I spent a few years writing down my thoughts about how one should approach problems. If you're looking for a how-to guide, a cookbook, or a reference this book is not for you. If you want to learn by challenging the way you think, pick up a copy.
They say that old flames rekindle... my second laptop (and 3rd and 4th) were sony VAIOs. Then I went Mac.. (still am).. and now I've started an affair with another VAIO. I hope my Mac doesn't take it to hard.
I bought a new toy today. I know everyone has a laptop already -- as do I. I have a 17" TiBook running 10.4 and I'm more or less happy with it. However....
I do enough development on different unicies that I felt having a laptop that was Intel-based would be worth while. So, I bought a new Sony PCG-6G4L (or a 460S, depending on which label you read). It's nice. 4lbs, 80GB and 1.7GHz. It only has 512MB of RAM and I wanted to upgrade to 1.5GB, but CompUSA just didn't want my money (they didn't stock DDR2 RAM).
My goal? To run Windows XP, Linux 2.6.x (FC4 maybe, or CentOS4), FreeBSD 5.4 amd OpenSolaris or Solaris 10. Lately, I've been responsibly (or at least partly) in making sure that our products package well and run well (out of the box) on a mind boggling number of platforms. We have a very large (20+ machine) build/test/support cluster hosting every platform/OS/architecture we sell for... however, I'm often not near the office and I though it would be nice to be able to test that sort of stuff on my laptop.
Typically, I do core development on a cluster box or on my laptop... Mac OS X's FreeBSD user space is so close to FreeBSD 5 that I can do a decent amount of core development work, data structure profiling, bug fixing, and new innovation on that. However, having a Mac OS X box doesn't help be the slightest bit in building a Sun Solaris 10 32/64 bit build and testing the pkgadd and pkgrm processes.
I haven't had to set up "dual boot" (let along quad-boot) on a box for at least 5 years. I have an eerie feeling I'm going to be spending the first few days reinstalling OSes to get them all to coexist happily.
I've been a long time user of Spread. I would say that I have a decent understanding of how it works and how it runs. I was, after all, in the lab (during graduate school) with all of its creators and even managed to contribute some code.
When I left academia, I did quite a bit of software building and analysis. This led to architectural design of enterprise systems. And a division of my company has a very strong focus on enterprise software development and release management. After learning all that I have about software in the real world, I've had concerns for a long time about the well being of projects, such as Spread, that have academic roots without tremendous input from professional software developers.
Just recently on the Spread list, Spread 4's emminent release was announced and a few features were announced. As expected, these features are almost entirely product function enhancements. Don't get me wrong, several of the features were much anticipated and the others allow for more efficient high-level protocol design. But what is missing?