from the safe portal of your desktop.
• Sandbox your desktop: Your Windows desktop is mission critical and yet, it needs access to Internet. You can setup Linux or Unix host and run Windows Desktop in Sandboxed environment

Without slipping into slush of jargon, virtualization also happens to be coolest way to flaunt your nerdery.

Java is not a favorite language for many. But you find yourself sometimes using it. Feeling of frustration simmers while setting up development environment, especially if you come across two variables – JAVA_HOME and CLASSPATH. Java newbies often get confused what these variables mean. What role do they play in java jigsaw puzzle?

Sole purpose of JAVA_HOME is to specify absolute path to tools.jar. This can allow a java application to invoke the compiler from inside of a code sequence. Applications like tomcat and ant often need this feature for their normal working.

CLASSPATH points to extra libraries you plan to include in your code. If you are referring to any symbol in namespace stored in particular library. That library should be in the CLASSPATH. Most of the time you would be putting hordes of jar files in CLASSPATH. But ones in a while you could be pointing your CLASSPATH to the top-level directory of your own code. You can also set this via -cp directive in java and javac.

To illustrate an example, I have written a rudimentary program to setup JAVA_HOME and CLASSPATH on unix platform

#!/usr/bin/env python

import os
import glob
SEP = os.sep

def JavaHome(pathList):
    for item in pathList:
        path = SEP.join([item, 'javac'])
        if os.access(path, os.X_OK):
            return os.path.dirname(os.path.dirname(os.path.realpath(path)))

def ClassPath(path='.'):
    classpath = ':'.join([os.path.abspath(a) for a in  glob.glob(SEP.join([path, "*.jar"]))])
    return classpath
def pathList():
    '''Make this OS specific'''
    return os.getenv('PATH').split(':')

if __name__=='__main__':
    path = pathList()
    print "export JAVA_HOME=" + JavaHome(path)
    print "export CLASSPATH=" + ClassPath()

If your javac is placed in lets say /usr/lib/jvm/java-6-sun/bin/javac. Your JAVA_HOME will simply be /usr/lib/jvm/java-6-sun/.
CLASSPATH on other hand will differ from application to application. Interestingly if you find yourself using certain jar files very often, you should place it in standard CLASSPATH which should be $JAVA_HOME/jre/lib.