Changes Ahead

For the past few weeks, I've had a few ideas brewing in my head. The change from .NET 1.1 to 2.0, as far as my site is concerned, is not only going to have new backend code, but a new design as well. I want to put forth a complete site with working search, comments, archives, etc. In the past, I've usually only implemented one or two features and told myself I'd get around to the others in the future. As we can obviously see from this site, I have yet to add them after 2 years of turning this site into my (somewhat) blog.

If course, what are features if you don't have a decent way to present them. A certain feature I want to implement requires animation. It has been a long time since I did anything with my animation library (if you can call it that), and after looking over the web, I've found quite a few nice libraries that enable me to quickly implement these ideas. Using someone else's code is certainly a change for me; I like to reinvent the wheel, just to prove to myself that I can do it. I've taken a liking to Script.aculo.us, in more ways that one.

After looking over some of the code, I noticed the authors use a different method of passing information to methods: using a single object as the method's argument. Granted, this is nothing new, but it provides an easy to read method of passing information to a method; and one I think I will begin to use.

For example, the new version of XWebMenu focuses more on XML because more people want to implement a menu using XML (in the current version of the widget, XML was added on after the script was written). However, I refuse to leave out the API. So to create an XWebMenu object, you'd only pass one object to the creation method:

var myXWebMenu = xwebMenu.createMenu(
{
name : "myMenu",
attachTo : document.getElementById("divMenuContainer"),
behavior : xwebMenu.behavior.standard
}
);

Of course, this approach requires a standard set of property names to check for, but you no longer have to worry about the order of arguments, you know exactly what information is being passed to the method, and it could provide for a much easier emulation of overloading.

The next change I foresee is a change in hosting providers. I've been with Hosting4USA for a while now, but they're dragging their feet with moving to .NET 2.0. I can get what I get here, plus a bit more, at GoDaddy.com for a cheaper price (and they already provide .NET 2.0!). Granted, I don't have any .NET 2.0 code finished, but the fact that it's available right now says something to me.

Finally, I plan on changing FooReader.NET; more specifically, I'm removing the client-side version of XParser and replacing it with a server-side component that'll send data in JSON format. Why? I like JSON. Development on the JavaScript version of XParser may end, as I'm fairly pleased with it; it's up in the air at this particular moment. Some people have pointed out several flaws with the code... of which, some are worth looking into to fix; some are not (it's f'ing JavaScript, not a feature filled, bonafide, has to check all namespaces class). The result could push FooReader.NET into a more feature rich application, being able to support multiple users, email, as well as RSS and Atom.

Anyhow, that's what's on my plate right now. I'm pushing to get the new site and admin tools complete so that I can go ahead and make the move to GoDaddy. Past that, who knows?

12/3/2005 12:00:00 AM | Tags: JavaScript, DHTML, Web
© 2008 Jeremy McPeak