Oct
28
2004
I’m still very impressed with Ubuntu, and plan on putting it on some old PC hardware I have for a friend that needs “a computer to check my[his] email, and surf the web.” The hardware isn’t quite up to WinXP power, but should run Ubuntu fine (my test machine is a similar one, and had no problems with Ubuntu, 500mhz P3, 256 mb RAM. )
Although the system for a friend could run WinXP, I’m selfishly using him as a guinea pig, to answer the question, “Can a guy that knows little to nothing about *nix realistically run a nice Linux Desktop Distribution as his primary desktop?”
I wouldn’t do this with Fedora or “straight” Debian, but I think Ubuntu might push the usability factor up enough to consider the experiment.
Oct
28
2004
Wow! I don’t want to jump to conclusions, but I may have found my new favorite desktop distro of Linux. A slick desktop, with Debian underneath? I’m not sure what else I could ask for.
Ubuntu
Oct
25
2004
I made the switch from Moveable Type to Wordpress. I tried to upgrade to the 3.x free release of MT, but ran into a few problems. I get the feeling that I could have worked through them, but got frustrated enough to just switch.
Wordpress has been pretty impressive so far. I need to spend some time working on the templates a bit. I’d like to get back to my old site’s look. The interface is great, and the underlying database schema is pretty slick. All in all, I’m impressed.
Oct
21
2004
I’ve started to write bits and pieces of the Python Netbackup toolkit that has been kicking around in my head for a while. I’d like to have a library of tools that make scripting common Netbackup tasks easier with Python than Perl. I’m excited about the project, especially because I’ve met another Netbackup Admin in chiPy that’s interested in working on the project too.
I’m using this as an opportunity to try unittest for the first time. I think I understand why unit tests are important, and the advantages of writing them before you write your code, but it’s slow going. I had a few short pages of notes on how I’d like to break the library down from a high level, but I hadn’t really thought out the actual methods I’d like to build.
So, I guess the first thing unit testing has taught me, is that I need to think out larger projects more than I typically do. The handful of “bigger than a script” programs I’ve written have mostly been a quick sketch on paper, build out the classes based on the sketches, then start filling in methods as needed. If I really want to write test cases before writing the classes, then I need to do more work up front.
I decided to try and do the initial design of the project in Leo. So far, it’s been a real help. It’s nice to outline a class, and then fill it in. I’ve already seen areas where a method seemed like it was in an inappropriate place. I’m not sure I would have seen that if I was coding straight.
Oct
15
2004
The October chiPy meeting was last night, it was a great night. Ed Ream, the author of Leo presented Leo to the group. I’ve been using Leo for about a year now, off and on, as a project management outline tool. I’ve known that you can use it as a literate programming tool, but have never been able to get to that quick, “AH! That’s why it’s a programming tool…” moment. Thankfully, Ed pushed me up over the bump, and I know understand why it would be useful outside the straight outlining ability.
Also, I was one of the raffle winners last night, so I’ll be reviewing “Foundations of Python Networking”. I read a chapter on the train on mod_python. So far, the book is great. There’s a lot of stuff in here that I’ve messed around with before, but had a tough time learning from the standard library docs. I think this book is going to help a lot. There’s a section on multi-tasking, specifically fork v. thread v. asynch that I think is going to be great. I’m starting to get to the point in my code where I need to buckle down and start writing some mult-task stuff. I’ve written some threaded Java for class, but have never been able to get anything more than a thread demo program working in Python. I think I’d prefer to stick with forking, just because it seems more natural to me, and has less underlying gotchas. Regardless, I think that section will really help me out.
I’ll probably post as I read the book, as notes for the final review.
Oct
15
2004
This was posted to the Daily Python page, but I thought it was interesting. When I have time, I’d like to look at the source.
http://www.sharp-ideas.net/archives/000032.html