Dylan Smith

Yesterday the BBC reported that Rick Astley has been named as a nominee for “Best Act Ever” at thsi year’s MTV Europe Music Awards. The internet has spoken and MTV have listened. So can the “Rickrolling” meme really break through into the mainstream?

Rick is up against U2, Britney Spears, Christina Aguilera, Green Day and Tokio Hotel. Certainly formidible opponents, but I’m fairly confident Rick is going take the award. And what a sweet, sweet thing that will be. Not because I particularly like Rick, but because I can’t wait for Bono’s reaction to being beaten by Rick Astley.

After all, remember how much Bono HATES being #2?

Oh please, oh please…

Hmm, this is interesting. Flash CS4 has just been announced officially, and there is some concrete news of the new animation system that they are introducing. The humble Tween as we know it, is gone!

Flashthusiast has an interesting article on the new motion system, which seems to be a huge improvement. I haven’t been able to check any of this out just yet, but it’s definitely got my attention.

This is pretty incredible. The Independent Electoral Commission of South Africa just announced the 2009 Elections, and users who tried to visit their website using non-IE browsers, or on Mac machines, were greeted with this message:

Our server detected that you are using a browser or operating system (e.g. Netscape, Mozilla Firefox, Google Chrome etc.) which is currently incompatible with our website. The current website is only compatible with Microsoft Internet Explorer V4 (and upward) on the Windows operating system.

The website is under construction in order to rectify this.

We apologise for any inconvenience this may cause.

Are you kidding me? How did the IEC get all the way to 2009 with this kind of insane web policy in place? I despair…

I’m at Google’s Developer Day in London today (shamelessly posting over their free wifi right now). First up was the introductory keynote, in which Tom Stocky provided an overview of Google’s vision for the “open web” and the range of products and open source tools that they are building to support it: Chrome, AppEngine and the “Cloud”, GWT and Android.

12h15: AppEngine
I’ve just finished watching a presentation on AppEngine, which discussed some of the product limitations (script execution times, CPU and storage) and some “beta” costs if one wanted to extend the server resources beyond what is offered for free. At the moment, the consensus is that it will cost $40 to double the free resources, which is quite reasonable.

A lot of the discussion was around Bigtable, which is the database behind AppEngine. Since Bigtable is not a relational database, but rather an object-oriented database, this is quite a big learning curve for most web developers. Some Python code examples were presented showing how to keep track of entity counts so that you don’t have to smash Bigtable whenever you want to count entities (this is not a built-in feature of Bigtable).

It was quite an interesting presentation, and AppEngine certainly seems like a great way for developers to quickly and easily get applications up and running. There was also discussion about more language support (Python is the only available runtime at the moment). Mano was quite tight-lipped about this one, saying only that they will stick to languages that “begin with a vowel or a consonant” . I’m guessing Ruby will be next, maybe followed by PHP.

Next up is a deeper look at the tech behind Google Gears.

13h30: Gears
The Gears presentation was awesome. Aaron Boodman presented, and I believe he is an engineer on the Gears team. For those who don’t know, Gears (formerly known as Google Gears, but renamed to take the focus off Google), is a browser plugin that strives to accelerate the uptake of new, standards-compliant browser features across the web. Historically, browser developers would implement proprietary features and APIs into their browsers, which makes writing cross-platform code impossible and each browser would offer a different set of functionality to the developer.

Gears is a set of JavaScript APIs that expose new functionality in a secure way, and currently has ports for all the major browsers. I suppose in architecture it can be compare to the Flash Player – the same SWF will run in exactly the same manner on all platforms with the Flash Player, and have the same set of functionality available.

Gears started out by providing a means of storing data locally for offline use of web applications, but now provides new APIs to help with geolocation, multiple file upload, handling of large binary data (and the transmission thereof) and the ability to run background threads that won’t clog up the UI. Gears is also the code at work behind Chrome’s “application shortcuts” – the ability to create a desktop shortcut to a web application, that will open in the browser that created the shortcut.

The Gears FileUpload and Blob APIs are currently being used on YouTube to allow users to upload multiple videos, and resume failed uploads by splitting the video data into chunks and providing the ability to resume an upload at the last failed chunk.

The Gears project is quite exciting, and a great way to get common functionality into all the various browser platforms. I’ll be attending a Code Lab later in the day to test drive the Geolocation API.

15h00: OpenSocial
Wow, that was a long presentation. Apache Shindig has Java and PHP reference implementations of OpenSocial to make integrating it into container sites easier. Personally, social networking doesn’t really interest me all that much. OpenSocial is a great idea though.

17h30: Google Web Toolkit
This is just about cool enough to brush up on Java for. I’ve heard whispers about this before – a cross compiler that allows you to write all your application JavaScript in Java. The presentation focussed mainly on the technical case for this approach, and put forward quite a strong argument.

They claim that GWT generates optimised, obfuscated JS that executes faster than anything you could hand-code yourself, because their compiler apparently knows about all the quirks and best practises for each browser and will generate the best possible JavaScript for each platform. Also, they claim to eliminate memory leaks because a guy at Google who worked on the compiler is also a serious guru on JavaScript memory leaks.

They also include a history management API (RSH compliant), and they use “deferred binding”: the compiler generates an obfuscated codebase file for each platform, named .cache.js, and a bootstrap JS file that is used to detect the platform and load the relevant codebase. The *.cache.js files can be permanently cached by the server, because if the code changes, the filename will change as well due to the MD5 naming convention.

Another major benefit is the maturity of IDE’s for the Java language compared with JavaScript. By coding in Java, you get all the features of tools like Eclipse, include proper refactoring, code completion, and debugging. You also have the ability to use existing Java unit testing libraries.

GWT seems seriously impressive, and I’m keen to give it a bash to see how much time and hassle it really saves.

With the release of Firefox 3, the SpreadFirefox people are aiming for a Guiness World Record for the most software downloads in 24 hours. To participate, just head over to http://www.spreadfirefox.com/en-US/worldrecord/ and pledge your support. They will drop you a mail when it is released, and all you have to do is download it.

I heard whispers in the wind that an IE8 Beta is out soon. It’s incredible how the release cycles for IE have been accelerated in response to Firefox’s success. I think it’s fair to say that without Firefox, we would still be using IE6 in 5 years time, so the world owes Mozilla a debt of gratitude. I shall remain cautiously optimistic about IE8, unless the Acid Test thing was a manufactured result.

Wow. That’s out of left field.

It appears Sun Microsystems has announced today that it has acquired MySQL. I’m a huge fan of MySQL, having used the LAMP stack as my primary development environment for some time. I’m not sure what Sun paid, but I’m sure it’s worth it, as MySQL is a widely used and much loved technology. I just hope Sun has good intentions.

Read about it on MySQL’s blog.

Update: I just found out that Sun paid $1 billion. It sounds like a lot, but I think it’s pretty fair actually: MySQL powers some of the largest sites on the web, and millions of smaller ones. I’m also glad Sun got there before Oracle. *shudder*

This is slightly old news, as it was announced before I had started this website, but is nevertheless worthy of repeat:

…I’m delighted to tell you that on Wednesday, December 12, Internet Explorer correctly rendered the Acid2 page in IE8 standards mode. While supporting the features tested in Acid2 is important for many reasons, it is just one of several milestones for the interoperability, standards compliance, and backwards compatibility that we’re committed to for this release.
Source: IEBlog

Like many other web developers, when this piece of news hit, I had to do a serious double-take. I do try my best not to jump on the Microsoft-bashing wagon; I sincerely believe that Microsoft are finally taking the issues of web standards quite seriously, and this IE8 teaser certainly shows that.

Microsoft are a massive company with an extremely vested interest in the web, so it only makes sense that they should have employees who are active within the working groups who help to define and support web standards, and lobby for their actual implemented by browser vendors.

More importantly for me, thought, is that I doubt Microsoft is devoid of web developers who are actually passionate about web standards. Standards make sense, and developers are usually logical creatures who value things that make sense (and therefore save them time). Talented developers at Microsoft who work on (and with) browsers every day of their lives are quite likely to care as passionately about standards as the rest of us, if not more so.

The major stumbling block with IE until now has been the lack of competition (which always helps to drive innovation), and the need for Microsoft to support it’s proprietary browser features. I think Firefox deserves at least some credit for this achievement, as the Mozilla project definitely lit a fire under Microsoft’s collective arse over the last few years.

So, in conclusion, I think Microsoft deserves a supportive “well done”, rather than a grumpy “it’s about time”. Now if only the paranoid IE6 installed base would let it all hang out a bit and actually upgrade to IE7, we might have a hope of I8 being adopted within a year of it’s release.

As I sit here typing the first post on my first-ever blog, I wonder how I managed to make it to 2008 without actually ever having one. Perhaps with the likes of Facebook, I felt I was putting enough User-Generated Content™ out there. OK, so I once signed up with Blogger, but I never got beyond the first post. I intend to improve on that performance here.

I’d like to make it clear that, like Maddox, I hate the word “blog”, and I hope to never have to use it ever again. WordPress just happens to be a fantastic publishing platform for any web content, and therefore was chosen as the engine behind this website.

As a keen technologist and developer, much of the content on this site will revolve around tech, and I hope to put up some code samples and pet projects that I’m working on, as I have benefitted much from the Open Source movement, and feel that it is only fair to share and share alike.

Enjoy!