The problem with software in general

Okay, so earlier last week I was having a lot of trouble with firefox. I am a really lazy bastard, so I did the wrong thing. I didn't try to find the source of the problem with the development team. I deleted and exiled firefox from my system for a week, but now it's looking to be very permanent.

A few issues I have with firefox. First, it's bloated. I have a thing in general against bloated software unless that bloat isn't apparent, or if I am starting off with other packages in the general field. Back when I had Windows, firefox was not. I started off around the beginnings of version 2, and it flew on my Pentium 4 processor, clocked at 2.4GHz. It was much faster than the Internet Explorer, which I switched from. Not a problem, until now.

Well, firefox gradually became more and more slow, and I liked it less and less. First, they added features in firefox 3 that I wasn't crazy about, and I had to add plug-ins to remove something that wasn't in the code before, as well as change a couple of obscure configuration variables! Namely speaking, it was the 'smart search' feature. There are two reasons I don't like it. First, it took up two lines instead of one, which made it harder for me to glance through the list of urls and pick the website I wanted. Second, I already new the /exact/ url I wanted, so the search feature that searched anywhere within the title/URL was stupid to me. If I wanted to search for something, I always googled it. I'll admit, the addition of being able to use `google foo' was accepted, but I'd rather trash that feature than have the other two annoyances tacked on.

Why is it that version 2 of firefox flew, but now firefox seems to require a multi-core? firefox flies on my new laptop with a Dual-Core T4200 @ 2.0Ghz. My philosophy is that software should not need to require more excessive resources in order to function properly, but merely use them if they can and not out of need.

This is the reason I changed from firefox to uzbl.

uzbl is a very minimalist web browser, and its tagline is "Uzbl follows the UNIX philosophy". I find this true. Now, it may not adhere to strict definitions (or maybe it does, and my ignorance is talking), but I find this to be highly true. Its utilization of text streams for near everything makes it much closer to a unix philosophy than almost every web browser out there today will ever hope. It's highly configurable, and I give it a recommendation to anyone who doesn't like bloat. For most, it will be a novelty, but for me, it's productivity.

The nice thing about uzbl is really how its core binary is just a rendering spot for WebkitGTK and it uses other minimal libraries to do very basic things like just get the pages. Its magic lies within text streams. It makes heavy uses of pipes and sockets, and it strings together either scripts to make things work or applications to make things work more easily. For example, the defaul download script had a basic amount of functionality (just save the binary to the home directory). I wanted just a bit more. So, now I tooled it to notify me with the stock notify-send binary from libnotify and save to a different directory. That's all the functionality I need. If I wanted more, I'd have to make my own program, and I might do it so I can save under different directories.

Now, there are some pitfalls. It isn't the glorious end-all browser, yet. It is still in its alpha stages. Its configurability is somewhat limited, as I have found so far. And it inherits the bugs from Webkit, which I can't get the debugging binary to compile, and fairly limits what I can debug on Webkit. Since I always plan to have my laptop with me at home, I can just use firefox on it. It gives me an incentive to make the netfifo script I'm working on, as well, so I can quickly copy urls and text files over the network.

I'll post that up once I'm done. I keep getting sidetracked on working on the script with my Python book. And that gets sidetracked by Reddit ^_^