Sunday the 11th, 8:23pm — the day of my first departure. Having decided that Portland probably wasn’t worth a detour, I booked the Coast Starlight to Seattle, 24 hours by rail, along with another train from Vancouver to Toronto on the 23rd. (More on that later.) My station of departure was the San Jose Diridon. The train arrived a bit early, so I spent about an hour exploring the different cars and waving goodbye to my family. 

After living in a dark, enclosed space for a few years, I’ve decided to stretch my legs and travel around for a while. They say that routine makes your life pass by a lot faster. It certainly felt like the days were turning into months far too quickly at my previous job. Perhaps the secret to staying young forever, then, is simply a matter of taking in new experiences every day. Maybe it’ll be possible to get some of those months back while looking down from the Sears Tower (sorry, Willis), wandering around the streets of Barcelona, or taking in the neon lights of Tokyo. I’m already a few weeks into a train trip around the United States and Canada, and it’s been wonderful waking up to a new skyline every few days. I’m also working on a few apps and games on the side, leeching Wi-Fi from coffee shops and turning libraries into makeshift offices. With any luck, these projects will turn into a viable career at some point in the near future, and then — I will be able to truly call myself a digital nomad!

I’ve been planning a trip around the United States for the past few weeks. I’ve got my bag, my clothes, my documents and even my portable bucket. Now all I need is… an itinerary? Rats!

Since this trip is intended to be a preliminary journey before I head off to more distant lands, I want to take most of it by train. It’ll be nice to see a fairly sizable portion of the country from the window. I might end up regretting it when I’m sitting in a 48-hour Amtrak car that’s delayed in the middle of the boonies, but hey, it’s better than refreshing MetaFilter or Hacker News for hours at a time.

I drew up my initial plan a few hours ago. Seeing as how I didn’t want to meticulously plan every detail of my drip ahead of time, I decided that a flow chart would be the best way to do it. (It’s like a choose your own adventure book in real life!) The best online tool I could find for this purpose was called LucidChart, by virtue of the fact that it featured automatic curved arrows. Free for personal use! Really cool.

Here’s the semi-finished map (click for larger):

(OK, so I know a true sysadmin would be talking about the wonders of rsync in this post, but please bear with me. This was a Windows problem and I wanted to use the right Windows tool for the job.)

I recently got a Synology NAS unit, and I needed to copy a couple of terabytes worth of data from my external NTFS drives over to the new RAID array. What to do? First of all, it might seem obvious to most of you, but I just want to make it clear: Windows file copy was simply not an option. Among other nasty things, my drives contained backed up user directories from a bunch of old Windows installs, meaning that they were riddled with broken permissions, confused symbolic links and NTFS junction points, and paths that were somehow too long for Windows to handle. Windows file copy would have choked about halfway through.

No, this was a job for some serious command line jujitsu. And thankfully, Windows has a built-in tool that was perfect for the job: robocopy. Just like rsync, robocopy has dozens of incomprehensible flags that you need to spend hours studying to fully understand. Fortunately, I’ve whittled them down to an abridged summary in case you find yourself in a similar situation.

I’ve been doing some serious hardcore scanning lately, and now I’d like to enjoy the fruits of my labor on my iPad. What to do? Obviously, reading image by image is a crappy option. A typical e-book format like .epub might be OK, but in its current incarnation it’s really more optimized for text than image-heavy content. So PDF is clearly the way to go. (Well, I suppose you could also use a comic book format like .cbr and .cbz, but there’s no way your average reader would be able to use it.)

And how exactly does one create a PDF? If you’re starting from scratch, you might use something like Adobe Acrobat. But this isn’t really a good choice for lumping a bunch of images together, since it costs $$$ and also applies all sorts of nasty behind-the-scenes transformations that you probably don’t want. Ideal would be a simple tool that lets you combine images, one or two per page, at their native resolution, with no additional compression. And it should require as little fiddling as possible.

Fortunately, there are PDF frameworks out there that can be controlled entirely through code. I’ve done some looking, and it seems that iText for Java is really the gold standard. (I’d prefer to use a scripting language, but setting up IntelliJ and adding the framework is hardly a hassle. The whole thing took me about 20 minutes to install and fully configure, though I admittedly already had Java installed.)

I’ve had a (rarely-updated) development blog at abstractrose.net for about a year now. It runs on a self-hosted Wordpress installation over at NearlyFreeSpeech and pulls images from Amazon Web Services. Thanks to the maturity of the Wordpress codebase, managing it has been a snap. But the stress of learning about security, finding the right plugins, dealing with spam, worrying about SEO, and other fiddly factors is still significant. And even though NearlyFreeSpeech hosts their own Apache, PHP, and MySQL services, it still took a couple of full days to get everything up and running. (Guilty: I’m a web development newbie.)