Today I found out that Don Mitchell has a blog. While I’ve only briefly met Don a couple of times, I’ve been aware of his computer graphics work for quite some time, so it was good to see his writing on other interesting topics. I wish my blogroll had more crazy smart people writing about whatever they are interested and passionate about.
He’s on my daily blogroll now.
Well, I suppose I should have seen this as coming. Our adopted feral friend Scrappy has been fighting off a skin infection, and we have had him on antibiotics for the last two weeks, and he went back in for testing, where they test for, among other things, FIV antibodies.
Scrappy tested positive.
It makes me sad. He’s a wonderful cat.
As I’m sitting here with my iPhone, eagerly (and as yet vainly) waiting for it to be reactivated via the iTunes store, I was trying to figure out what the failure mode was that allowed this kind of (by appearances on the iPhone support forum, and news from various bloggers who are in line to buy new 3g iPhone) serious problem to occur.
Here’s the deal: preceding each and every iPhone upgrade, there would appear to be a needed iTunes 7.7 upgrade. This upgrade itself was sizeable (to be honest, didn’t mark down the size). Then, the iPhone 2.0 firmware update had to be downloaded, which was 218Mb additional. These downloads were accomplished without incident. But then you go to activate the phone… and… problems.
Let’s say just consider the firmware update. 218Mbytes let’s say takes… one minute. That’s about what it took for me via comcast. Whatever the rate of new customers getting this update are, they are able to keep up with this load. But they are not able to process the load of (presumably) getting a small amount of information back and forth to your phone to activate it in the same one minute time frame.
In a way, I guess that it isn’t too surprising: after all, web servers which are optimized to serve up static content are very, very fast. Back when I used to run thttpd, it could saturate 10Mb ethernet links running on a 486. But it also suggests that Apple should have realized the assymetry, and throttled downloads to allow their downstream processes to keep pace. It would have avoided a lot of frustration.
Me? My iPhone is still bricked into emergency only mode, and now I’m getting -4 errors from the iTunes store. Patience is a virtue. My phone lives again!
My cat woke me up before 6:00AM this morning to remind me that today was the Apple 3G iPhone release date. Buying a new phone isn’t in the cards right now, but I suspected that there might be the 2.0 firmware update for my old phone. Sure enough, I powered up the laptop and got a notice that there was a new 7.7 iTunes update. Downloaded it, installed it, plugged in my iphone, check for updates, and…
Nuthin. Still 1.1.4.
Hmmph. Shutdown iTunes, restarted it. Checked for updates.
Eureka! 2.0 currently downloading, 218Mb.
I’ll let you know how it goes.
Addendum: All the cycling of “backup, installing, verifying”, seemed to go reasonably smoothly. Now my phone is in a state where it seems to need to talk to iTunes, and the iTunes needs to talk to the iTunes store, but it’s just not connecting. I’m guessing their servers for unlocking are currently completely swamped because of the 3G release. I’m trying to be patient.
Addendum2: Frown. Now my phone is in the mode where it can apparently make emergency calls, but it is requesting to be connected to iTunes. When I do so, I get this:
I suspect this magic error code is just that it’s activation servers are overloaded, but I’m not very happy.
Addendum3: It appears that my surmise was correct. Going to Apple’s support forums:
Confirms that lots of people are having this magic “-9838” issue, and that Apple has confirmed that it was for inadequate bandwidth on their end. Sigh. I’m gonna go take a shower.
Addendum4: Word of the day:
Apparently word is spreading that activations are just horribly slow. My own iPhone has gone from having that -9838 message, to just spinning endlessly, waiting for the iTunes store. As Henry Jones would say, “Our situation has not improved!”
Addendum5: Around noon (after about six hours), my iPhone finally phoned home and I know have a working iPhone again. More later.
My wife noticed a mommy and daddy quail at our fence, with a whole bunch of babies squabbling around in the grass at the bottom. Couldn’t get any good pictures of the babies, but shot this video of the parents:
Not sure the autofocus got the quails as well as they could have, here is the best still picture of them I managed to get, full resolution.
This came across one of my mailing lists: a complete description of the mechanical computer used to solve torpedo targeting problems during WWII. It’s really very cool, and could serve as a great intoroduction to how mechanical computers work. Very neat.
Well, I decided to try again to record some of the beacons that are on 30m. I recorded 1 hour of what sounds like white noise starting on July 5 around 18:30 UTC, and then ran it through an FFT and mapped out the frequency ranges that represent 10.14000 to 10.140200 Mhz. Here’s a processed version of the spectrogram:
Strictly speaking, only the lower half of the image is within the beacon band, but I haven’t actually calibrated my FT-817, so it might be off a few Hertz. The signals aren’t strong enough for me to identify any of them, but I might actually be getting something.
Here is the naked, unprocessed spectrogram.
Most people who do this kind of listening use Argo, written by I2PHD, and indeed I would except for one thing: no boxes run Microsoft Windows in my house. Well, that and I am kind of interested in the kind of DSP algorithms that enable us to do this kind of stuff. So, I tinker this stuff together for the Mac/Linux.
Addendum: I left the recorder going when I went out today. Around 0:00 UTC, I recorded the following near hour of the band between 10.140 and 10.140100 Hz. Whoohoo! My first real beacon. Not sure who it is yet though.
Addendum2: Hmmm. Figured it out! It’s WB3ANQ I think. His grid in FM19rc is almost 2450 miles away, using a power level of only 200mw!
Hello Knights, I have joined in on 30 meters 10.140.095 to 10.140.100 Mhz QPS ref. 200 mW with my SNAKE format /\/\/\/\ 30 second period as of 1900 UTC -- Larry Putman WB3ANQ Pasadena, Maryland FM19rc
Courtesy of Lifehacker, here are some neat pictures of a guy who added an office to his home by using a pre-fab 8×15 shed which was wired, insulated and dry-walled. Very nice. Wonder what the zoning issues surrounding this kind of installation are.
I’ve been bitten by the QRPP/QRSS bug, and am considering a project where I build a MEPT (that stands for Manned Experimental Propagation Transmitter, apparently), which is a simple beacon transmitter that many people are using to experiment with on the 30m band. QRSS activity is usually in the range from 10.140 to 10.1401 (yes, only 100 Hz of spectrum space), but above that some other people are using KJ1T’s WSPR (“whisper”) mode, which is an FSK digital mode. I don’t have a decoder for it (KJ1T seems to only have written a Windows version of it, and it’s new enough that a Linux version doesn’t exist) but I can see some of the signals in the recording I did on 30m this evening.
The spectrum goes from about 10.1406 on the left to 10.1410 on the right, with about five rows per second scanning down. You can see several WSPR signals, and the wider signals are PSK31.
I blogged about “Dramatic Chipmunk” one year ago today. My, how time flies.
