Yeager

I just finished reading Yeager (Chuck Yeager’s autobiography), having retrieved it out of the local library’s "buck-a-bag" book section along with a few other selections. It’s a pretty decent read, and paints an interesting picture of a man who was in the thick of many of the most exciting aviation developments from the ’40s through the ’60s. He is a blunt, determined individual — heavy on common sense, lighter on formal education. While his interests are not limited to flying aircraft, it is clearly what drives him — his passion for doing so, and the enormous amount of time he spends doing so, define him as well as make him more able to survive the dangers of war and being a test pilot.

It is a drive that I can somewhat relate to — I have been known to get wrapped up in my work and any number of subjects that I find interesting, although probably not nearly to the degree that Yeager does. In a sense, the nomadic life of a career military man can really reinforce those tendencies — moving constantly, and spending most of your time in remote locales with your co-workers and support staff (who also happen to be your buddies) means that you’re talking shop and thinking about "work" all the time. The book periodically drops into anecdotes written by others, including Yeager’s first wife, Glennis, and many of his friends — the ones written by Glennis Yeager vividly illustrate the hardscrabble life of an Air Force family in that era, and her words contrast quite a bit with the passages written by Chuck, a man who is simply thrilled to be moving along from challenge to challenge. "Slick" Goodlin wanted $150,000 to risk his neck breaking the sound barrier — Chuck Yeager only wanted (and got) his normal Air Force pay for doing the same thing, and for many years his family was living a very modest existence in spite of his fame.

The book, apart from the sections relating to the pioneering X-1 supersonic flights, is fairly light on the technical details of flight, but provides a lot of interesting detail about the post-war test flight programs. Particularly interesting is his general contempt for the early fliers of NACA (the predecessor to NASA) — almost without exception he calls them out as lightweights and deficient pilots, reluctant to seek advice from military fliers and especially Yeager, a lightly-educated hillbilly from West Virginia. The accounts of his near-misses are also pretty interesting.

Another aspect of his life (on which considerable space is spent in the book), one which doesn’t match up with my personality, is all of the general hell-raising and debauchery in which he engages. He is pretty much the living embodiment of what you might call a hell-raising rock-and-roll lifestyle (except back in the day) — boozing, chasing skirts, engaging in high-speed pranks and hijinks, and generally bending or breaking every rule that stood in his way. It comes as no surprise to me that, after his wife’s death, he eventually remarried, to a woman 36 years his junior. What is surprising, though, is that apparently he is engaged in a lawsuit against his own children related to the marriage and handling of his financial affairs. Crazy stuff.

Overall, it’s a pretty decent book, and if you’re interested in the subject matter, it’s worth checking out.

Oliver’s Second Birthday

We recently celebrated Oliver’s second “birthday” (when he came to live with us) with a very special treat, one that he was more than happy to devour: smoked salmon.

Oliver eating salmon
Oliver eating salmon

It was impossible to get a good picture of him eating, because his head was firmly planted in that bowl until all of the salmon was eaten.

Mia also got some salmon (in the other bowl in the picture). She only ate a little bit, though, so of course Oliver went and ate the rest of her salmon after he finished with his bowl.

Happy second birthday to my little buddy Oliver!

The Mobile Browsing Experience: Getting Better?

Sandy recently got an iPhone 3G, and I’ve been fairly impressed with the web browsing experience (through Safari) on it. While the JavaScript support is somewhat flaky, the mobile browsing experience on it is close enough to the desktop experience to be acceptable. The responsiveness of the browser is also quite good, at least on pages that don’t go overboard with huge amounts of content or scripting. I figured that I wouldn’t be able to get a similar level of functionality until I got a new phone, since my experience with the existing browsers on my Windows Mobile phone has been less than optimal.

As it turns out, Opera released a beta version of Opera Mobile 9.5 a couple of months ago. I didn’t know about it until recently, and I only got around to downloading it and trying it out today. It is surprisingly good, particularly compared to IE Mobile and the earlier version of Opera that I had tried. They are clearly trying to emulate some of the Safari interface, albeit without a multi-touch interface. Double-tapping zooms in/out of pages, and dragging motions pan around the page. It’s unfortunate, however, that the reliance for the stylus for the preferred interface (particularly zooming) makes it less useful when navigating via the scroll wheel. The responsiveness on simple pages is good (which is essentially the only way I can compare performance against IE Mobile), and more complicated pages also perform well once they are loaded and the layout has been computed.

It appears to be based off of the same rendering technology as the Opera desktop browser (which is a change from earlier Opera Mobile versions). As smartphones and PDAs get more capable (with more memory and faster CPUs), this strategy makes sense — and as an end user, I appreciate the fact that I can be browsing the "real" versions of web pages instead of crippled or non-existent mobile versions. The JavaScript support is decent — most simple things I have tried have worked, although I couldn’t use the fancy WYSIWYG JavaScript editor that the latest WordPress ships with. (I would have been shocked if I had been able to use it, to be honest.)

So far, so good — unless I find something terribly wrong with it, I’m ready to use the new Opera Mobile as a replacement for IE Mobile. What’s even stranger is that, once it goes final, this might become the first Web browser for which I have ever paid money…

(If you want to try it out, you should actually be able to run it on the Windows Mobile emulators. I’m not sure about the actual utility of running a mobile browser on a desktop via emulation, but if you want to see how the other [non-iPhone] half lives, go right ahead…)

Getting Rid of 3 1/2″ Floppies (Somewhat Safely)

A few months ago I wrote about getting rid of old hard drives safely, by destroying the data contained on them. I was going through some more old stuff recently, and came across some relics — 3 1/2″ floppy disks. None of them really contain anything sensitive or that I’d want to keep, but at the same time, I’d sleep better at night knowing that no reasonable person would be able to access any of that data again.

I didn’t find much helpful information through a quick Google search — a somewhat dubious eHow article that advocates stabbing the media through its retractable cover, an article that mentions shredding and degaussing (two options that are not available to me, given that I don’t want to spend any money on this), and a technology columnist who recommends shredding as well as “a large hammer and a stake.” Granted, I could slap each disk in and overwrite it according to security sanitizing standards, but that could take a very long time.

Instead, I decided to investigate simple physical destruction of the magnetic recording material. I was pleased to discover that you can quickly disassemble a disk, separate the recording surface from the hub of the disk, and then shred the recording surface in any ordinary office shredder. Here’s how:

  1. First, lift up the metal shutter on the bottom part of the disk. It may be slightly easier to grab the edge of the shutter if you bend the disk slightly.
    Lift up the shutter
    Lift up the shutter
  2.  Next, you can slide the shutter over slightly, and then rotate the shutter out of its normal track.

    Unseat the shutter by sliding it over slightly, then lifting
    Unseat the shutter by sliding it over slightly, then lifting
  3. Pulling the shutter off may also release the spring mechanism that normally keeps it in place. If not, you can grasp the edge of the spring and pull it out, or simply wait until the floppy has been split open to remove the spring. Discard the shutter and spring.
    The shutter and the spring that holds it in place.
    The shutter and the spring that holds it in place.
  4. You can then pry open the floppy disk using the groove (normally covered by the shutter) indicated by my finger here. You can just use the tip of your finger — you don’t need to use a screwdriver or anything special.
    Pry here
    Pry here
    What it looks like when you begin prying the disk apart
    What it looks like when you begin prying the disk apart
  5. When you have pried the disk open, you will be able to take out the actual media platter. Once you’ve done this, you can throw the rest of the disk into the trash.

    The media platter, removed from the split disk
    The media platter, removed from the split disk
  6. Finally, you can remove the media ring from its metal hub by simply peeling it off. The adhesive should break very easily, leaving you with just the magnetic media itself. You can then feed the media into a shredder, or manually cut it into pieces.

    Peeling the media platter from its hub
    Peeling the media platter from its hub

This process takes about 30 seconds to accomplish in total, which is pretty reasonable for the relatively small volume of disks I need to destroy. Cutting or shredding the media isn’t 100% reliable, as you can still read data from the individual pieces, but it’s certainly good enough for my purposes.

Death Watch: CD Singles

I happened to be browsing through Amazon and eBay the other day, and got to thinking about the CD single format. I used to be quite the avid collector of these back in the late 90’s and earlier this decade — you could often find hidden gems among the b-sides, and I just liked the collecting aspect and the different art. Admittedly, they would often collect dust on my shelf after the first few listens, but my CD rip project has now made it much easier to listen to them at a moment’s notice. (I listened to the early Brave Captain EPs yesterday, for the first time in probably seven-plus years, and discovered that my music tastes have shifted such that I find the electronic-experimentation bits much more palatable now. This is, by the way, the reason why I never get rid of CDs — there have been many artists that I have had second thoughts about…)

Fast forward to this year, and you’ll find that the format is just about extinct. You can’t even find them in music stores anymore, and even the major UK retailers have stopped stocking them. Over there, the format underwent a precipitous 90% decline in unit sales from 1999 to 2007, and I wouldn’t be surprised to see similar figures for the US.

I can’t help but feel a bit sad about this — I consider the CD single an integral part of my education as a popular music fan and collector. (One of the most important lessons learned: the b-sides for singles from a band’s first album are almost always the best they will ever release. Why? Well, by the time they get around to recording an album, they tend to have a large repertoire of songs laying around. For later albums, they don’t have that backlog to rely on.) Another concern is whether or not "official" remixes will die out — part of their raison d’être was to fill out singles.

I feel like the industry had a hand in killing the CD single format — the move to multi-part singles and the reduction in the number of tracks (partly to satisfy chart rules) smacked of greed. While it didn’t stop me from buying them, I’m sure that the consumer market as a whole got tired of it. Three part singles were the absolute nadir of this phenomenon, though thankfully they were short-lived.

My feelings echo many of the commenters in this thread. There’s just something different about clicking on a link and downloading an MP3 — granted, I am coming around to the idea given the convenience of having my music library instantly accessible, but only grudgingly so. Coming soon: a screed regarding the practice of buying just one or two songs from an album instead of the whole thing…

Matthew Sweet at the Coach House

Last night I saw Matthew Sweet play at the Coach House in San Juan Capistrano. He’s on a mini-tour to promote his new album (out since a few days ago), which I picked up but haven’t listened to yet. The show was OK — heavily weighted towards the new album, which I am cautiously optimistic about. The venue, which I’d never been to, is pretty nice but kinda weird — looks like it used to be a large sit-down restaurant in a strip mall.

It sounded like he was winded or short of breath at times, which was not good — what’s worse is that he actually said something along the lines of, "Well, I’m not as tired now as I was at the show a couple of days ago!" While it may be overly dramatic to say that time has not been kind to his body, he has clearly put on some weight (and he was a pretty big guy when I first saw him live 13 years ago), and I have to wonder if that is affecting his performing ability. I should also note that his sartorial choice of a mesh-back hat did not add to my estimation of his health.

There’s also an interesting interview in Performing Songwriter with him, talking a bit about his career, The Thorns, his cover record, his fear of flying, and his new hobby of pottery. (For the record, there was no pottery for sale at this show, so I can’t pass judgment on it.) It’s sort of interesting that he acknowledges the changes that have taken place in the music sales landscape in the last few years, and how they will probably allow him (an artist who’s moderately well known) to succeed modestly on his own terms. At the same time, though, he mentions that one of the reasons he signed with his new label is that they do artist development — this seems strange to me, because after 20-something years of producing music, you’d think that you would already have your market and fanbase pretty well defined. An anecdote in support of this idea: I was probably one of five or six people at the show who were under the age of 30 — there probably aren’t too many new Matthew Sweet fans out there, even if he did have a track in Guitar Hero II…

Also, his long-time drummer Ric Menck looks like he stepped off the set of The Planet of the Apes — with sunglasses and a hat on, he really looked like an ape. Pretty crazy.

Bureaucracy

Recently, I went to the DMV in order to renew my car registration, as well as renew my driver’s license. Now, if this were a very tired comedy shtick, I would proceed to complain about how this experience wasted an entire day of mine and destroyed my soul.

Instead, because I try to give things a fair shake, I’ll report the truth. Namely, that the DMV experience (at least in California) is not a soul-destroying experience at all, and is actually much closer to the ideal DMV experience than comedy hacks would have you believe. In the 8 years that I’ve lived in this state, my DMV visits have typically been pretty efficient and without complication, never having to endure more than a 15 minute waiting time.

Part of that is preparation on my part — the California DMV has a "wait time estimator" for each office, and I consult that to choose a time to go over there, minimizing my waiting time. Even taking that into account, though, I don’t feel like my time is wasted at the DMV. The signs in the office are very clear, and the "take-a-number" system keeps things moving smoothly. The employees keep up a steady pace, and I have never had to get into a long discussion or argument about a form, payment, or other issue. Granted, my transactions there have always been pretty straightforward. Yet, I have dealt with many other organizations (which deal with many fewer transactions) that prove to be far more troublesome, so I don’t feel that my praise is unjustified.

All in all, it’s a pretty streamlined bureaucracy now. So, to all comics out there who are starved of material: making fun of the DMV was low-hanging fruit back in the 1990’s. (And yes, that ridicule is certainly a reason why improvements were made.) But it’s probably time to find a new whipping boy…

Backing Up My Music

After finishing my CD ripping project, one of the first things I considered was, "How should I back this stuff up?" The last thing I want to have happen is for my hard drive to fail, necessitating a repeat of this epic project.

My brother had suggested just using another hard drive or some sort of NAS, but that solution can still suffer from hardware failure (depending on how much money you spend, and I am not looking to spend a lot). I wanted some redundancy, and the ability to have an off-site backup. However, the thought of having to push 45 GB of data upstream to the Internet is unappealing — that, plus the fact that storage service sites would cost about $60/year minimum, made me look elsewhere. This is music I’m interested in backing up — not anything mission critical, and not necessarily something I need remote access to, but at the same time something that I do want to be backed up.

Eventually I decided on simply burning DVD sets, and sending one set to my brother to act as my off-site backup. I looked around a little bit for any sort of application that could help in this regard — however, searching for "DVD backup" on Google just returns a ton of results of products for "backing up your video DVD collection." Sourceforge‘s backup section is a somewhat barren wasteland with no clear indication of tried-and-true solutions that would do exactly what I wanted. The backup utility included in XP doesn’t really deal with the type of backup I wanted (splitting files across removable non-tape media), and I don’t have a tape drive to use with it either. I eventually just rolled up my sleeves and split the collection manually, burning DVDs with the copy of Nero that I got with my DVD drive.

Unfortunately, while Nero claims that it clears the archive bit from files written to a CD, this doesn’t appear to be the case. This will make future "incremental" updates more difficult for me to manage — I’ll have to figure out some kind of strategy for dealing with this. (This may result in me writing some custom software to handle exactly what I want…yeesh!)

This backup strategy will be augmented with file synching to another computer we have, probably using the Sync PowerToy. I haven’t set this up yet, but it looks like it should be fairly straightforward.

Braid

Braid is a very good game that overcomes some ham-fisted writing and overly ambitious themes. I won’t really dwell on talking about the game in this post other than to say that I recommend it, and that there are some mild spoilers in the remainder of this post.

I saw a post by the author commenting on the fact that getting the game to run on the PS3 would be a "gargantuan effort," and that got me thinking about the reasons why, along with how the game engine works to begin with.

First of all, the time-rewinding mechanism appears to be using input recording rather than state recording. That is, the state of the user’s controller is stored for each frame of the game, and is used to control the player’s character during time rewinds/fast forwards. I base this idea off of the fact that, in the world that uses your "time shadow," the shadow’s actions are based on the player’s controller input, and the replayed actions can be affected by the current state of the world.

Naturally, there is additional information stored to assist with the time mechanics of the game — input alone is not enough to reconstruct the state of the world. Jumping, for example, plays a significant role in the game, and being able to "rewind" a jump requires knowledge about when and where the jump started. (If it were possible to efficiently reconstruct the state of the game up to a certain time, you could avoid storing the "where" part, but I don’t think this is the case.) Additionally, the behavior of AI actors like the bunny and cannons would require some state to be saved in order to properly rewind or replay their actions.

The game runs at a steady 60 FPS. If we assume that replay data is stored at the same frequency, and we assume 128 bytes of replay data per frame (a number that I simply pulled out of a hat that said "reasonable assumptions"), then 15 minutes of gameplay results in about 7 MB of uncompressed replay data. This back-of-the-envelope calculation leads me to believe that the reason that Braid would be difficult to port to the PS3 is not related to utilizing more than 256 MB of RAM. (The Xbox 360’s 512 MB of RAM is unified, whereas the PS3 has a hard separation of 256 MB of main memory and 256 MB of GPU memory.) There should be plenty of memory for the replay data on both platforms, given the relative sparseness of the levels.

With that possibility ruled out, what does that leave? I think it’s related to the game engine (and replay system) architecture, not the amount of data being created. My guess is that the updates performed by the game engine are, by design, fully reversible, in order to minimize the amount of replay data that needs to be stored. Furthermore, in order to allow for really fast rewinding/replaying (up to 8x, while remaining at 60 FPS), the game update tasks are parallelized.

The game state that is touched by these parallel steps, however, is large enough such that it will not fit into the 256K memory on each SPE of the Cell processor. Furthermore, the game update phase is not organized in any way that would permit "windows" of the game data to be sent to (and modified on) the SPEs. This is pretty much the worst-case performance scenario for the PS3 and the Cell — the need to parallelize algorithms that touch a large data set in a random-access manner.

I don’t think this is an insoluble problem, but a solution that would work well on the Cell would likely look drastically different from what you would write if you only had to consider the PC and Xbox 360. And when you consider the size of the programming team that created Braid, it’s easy to make the choice between rearchitecting the code and moving on to the next game by scrapping the notion of a PS3 port.

Please note that all of the above is just speculation — I don’t have any idea if this is actually the case, but I like to speculate about game engine design and implementation, since I’ve had quite a bit of experience with it…