Email is tough. It’s very, very complicated, which is a big problem, because from the outside it seems so simple: it’s like some headers and then a body or two, right? I try to advise people that Mail::Message is not as crazy as everyone implies, but people really like to use something really, really simple, and it leads to problems – one of which is trying to figure out where the ideal distinction between “just enough” and “too much” is.

I’ve been running Leopard for just over a week, and it’s a really mixed bag. My memory is that Tiger and Panther were only annoying because they failed to fix things I wanted fixed. Leopard has broken quite a bunch of stuff, and I afraid that some of it just isn’t going to get fixed.

For quite a while now, I’ve used SuperDuper! to do backups. It’s basically a nice GUI over a rsync. The common use it to tell it “make drive B into an exact duplicate of drive A.” It does so as quickly as possible, changing only the files that need to be changed. This means that after your initial backup, you can make a new backup very quickly by updating the initial backup.

Since the very, very early days of Sub::Exporter, it was clear that it would be really useful to be able to replace the way that exporting itself occurs. That is: the way that routines are generated and installed, based on configuration, gathered collections, and other user arguments to the generated import method.

I first started using procmail around 1993, I guess, when I got my first access to the Internet through my dad’s alumni account at Lehigh University. The only programming language I knew was C (and things that don’t count, like Logo and BASIC), so procmail was like magic.

Earlier this week, I read the final of the Fletch books. (Actually, I read the final Son of Fletch book. The distinction isn’t that important.) I’m looking forward to reading the Flynn series, which is something of a spin-off.

I liked Damian Conway’s book Perl Best Practices. It had a lot of sound advice that can help a programmer or programming group decide on a set of house rules. For those who aren’t interested in making a lot of decisions, it can even be used as a pre-built set of standards (although a few of its suggestions, generally those involving modules releaed by Damian for the book, are untenable).

Yesterday, I called Double Click, my friendly local repair place, and I said, “Um, what’s the deal with my MacBook? Where is it?” The guy said, “Oh. I just left you a message: it’s here!” I was hoping that would be the case. I left my phone at home, so I got to spend much of the day fantasizing about what kind of awesome voicemails I would have: I’d have won the lottery, my laptop would be in. Heck, maybe just my brother would be visiting. Anything!

Well, I’m in the back seat of Walt’s car, and I’m tired. We’re in central Pennsylvania, cruising east toward the Lehigh Valley. I think the workshop went very well, both as a whole and as far as my participation. I’m definitely looking forward to PPW 2008.

I’m at the Pittsburgh Perl Workshop this weekend, giving a few talks and listening to a few others. I’ll write more about that later. Here’s the coolest thing I think I’ve learned so far.