Walk Softly and Carry a Big Schtick

Progress on two fronts: recovery of Hazel's Picture Gallery using the Wayback Machine archives, and our entire music collection has been installed on our new Jukebox Server.

I've been slowly, methodically, going through the Wayback Machine (Internet Archive) looking for photos from my server. I've found everything they have of my pre-2000 photos, and I'm putting them back on the website. For example, you can see one of my old pictures of lyzard13 at lasfs (back in 1999) is back on the site.

I had previously been holding off doing this, because I knew I would mostly just get back thumbnails, and I thought I'd get better results sooner if I concentrated on the recovery data from the old drive. But I got bogged down with the matching-up process on the recovered files, because the two big batches I was looking at were from a Grilled Cheese Invitational -- lots of photos of faces I don't recognize -- and from various weeks at lasfs, which would be lots of similar photos of faces I know. Either way, tricky to match back up, so I put the recovery project on hold. And laconiv needed website work, and I've been rewriting my database tools to use MySQL, so I've been keeping busy. The database rewrite means I can track whether I've lost a thumbnail image, a viewing size image, a high-resolution image or all three, which helps in figuring out where I am on the restoration project.

Well. It occurred to me that the Wayback machine project would have three or four benefits to varying degrees. First, I'd get back a bunch of thumbnails, which would mean that for people casually browsing the website it wouldn't look quite so moth-eaten. If you look at these pictures of Maria and her friends you'll see it looks like the page is mostly back in place. The reason it still says "10 out of 10 pictures lost" is because none of the large-size pictures have been recovered yet.

Second, I would get back some of the viewing-size pictures, one or two of them anyway; click on the photo of lyzard13 above right, scroll down and click on her again, and you'll see I have that one back.

And thirdly, because the thumbnails come back complete with filenames, I can use the recovered-from-archive.org thumbnail images to help me place the recovered-from-the-disk-drive images. Like a ton of bricks, this concept hit me; I'd left myself a bookmark to get back into the 1997 photos, and over the past week I've been grabbing them and re-uploading them.

From a "data recovery progress" standpoint, my percentage-recovered indicator has only gone from 32% to 33%. But some of the pages look a lot better.

And, in other news: The Jukebox Server has been built. I've talked about this for years -- almost a decade -- and I realized earlier this year that I'd better get cracking on implementation if Maria was ever going to get any use out of it, instead of her eventually moving out and concluding that my talk about that project was just a bunch of hot air. The project has been underway for several years, as far as ripping MP3s from our 1400-plus compact discs is concerned; all of us have already benefited from it as far as listening to music on our computers and playing MP3 cd-roms in the car. But the grand project -- all of our music loaded onto one computer, for easy random access of all tracks, with eventually a moderately-clever playlist generator built into it -- that's been stalled.

Well. We've got it. It's sitting on our living room table right now. It's a G4 Mac mini with a big external Firewire drive, and every single CD has been encoded to MP3. There are a handful of recent acquisitions sitting on my work computer that haven't been incorporated into the server, but I'll clean those up real soon now. I'm going to go out and get a WiFi card for the server so we can stream music from there to any computer in the house; I'd been thinking of plugging it into our Airport Extreme or Airport Express stations, but it's not ready to be hidden away just yet and I want us all to get to use it in the near term. (A few months ago I promised Maria it would be ready before the 4th of July weekend.)

(The new Intel-processor Mac minis are in the store now. Version 1.0 of a new product. Oh joy, who wants to be part of the gamma* test group? I figured I'd better act quickly if I wanted a G4-processor model, especially if I wanted one new, which is part of why it all happened in a bit of a blur...)

*Gamma testing = the experience that unsuspecting customers have when they buy a product that's been released immediately after the beta test phase. (Gamma is the third letter of the greek alphabet.)

I've learned something interesting about Apple's "iTunes" -- the shuffle-play isn't nearly as random as I expected it to be. Because of how the music has been loaded onto the computer, great big chunks of classical music are located "near" each other, and the "next random track" tends to be not too far away from the one playing. In other words it's been playing just classical music (even though it's only 6% of our music collection) for the last three hours -- and the hour before that was a mix of classical and other instrumental stuff. (I knew something odd was going on when out of four tracks in a row, three of them were different versions of "Also Sprach Zarathustra.") As it happens, I kind of like classical music for a low-key it's-late-at-night atmosphere, but it's not quite the "pick a random track of music" I'd expected to hear!

[Edit: Shutting it down and re-launching it cleared up the problem.]