Random Rambles

Itz been one helluva week really. Yeah, with no work and me lazing around at home (On Vacation:) ) , wondering whether the sky looked this blue on a Monday morning I transcended the normal human thinking and realized the true meaning of "InActive".

Talking of Action, Technocrati has now come up with the list of top sites that are getting Attention - Technocrati Attention Index . Youtube tops the list as usual but was quite surprised to know that Ny times comes second. Weird. But given the current scenario not very surprising either.

Sun steals the show.. and I am back

It isn't a good sign to pop in once a while and check if my blog still exists or blogger.com has decommissioned it citing inactivity. So made up my mind today with some reminder from one of my friends to come and scribble.

Been doing good albeit hectic schedule. 2008 hasn't revolutionized my day to day life yet, so nothings much changed from where i was earlier :).

Came across the latest developer.com awards here :-
Dev.com

Seems Sun products have won in most of the category. Good to see that the Scripting feature for Java has won the best award. Hoping that next year, closures will be in the race as well for this.

Java has undergone a complete facial remake in terms of features and extensions. With Java 5, the emphasis was not only on just API extension but to include in a lot of language extension, developer friendly typing features as well. Generics is one good example. Though many argue that generics is not quite robust and has some fundamental flaws , from a developer perspective, i think it satisfies what it was meant to. Strict typing was a major difference that put Java in top of many dynamic pgmning languages and by making it compile time, genrics helps handle unneccesary surprises.

At any case, the point I was trying to say was Java SE 6 has also won in the developer.com category. And with Java 7 including some better functional programming practices within it, i will place my bet on it.

Closures in Java - What lies beneath ??

I was listening to the "Closures for Java" tech talk by Neil Gafter. The presentation was very impressive and the content was equally good. Yet,one basic question remains unanswered. Do we need closures in java at all ??

I have nothing against closures or it being added to Java. But i like to raise few debatable points:

1) Closures is not something new. According to Wiki:
The concept of closures was developed in the 60’s and was first fully implemented as a language feature in the programming language Scheme

It is an integral part of many functional programming languages. But the recent addition of closures to Java signals the functional + OOP convergence which i think is nice to have , but not necessary. Anything we can do by adding closures can be done without using closures. So there should be some really strong use cases to justify why a developer should choose closure over his routine coding pattern.

2) Java has Anonymous Inner Classes. But how many of the developers end up using this in their day to day job ? Closures, as i understand provides a lot of features which we can't do with the current day Anonymous classes. But the question is why can't we enhance the anonymous inner class itself to provide all the missing functionality ?Why go for a concept which requires its own grammar and semantics ?

3) Java is Turing complete language. I agree, this doesn't justify closures shouldn't be added, however, Java is complete in its own OOP way. Java was not built to be a functional programming language. If closures are required, why cant we just use a JVM compatible FP to provide that feature ?

These were some of the basic questions that came to my mind once i heard about closures. Will post the answers once i find :). In the meanwhile, if there are any closure enlightened souls, please enlighten me as well.

And by the way, Wishing others & myself a very happy new year - 2008 :).

The Problem of Plenty

After a long time, bak to blogosphere. Not much of an update i have to share.
Anyways, decided to give a techie colour to my blog here upon simply coz i dont have anything else to write about.
Here is what i am trying to implement now..
XML we all know is The mode of data Transfer across different systems. But how much of an xml can you transfer. If u look at it, the size of a XML file is usually much higher than the text file that couuld be created with the contents removing the tags.
Hence is there a suitable algorithm to compress Xml. Google revealed that, the usual Huffman, LZW, work well with text files but not with xml.

Let me know if there is a better way of compressing XML.