Dan Herbert's

The blog Toolness reports on a really interesting idea for client-side web languages. The idea is to write a compiler for languages like C or Python that will compile code into low-level bytecode for a Flash Virtual Machine. This would allow you to write C programs that would run on the browser’s Flash plugin.

The proof-of-concept compiler has already been demonstrated to compile and run existing software such as a Nintendo emulator in a browser. The idea is that existing C code can be compiled by this new compiler and then run, apparently fine, in a browser.

What I think makes this really interesting is that this sets the way for some really awesome possibilities for web development. I love Flash. I’m really happy it’s the most popular method of including multimedia on a web site (when JavaScript is not an option). However, I don’t like ActionScript very much and I think it could be much easier to develop in a language I’m more comfortable with. This is why a technology breakthrough like this is so important. ActionScript is a little bit of an odd language (in my opinion), and there are languags that I’d much rather be coding in.

The fact that existing code can be easily compiled means that existing software could be ported to Flash and run in a browser. I think that’s a pretty awesome idea.

One of the first things that came to my mind when I heard about this was the issue of security. It seems like any code run in this manner is still secured by the Virtual Machine in the same way that ActionScript is secured when running in a browser.

---

As a little side note, I just remembered that Microsoft’s Silverlight is capable of doing essentially the same thing with any .NET language (C#, J#, VB.NET, Lisp, Ruby) What I think is so awesome about this is the idea that the much more open and widely used Flash is the technology that is being used here. This means more platforms will fully support the technology once it’s released.

I can’t wait to write Flash stuff in Java (no actual word on Java yet, but I can hope) or Python.

As a web developer, one of the major things I’m interested in knowing is the general market share of browsers that are out there. This is a tough thing to measure since unlike most markets, there is no purchasing going on and no physical object to buy or own like with cars.

The best way to measure browser market share would be by getting a large sample set from a very popular web site with a very wide range of users from the computer literate to the computer illiterate.

W3Schools makes their browser statistics public, however this is very non-representative of the actual market share since it is a site geared towards web developers. It’s no suprise that their site (at the time of writing this) reports firefox with over 40% of the market share, with IE6 & 7 taking up the rest.

A 40% market share is great on a technical web site, but that says nothing for the average user. A better measure would be a site that is generally popular with the web browsing population. A few that come to mind are Facebook, MySpace, and YouTube. These sites have yet to release browser stats that I have been able to find.

Another good metric, which I feel would be the absolute best for reporting market share would be the browser statistics of major search engines such as Google, Yahoo! or Microsoft Live. As far as I know, these companies have not directly released any of their stats about browser usage. Google did at one point years ago when they published their Zeitgeist, but they no longer do.

What I discovered the other day however, is that it looks like Google has released some numbers for browser market share. This is a pretty big deal since it shows what browsers are being used, which is important to know when designing a web site since it’s always best to know what browsers to test your web site on. It seems like everyone has missed this however, thanks to the fact that it was released for a different reason than just “reporting browser usage.” Google released their statistics as part of a study on browser upgrades and what groups of users upgrade their browsers. Although it is very interesting to know how hard it is to get people to upgrade their browsers, I found it more interesting to see Google’s numbers for what browsers people use.

The study doesn’t give extremely detailed numbers as to what versions of each browser are being used, however it does show the use of Firefox 2, IE6, IE7, Opera 9, and Safari 2 & 3. The numbers appear to break down like this:

• 79.4% – Internet Explorer
• 16.3% – Firefox
• 3.4% – Safari
• .8% – Opera

The study did not seem give very specific numbers on older or more obscure browsers, which I would have liked to have seen just out of curiosity, but it at least gives a good idea of the general browser market share of the world. It also shows how bad people are about upgrading their browsers. I think this is really bad since IE6 is considered the most non-secure browser ever released and people haven’t upgraded yet. If you’re reading this and know anyone who might be using IE6, you owe it to yourself to try convincing them to upgrade, if only to make them more secure.

Overall, even though I find it suprising that it looks like the IT community seems to have ignored this as nothing more than a study on users’ upgrade trends, I think it gives a great insight into what browsers the general public use, which means web developers like myself know where to focus our energies. It looks like IE6 is here to stay unless web developers start refusing access from IE6.