I have tried to use Facebook many times, stopped using in between and again went to it. But I was never happy. The problem which you all may know is ….

Every friend is same. What I post will be shared with every one. Which is useless for 90% of my friends. At any time I post things which are relevant only to 10 to 15 % my friends.

I have tried to see if there is any groups concept. I was not able find.

Then I thought if there is some thing to beet FB, then I think some thing which provides this feature. And I am right.

I believe that Google+ is exactly what I have been waiting for. And there will be massive shift of users from Facebook to Google+. It will reach Facebook user levels in a rapid speed, a speed 10 times faster than the Android adoption.

Then I thought, Is it possible that Facebook introduces this features sooner and save some users? I don’t think that is possible. People are happy with the thought of moving on. They need some thing to move on to. It is in our nature. And I don’t think it is easy for FB to imitate Google+. It is not technical issue, it is philosophy.

That is why I said “This too shell pass” (Facebook).

Eagerly waiting for Google+ account.

RIM has its own App Store called “App World”. And with recent release of App World 2.0 we are seeing better results. App World 1.0 is used to generate total sales around 6000*.  After the release of 2.0 with new features like Credit card billing and carrier billing we estimate that it is making around 10k* sales per day.

Anyway one can see that sales in App World is like 150 times less than what it is in IPhone App Store.

Now instead of strengthening the functionality of App World, RIM choose to invest in some other App Store called “PocketGear”. This is totally insane from the point of view of any BlackBerry Developer.

As a developer who have participated in every store we can tell that App World is producing 100 times more sales than PocketGear for our applications.

Now only RIM can answer why they choose to invest in PocketGear.

Note: These numbers are total sales for all applications. And they are approximate numbers.

But today when I started MacBook in office I found that I am unable to connect to net. Found that I am using IP number of series 169.*.*.* which is not what I expected. It should be in the series of 192.168.*.*

I tried renewing the DHCP lease. But in vain. Same IP is coming again. Finally decided that it is the problem with the DCHP Server.

But there is no way to find the DCHP server that allocated that wrong IP number. I search for 15 min then finally I found the command in a website.

ipconfig getoption en1 server_identifier

BTW this command only works in Mac. Here en1 is the network interface.

Then I found that my DHCP Server is actually down at that point. So that number which I got is allocated by the OS by default. Not the problem of DHCP Server. I started the server and renewed the IP and every thing worked perfectly.

After a day, I have noticed that my RCP application is not working. After spending almost 1 hr I find out that it is the problem with SWT Browser component.

The exception in the lof is as follows

org.eclipse.swt.SWTError: No more handles (java.lang.UnsatisfiedLinkError: no swt-mozilla-gtk-3349 or swt-mozilla-gtk in swt.library.path, java.library.path or the jar file)
at org.eclipse.swt.SWT.error(SWT.java:3589)
at org.eclipse.swt.SWT.error(SWT.java:3481)
at org.eclipse.swt.browser.Mozilla.create(Mozilla.java:323)
at org.eclipse.swt.browser.Browser.<init>(Browser.java:109)

If you also came across such an issue, here is the fix.

http://www.youtube.com/watch?v=8mwKq7_JlS8

Anyway I felt happy when I realized that we can use JQuery’s advanced short code to extract data from various web pages easily. Till now I have written so many programs in various languages to extract data from HTML files. But no program is similar to any other. It is due to the fact that each page has its only layout and the data I need to extract are different every time.Now using jQuery and HTML tool-kits like Cobra we can easily extract data from any HTML file without using regular expressions.

I wanted to develop an library which uses jQuery and Cobra’s API and extract information from HTML file using a standard specification. This specification will address where and what information to be extracted, and the result will be objects of the data. I have this dream from long back. But jQuery may help it get started. Thank you jQuery.

I hated make and nmake all the time I worked on Linux. The syntax of the Makefile was so complex for me to understand it. I have compiled so many projects with make, but still don’t know how to write a file myself.

Thanks to Java, I stopped writing programs in C,C++.

Today I thought of checking what exactly Ant is, after working in Java for more than 5 years. After reading the manual I thought I should have checked it long back.

I don’t know how I managed these many years without using Ant. From now onwords I decided to use Ant for all my Java projects.

Check this for Ant tutorial.

If those many software developers are working under me, I would have replaced all this software we use and replace with artificial intelligent software which will do every task we normally require( OS, Email, Messaging, Collaboration, Security, Office Applications, Databases, Enterprise Management systems etc etc.)

I personally feel that that is the future of computers. There wont be any softwares in that systems that we presently know.

Only one software and it is dynamic. It will evolve its capabilities of doing each task. It learns any unknown task from some one else. U talk to the computer. You will not type. OR you just need to imagine, no need to talk also. Even though such input devices seem to far, I don’t see such softwares as too far.

40,000 could have replaced all known software applications (excluding scientific and mathematical algorithms).

I will do prove it one day…

I have written an article long back about next version of OS. And my new idea of CBI may take me one step near to my dream.
It works like this….
Say for a task of designing a website with no graphics.
If the users tells the computer how to design (similarly how he tells the designer to do it..) he may need around 100 words to convey his thoughts.
If number of meaning full and context based sentences may be around 10,000. Not more than those sentences are possible in the task of website designing.
If every word (in the 100) have a properties, actions, attached to it..we/the computer can conclude the meaning of each sentence in 10,000 words. So in affect we are actualy telling the computer what we want and it is poerforming the task for us.

At present we dont have any such ability. We need to do what we want. And some tasks are so complex to do it for any person who is not a professional desingner. But if common language interface is possible with CBI then surely it will be having more target audience than what all other software do.

This CBI can be applied to any contaxt and the final task becomes very small. Because in each context there wont be more than 100 words to process. And we think we can tell meaning of 100 words to the computer…

We are not trying make it understand what we mean.. We are trying it to do what we mean. (it means computer)
You dont need to understand some thing to do it..because ultimatly it is our programmers only who are doing….
We are just giving more complex commands and arguments…nothing else.

• Dynamically Expandable at any time
• Should be enabled with Network Knowledge(the word I choose to describe the ability to ‘digitize, use, share the knowledge of all people connected to network)
• Programs should be integrated with OS. (you can not launch a program. OS will choose the program depending on the work it has to do for you)
• Programs are dynamic. (New versions get installed automatically. New programs are installed when needed)
• Programs are just made for you. (As they are dynamic…they will most probably made by one who have just same requirement as yours.)
• All programs are interconnected.. (As they will be part of the OS.. all are designed to take advantage of others)
• Programs dont hold any data. only OS handles all data (So that integrity between programs can be preserved)
• OS is totally adaptable, user centric.
• All adaptations are shared and ranked. So are the programs, developers, GUI.
• Dont do things manually. Just tell it. It will do most of things.
• Programs are not big like MS OFFICE. they are mostly small piece of code(<1kb) which understands just one or two words that the user is saying and just knows how to act for that command.
• Never re-invent the wheel (this is most important thing. The present system completely failed at this point)