Alex Pooley's Blog

Why don't you subscribe to my blog while you're here? I'm a freelance web developer and I blog about Ruby, Rails, and business online.

Go ahead and subscribe to my RSS feed. Thanks for visiting!

Names
Do you find it hard to name stuff? I recently named a component I created ‘Brian’, simply because I had to call it something. Some of the people at IBM seem to have a good naming convention going…

“We named the prototype: Queso, spanish for cheese, because we are
currently naming everything after the Mexican burrito place across the
street in Cambridge, MA. “ - http://torrez.us/archives/2006/07/17/471/

Msgpad
As for msgpad, I’ve decided to cleave the architecture in half. WHAM! Right down the middle. One half is a data server speaking ATOM Publishing Protocol, and the other half is a rich web client using javascript, etc.
For the last three months (and on/off before then) I have been irritated by the MVC architecture that is standard in web frameworks. For web “sites”, MVC is perhaps fine. But for web “applications” it is not good enough. A web application relies on rich user interaction. This will involve substantial javascript. I’m not talking fade in/out and other scriptaculous flare, I’m talking about something like editing an imaging (cropping, painting, etc).

The idea with the ATOM server is that you can create a collection of resources and perform SCRUD (search, create, read, update, delete) operations on them using standard HTTP (puts, posts, deletes, gets). A typical use case would be to HTTP GET an image resource, make modifications in the browser, and then PUT the modifications back on the server. The crux of the idea is that every “resource” (image in this case) has a URL that serves as a handle. So you could have something like http://beagles.com/image/12
which would refer to an image type resource with an ID of 12.
Google Calendar are an example of what I’m shooting for.

The Future

In times like this, I’m happy to live in Australia, far away from the global “action”. Provided we don’t blow ourselves up, fall in to a depression due to peak oil or an over heating China, I think we’re going to see a lot more web applications. The applications are going to share data, and that data will need to be standardised somehow. ATOM is a nicer way of doing this over custom interfaces.

There’s potential for applications to form some sort of infrastructure like the telco companies of the world. Lines of communications are shared and they charge to access each others networks. In the case of the Internet, I can easily see data shared between applications in a similar fashion (with user agreement of course). Perhaps more important
than data, I can see web services shared between companies. Want chat facilities embedded in your application? No worries, just plug in msgpad. Don’t know how to speak ATOM with the server? Not a problem, here’s a line of javascript you can dump in your page, and here’s the CSS styles you can modify. Well, you get the idea…
Incidentally, if you would like to embed a rich chat application in your own software, please contact me :)

technorati tags:atom, msgpad, crud

Blogged with Flock

I just read an article discussing Yahoo’s 22% share plunge. The bit that got my attention had nothing to do with the plunge though. It was a comment from Piper Jaffray & Co and their prediction of Yahoo’s share price by year end.

Piper’s prediction is that Yahoo will hit $36 by the end of the year (down from their initial $42 prediction). This is the same Piper Jaffray that made the prediction at the start of 2006, that Google would hit $600 by year end. Google is actually down about $25 (currently about $400). In fact, it’s even the same analyst at Piper Jaffray making these predictions, Safa Rashtchy!

These “analysts” are entirely over rated. They are no more capable than the cabbage sitting in the frost free vegetable compartment of my fridge. To be fair, I should probably sample a portion of these analysts predictions and try to draw conclusions based on that. But then, I would have to include Mark Stahlman’s prediction that Google would hit $2,000 by end of 2006, and all those other outrages oil, gold, etc predictions that have been circulating this year.
The way it works though is that if the analysts get it wrong, you won’t hear about it. If they get it right, it will make the first page of their glossy promotional brochure in 2007.
There are too many factors out of our control. I say limit your down side, open your up side, and brace for the best. Don’t fool yourself, and don’t listen to charlatans.
Consider this, if Ug the cave man could have predicted the future, then why didn’t he just invent the wheel and get on with it?

technorati tags:trading, life, philosophy, google, yahoo

Blogged with Flock

My blog has been real quiet lately! I’ve had a few half blog posts that I’ve ditched. But my excuse is that I’ve been hard at work :)

Progress is still slower than I always hope for. I’m continually irritated by the state of web application development. I’m doing something about this as I work through msgpad which is part of the reason why it’s taking longer to complete. Still, I know the problems I face everyday are the same problems faced by many thousands of people around the world. It just makes sense to solve these problems properly, once, and for all.

Eric Hodel and Ryan Davis have released a Ruby obfuscater called ZenObfuscate. It takes your Ruby, obfuscates, and then compiles down to native binary (dll, so, and bundle). Anyway, I think this is an important step for Ruby. Coding in Ruby saves me a lot of time and bring me general well being, but I can’t have people rip my intellectual property because it’s all I have right now.

P.s. This post was written using Flock. So far it’s nice, but I haven’t played much.

technorati tags:ruby, flock

Blogged with Flock