After quick googling I landed on http://opencv.willowgarage.com/wiki/ . Downloaded OpenCV from sourceforge.net and ran make. Gave some error that it needs cmake. I downloaded cmake and ran the installer again. It still failed giving a 200 line error that made no sense.

I then did heaps more googling. Someone suggested to do a “sudo port install opencv”. Tried installing through macports. Even that failed as openCV had dependancies on 30 different libraries. Some libraries weren’t compatible with x64 architecture. I tried installing the libraries one by one by forcing macports to find x64 versions, but that failed as well. Some libraries won’t just get along with me.

Then after 3 days of painful searching and googling and trying out every possible method out there, I came upon this website http://vislab.cs.vt.edu/~vislab/wiki/index.php?title=Vision . It had a pre-built openCV mac framework. It had a beautiful installer and it installed like a charm. Moral of story “Terminal Geeks sometimes need to learn on how to make installation process easier for those who aren’t experts  in the domain”.

Then was there was another challenge. How do you get the code to compile. After another day of rumbling and swearing at openCV. I came upon this magic line. “-arch i386″ was the key here.

g++ -O3 <filename>.cpp -arch i386 -framework OpenCV -o <filename>

Also you’ll need to change the C/C++ headers in the following form.

“cv.h” – <OpenCV/OpenCV.h>
“highgui.h” – <OpenCV/highgui.h>
“cvXXXXX.h”-<OpenCV/XXXXX.h> where XXXXX is a library like cvaux, cvcompat, cvcore e.t.c

Hope this blog post can save others time who have similar problems. I will be uploading videos and my experiments shortly.

There is a pretty interesting site called pleaserobme.com that does a very simple twitter search on “left home”, and shows their tweets.  A very handy site for people who want to be rich without working hard.

Now don’t get me wrong here, the Internet is one of the greatest inventions of humanity. Where there is people there is scams and spams.

It happens to be so that whenever we are having lunch at our office and get to a discussion about technology and various API’s. The topic of privacy pops in every now and then. Funnily enough it also happens every now and then that the conclusion that comes is that

“THERE IS NO UNDO ON THE INTERNET, PROCEED WITH CAUTION AND DON’T DO ANY STUPID SHIT”.

We live in interesting times. – (Quote from modius). More on that in the next post.

I have developed a few internal facebook apps before, in my free time just to play around. One of the apps I built periodically checks a website for picture gallery updates, and syncs it with a facebook picture album. Its a very simple app, but got me around to understand the facebook app anatomy.

I have been doing some custom facebook apps for clients (can’t talk much about it), and so far it has been a good learning experience. A few things I have learned about facebook that I didn’t previously know.

According to Facebook Statistics

• More than 400 million active users
• 50% of our active users log on to Facebook in any given day (200 Million)
• More than 35 million users update their status each day .

This is some ridiculously big figures. To put this in perspective, If all the active users of facebook formed a country, it would be the third largest country after india and china.

Now Imagine this, just considering if 200 million people log onto facebook each day, and you somehow got a way to convince only 0.5% of its members to pay up only $1 in an year, you could still be making $2 million in a year. That $1 could be broken up in micropayments of 20 cents. 20 cents doesn’t sound big does it ?

I recently read an article about “How to Spam Facebook like a pro – Insider’s confession” , A very interesting read about how big money will make people cross their limits on ethics. Don’t worry I won’t do any of those. I like programming to solve real world problems, not to create em.

Another article about “Zynga” (yep the company that makes the annoying farmville app). Largest social application dev house, with about 10 million unique visitors a day. Just shows, you don’t need to make a crazy 3D super graphics game. Even with a simple flash game, you can make millions. According to allfacebook stats, its total suite of apps has about 65 million visitors a day, and about 234 million active users. (yes! this blog post is all about millions. lol! )

Another simple example, is a facebook app call “Big Prize Giveaways” which holds some sort of competition, but to take part in the competition you need to be a fan. Guess how many fans they have ? Almost 3 million. With 3 million people as fans, a company could directly reach its customers, get to know where it needs to improve its product line, get a massive list of potential buyers e.t.c

To wrap up I would like to say, I have just begun my journey into what I previously thought was just a fad, but now seems like serious platform for writing apps.

So what do you think about facebook ? Leave your thoughts below in the comments section.

A bit of background, I spend 8 hours a day working with web front ends both in  javascript/html/css and flash/flex. I tend to use appropriate technologies where necessary. I believe to solve a problem ‘You need to use the right tool for the right job’.  HTML and flash aren’t quite substitutes of each other, both have some subsets overlapping but from a broad view you can see why each technology was created. I am a RIA (rich interactive app) developer and not a flash developer, I use whatever tool that best fits my problem.  Gskinner has a much more awesome blog post about it. Below are 5 reasons why HTML5 will still be friends with Flash

1. The language Actionscript 3 – is an implementation of an open standard ECMA-262 set by the Ecma Standards Body. It inherits its language ideas from javascript, java and a bit of C++, its not the world’s perfect language, but for web programming it has several advantages of javascript. AS3 is a proper Object Oriented language with Classes, Interfaces with much syntax borrowed from java. It supports typed objects meaning the compiler can add several optimizations at compile time to make it run really fast on the AVM2 Virtual Machine, which is open source. Having typed objects means I can get better errors when something goes wrong.  Javascript on the other hand is just in time compiled (JIT) or interpreted at runtime by the browser. It can’t afford to spend much time optimising the code that a compiler can. If an error in javascript occurs and its because of a type mis-match you generally spend more time debugging what went wrong.On the developer side, the IDE (Flex Builder in my case) would assist me in finding errors, where I may be passing wrong parameters. In addition Code completion would boost the time it takes to develop something. Now some of you would talk about Google Web Toolkit which works on JAVA. I have tried messing around with it, but it still doesn’t beat the speed of creating a quick working app with MXML and Flex Components.
2. API – When it comes to allowing developers access to creativity, flash has its edge. Some recent camera experiments at work. Doing all the maths but still having decent framerates. Its quite amazing how much of some web features are run by invisible flash movies, because some functionality is impossible to acheive somewhat or not at all on major browsers. e.g Copy Text to Clipboard, Multi-file upload (Gmail uses invisible movie clips to do this), TCP sockets, C compiled libraries on alchemy, Parrallel bitmap manipulation with Pixel Bender, Playing Mp3 sounds on background (yep that’s how you get Google chat/facebook chat pop sounds, Yahoo media player, e.t.c to work). Advanced Text Control. Now again HTML5 has tags, or proposed tags for most of above to work, however getting them to work alike in most browsers is still a massive task. Flash will keep innovating, e.g Peer-to-peer networking with RTMFP, HTML5 has no equivalent of this, may be GPU support and full openGL integration in future. To my knowledge there is O3D and other plugins but HTML5 has no proposal for any such javascript API.
3. Market Penetration - Flash takes about a year to reach about 98% penetration while HTML, errrrmm,  IE6 which is a 10 year old browser, still has a descent slice of the browser market share. In a nut-shell this means for flash to innovate and add new functionality it takes much lesser time than HTML5 to propagate amongst the crowd of browser. When is flash going to be ready ? The year 2022. Still a long long time for flash to play in the playground. Looks like we’ll have to wait for HTML5 (final specs) to run the browsers in iMars spaceships.
4. Same Code, Multiple Runtimes - Getting something to work alike, along multiple operating systems, is not a simple challenge. It makes a great deal of sense in terms of cost and time it takes to deploy something in flash/air. A good number of our clients ask us if they can deploy the same code to work as a desktop app as well. With simple code changes you can bundle the same app that runs on the web and make it run on the desktop with adobe air. New flash 10.1 will make the same code also run on mobile phones. As a content producer it gives me a chance to reach a mass of users, not worrying about what OS or Browser religion they are following. I do agree that even flash 10.1 still doesn’t have great performance compared to native code, and adobe needs to work hard on the mobile so that apps can have the same usable experience as iPhone apps. No one like apps that hog memory and battery. I would suggest adobe to take actionscript to the next generation (more low level control) and add more compiler optimsations. If google chrome’s V8 engine is that fast with javascript, I seriously think actionscript can gain huge performance benefits. Talking about HTML5, I would say adobe has the greatest power to bring mass market penetration. If you remember a while ago, there is was an ie-chrome plugin, which puts chrome’s html engine into internet explorer. Adobe already has webkit in its air runtime, It shouldn’t be so hard for them to bundle a HTML5 engine right into flash. So if a browser doesn’t support HTML5 you can fall back and use flash as the renderer.
5. Creativity & Interactivity – Although google has worked really hard to create the fast javascript runtime with Google Chrome’s V8 engine. The amount of time it takes to create an interactive app in javascript is almost double, I did some physics simulations in flash, doing them in javascript took almost 5 times longer. I just feel the canvas API is hectic to work with compared to the bitmap API. Having great IDEs such as Flash IDE, and Flash Builder get you as a developer some real work done. Aptana studio and Eclipse WRT are good, but still don’t offer the same level of code completion and other smarts. I would agree frameworks like jquery really making adding intereaction to pages quite easy. But looking at how flex solves the transitions, states, skinning, data binding, Its miles ahead of HTML5. This is going to become heaps better with Flex 4.  FXG >> SVG.

On the bottom line I would say, We can never exactly say if adobe decides to open source the flash player. Just as flex sdk went open source. Adobe doesn’t make much profits on flash player, they make $$$$ on tools e.g Dreamweaver, Flash IDE e.t.c It would be good to see flash sort of become a web standard. I understand they can’t open-source the video and audio codecs built into the player, but they should start slowly opening up flash internals where they can, so the community can kick in and help out. Adobe is definitely more open than macromedia, they opened up flash 10 bytecode specs, open sourced avm2, flex sdk, the as3 compiler, removed fees on flash player distribution. In terms of openness google >> adobe >> apple.(my view)

That rounds up my 3rd blog post. Let me know your views, I could be biased, not have the correct facts or info, be too optimistic, however I would be interested in hearing your thoughts. (please don’t mind my grammar, English is not my first language)

Compiled & Compressed Web Languages

Now I’ll borrow a few ideas from how flash movies are represented at bytecode. First the different types of objects are represented into tags, each tag is tries to parse into the object model by reading the appropriate bytes e.g first two bytes represent propertyX , then next two bytes represent property Y.  In this way it doesn’t have to property names. With tags SWF’s are extendible. Finally the whole swf is zlib compressed making it very tiny. Apart from image heavy adverts, flash designers who use vectors and code heavily in their swf’s can produce amazing content at almost minute sizes.

Comparatively even if you try to do the same thing with SVG and javascript, even with gzip compression you still get higher file sizes. Google’s SPDY (pronounced Speedy) proposes to transfer anything that’s text by gzip compressing it. My proposal is rather than just treating HTML, JS, CSS as text, treat them as a separate language. i.e After the browser parses HTML/XML , JS, CSS. The languages represent themselves in an Abstract Syntax Tree (AST), The AST has nodes to hold specific types of data (e.g nodes, attributes, childs). Now if there is a standard for the abstract syntax tree, the Node Types could be just represented as Integers/Shorts, and the the AST would transfer data without its keys, as the browser can get the tree back by looking up the Node Structures from a decided standard. The binary AST could then be further gzip compressed. This would have several advantages. In essence what I am really talking about is how SWF Bytecode is represented. Moreover Action Messaging Format (AMF3) is a binary protocol designed to send large amounts of data back and forth from server to client. There are some very interesting ideas to borrow from AMF. I will see if I can do a proof of concept of how much further compression I can get on HTML, XML,JSON, CSS, JS over a binary compressed AST protocol compared to just plain text. I am expecting to acheive more than 50% (Lets see how that goes). Below are some advantages of using my above proposed format.

1. The browser wouldn’t have to parse HTML,XML,JS,CSS like text, it would rather parse them directly as an AST, This would make rendering a bit quicker, as the browser can eliminate the parsing step.
2. Developers can still have Extended Nodes, which would allow developers to extend the tree with their own structures.
3. You binary format can still be converted back to its original text. It would still miss the extra spaces, comments e.t.c 98% of the web pages a normal user views, how many times do they check out the source. Speed is more important here.
4. Better compression, meaning better speeds.

Fragment Caching

Rather than just gzip compressing a page. It is quite often that a HTML page is just a template with some data changing. If the templates can be cached independently of the data, It would speed up load times as the browser would only need to load new data. Not just templates but sometimes headers/footers/menu/sidebars are same over and over again in different pages of the same websites. Caching them would mean less data sent.

Evolved Javascript

Javascript currently is an untyped functional language. This has its advantages but when developing large applications it gets in the way. There is no support for Classes, Interfaces and typed objects. Using a typed language lets the IDE help the developer write better faster code, It lets the compiler add better optimsations. Mozilla is already working on an ECMAScript 6 implementation. I am really excited to see how it goes. Flash transitioning from ECMAScript 2 (AS2) to ECMAScript 3 (AS3) added alot of features and made it possible to write Flex SDK as we know it. A powerful language is a building block for a powerful Web.

Let me start by telling you a short story about myself. My name is Noj, I have been interested in computers since my Dad bought us a first computer when I was about 15.  My Dad is the CEO of one of africa’s largest plumbing companies Trident Plumbers (yeah I built the site). He wanted me to be in the plumbing or the construction game when I grew up. When I started at University of New South Wales, I was pursuing Civil Engineering for the first semester.

I met an awesome lecturer by the name of Maurice who was taking ENGG1811 – Computing for Engineers class. The course involved a lot of programming  and I realized that while everyone else in my class was going O_O on the whole programming ’stuff’, It was all making sense to me, and I kept on getting full marks in labs and tests. My colleagues would usually come to me for assignment help, and I would show them around my code. (not let them copy though)

After the first semester was over, and one evening I thought, may be I was more passionate about code than I was about buildings or pipes.  Talked it through my family and changed my course to Computer Science. Since then I have never looked back. Software is awesome!

In my uni years I was excited making making cool stuff dance around on the screen. I started learning Flash on my own by reading kirupa.com tutorials. Used to do my wacko experiments, making simple animations, simple games. Met some awesome friends at uni who were into serious learning rather than just hanging around.

At the 3rd year I started working with AELP (Adaptive e-Learning Platfrom) which is a research project at UNSW. There I started my first proper paid job building a google map like tool, but for visualizing high-res cells. (I gotta tell you the brain cells at 100X magnification make up for some good wall papers).

After uni finished, I got a new job at RocketBoots (yeah I built the front-end for that site as well), which is still my current job. I work as part of a great team, doing Flex or Web front ends, sometimes working with coldfusion, java or even some php backends.