Home > Daynotes Home > Week of 4 August 2008

Photograph of Robert Bruce Thompson
Daynotes Journal

Week of 4 August 2008

Latest Update: Saturday, 9 August 2008 14:18 -0500

Paying for this Place
Visit Barbara's Journal Page

Monday, 4 August 2008
[Daynotes Forums]    [Last Week]   [Mon]  [Tue]  [Wed]  [Thu]  [Fri]  [Sat]  [Sun]   [Next Week]    [HardwareGuys Forums]

10:07 - Brian Jepson, my editor at O'Reilly, was kind enough to offer the loan of his first-generation Mac Mini to let me see what OS X is like. He also sent along a USB Wi-Fi dongle, because the 100BaseT socket in the Mac Mini no longer works.

My den Linux system had started overheating, probably because it hasn't been cleaned out in more than two years. It's a BTX system, which means you have to pretty much tear the whole thing apart to get to the CPU cooler. There's probably enough dog hair in that system to build a new dog. At any rate, I decided this was a good opportunity to take a look at OS X, so I installed the Mac Mini in the den on Saturday.

This Mac Mini has only two USB ports. I needed one for the Wi-Fi dongle, one for the mouse, one for the keyboard, and one for an external USB hard drive or USB flash drive, so I plugged a Belkin 4-port USB hub into one of the ports. I figured it might be a good idea to have the Wi-Fi dongle connected directly to the system unit, so I kept a free port on the system unit for it.

The system fired right up, and immediately recognized the Viewsonic VG-2021M display and set the resolution correctly. Unfortunately, that was about the end of the good news. I know people say that OS X Just Works, but I have reason to doubt that. The first problem was that it didn't recognize the keyboard, a bog-standard PC keyboard. It asked me to press the key immediately to the right of the left Shft key and then the key immediately to the left of the right Shft key. Okay, I did that, and it recognized the keyboard. That did seem a bit gratuitous, though.

The bog-standard Logitech mouse was running at a slow setting with no acceleration. I found the dialog to fix that problem and got the speed and acceleration set reasonably. It wasn't difficult, but this is something that Linux takes care of automagically, and which OS X doesn't.

Next step was to plug in the Wi-Fi dongle. The good news is that OS X recognized it immediately and brought up a Wireless configuration dialog. Unfortunately, there were no APs listed, although my D-Link AP in my office is less than 10 meters away. So I clicked on the Rescan button, and it popped up my AP in the list. So far, so good, although it seemed odd that it didn't find the AP without requiring a rescan.

I entered the SSID and WPA passphrase in the appropriate boxes and clicked "Connect". The AP disappeared from the list, and the status button showed Disconnected. Hmmm. So I tried again. When I went to the site survey tab, no APs were listed. I clicked Rescan again. Nothing. Again. Nothing. Finally, after about five tries, it showed my AP again, but the status showed as Disconnected.

So I reentered my SSID and WPA passphrase, at which point an error message popped up. I read through the error message details, and it said things about a kernel error and a bunch of other stuff. The only way to get out of that dialog was to click the button that would send the error report to Apple. Some chance with the system not yet connected to the AP, huh? But I clicked it anyway and after a couple of seconds a message popped up to tell me that that error report had been received successfully by Apple. Huh?

I clicked the Safari icon to fire up the web browser, and it brought up the home page almost instantly. Figuring that might have been cached, I entered a couple of other web sites that would be unlikely to be cached. They came right up as well, even though the Wireless utility was showing the system as not connected to the AP. So I went into my office and fired up the web-based admin utility for the D-Link WAP. Sure enough, the Mac was connected, at 54 Mbps with excellent signal quality. Hmmm.

So I played with Safari for about five minutes before I decided I just couldn't stand to use it as my main browser. I downloaded and installed Firefox 3 with no problems. I installed AdBlock Plus, the Filterset.g updater, and Flashblock, again with no problems. I logged onto my web-based email and checked mail with no problems. Then the connectivity died in the middle of loading a web page.

The OS X wireless utility was, as usual, showing me as disconnected, but the D-Link status page showed me as still connected with high signal quality. A few minutes later, I tried Firefox again, and it woked. Connectivity returned on its own. I did nothing. So I wrote that problem off to some temporary glitch and decided to configure OS X to use the network laser printer in my office. It couldn't find it, although none of our Linux systems had the slightest problem finding it. Thinking maybe connectivity had died again, I fired up Firefox again, which worked normally. OS X was connected to the network. It just couldn't find the IP laser printer.

Then connectivity died again and stayed dead. Well, not completely dead. Every once in a great while, it'd work for long enough to load one or two web pages, but then it would stop working. The good news is that when I fired up the OS X wireless utility, it now showed the link as connected. The signal quality fluctuates, but is always in the 96% to 99% range. The signal strength varies from about 50% to 75%, usually at the upper end of that range. But the stats screen shows a bunch of packet CRC errors.

Obviously, something is screwed up, but I have no idea what the problem is or how to fix it. The D-Link wireless router is configured correctly. I suppose it's possible that the USB Wi-Fi dongle Brian sent me is defective, but I have no real good way to test that.

Watching all this going on, Barbara said, "If you need it to do your work, why don't you just order a new Mac Mini?" And I may do that. But before I do, I'm going to use this one to play around with iMovie or whatever they call the video-editing application.

This system doesn't need Internet connectivity to do what I need to do. Well, it would need connectivity if I were going to use it permanently as my den system, but for video editing all I need is a way to transfer .DV files and finished .MPEG files to and from it. I was about to plug one of my external hard drives into the Mac and suck down the .DV files when I remembered that the Mac probably couldn't read, let alone write, the formats on any of my external drives, which are EXT3 and other Linux formats.

I decided to transfer files back and forth with USB flash drives instead. Unfortunately, I don't have any high-capacity USB flash drives. I haven't bought any since 512 MB was large flash drive and 1 GB was huge. I wanted something in the 4 GB or higher range, because some of the .DV files I need to transfer are 2+ GB.

I checked the Costco web site and was surprised that they had SanDisc Cruzer Micro flash drives at $33 for the 4 GB model and $40 for the 8 GB model. Then I checked NewEgg, and found they were even less expensive there. I ended up ordering two Kingston DataTraveler 100 4GB Flash Drives for $14.39 each, including free shipping.

In the meantime, I copied the raw .DV files to DVD and sneakernetted them into the den to load on the Mac Mini. The first video is made up of six .DV files that total about 5.9 GB. I copied the first five of those to a DVD and took that disc into the den to start it copying up to the Mac. The progress indicator on the Mac said that it'd take about 40 minutes to copy the files, which seemed a bit much. While that was going on, I returned to my office to copy the final 2.3 GB file to another DVD. When I returned to the den with the second disc, the progress indicator said there were 38 minutes left to complete the copy. Ugh.

So I decided to fire up Kino, a Linux DV editing app, on my main system. I opened a new project, dragged the six DV files for the first video over to the timeline pane, and started editing. Once I realized how things worked, it wasn't difficult at all. I thought I'd be literally editing the raw files. Instead, what Kino does is build a template file that includes the beginning time and ending time (or frame number) for each clip. The original files remain unchanged.

You can save in one of two ways. For original DV files from which I wanted to extract only one continuous clip, I used Overwrite mode. For example, the 001 file was Mary's talking-head intro to the segment. We did two takes, each of about 15 seconds, but with extraneous discussion and so on the DV file totaled about one minute. The second take was the better one, and it ran from something like 0:40.504 to 0:54.752. I set the start marker at 0:40.504 and the end marker at 0:54.752 and told Kino to save in Overwrite mode, leaving me with only the 14.248 second clip that I wanted to use.

For DV files that included multiple clips I wanted to use, I used Insert mode. After picking the start and end markers, I'd save in Insert mode, which prompts for "Before" or "After". Choosing Before puts a copy of the selected clip into the timeline before the main DV file I was working on. I could then edit that main DV file again to mark the second clip, save that clip in insert mode, go on to choose the third clip, and so on. I ended up with 14 total clips from the six original DV files.

Once I finished the timeline, I just clicked Export and told it to export as MPEG. The results were surprisingly good, although I somehow managed to get two clips swapped in the timeline and one clip ran a fraction of a second too long, leaving a partial spoken word at the end. The 720X480 MPEG is about 300 MB, which isn't a problem. YouTube takes videos as large as 1,024 MB. Unfortunately, the video is too long. The YouTube limit is 10 minutes--not for any technical reason but because they decided that most videos longer than 10 minutes were from TV programs and movies--and my video currently runs a few seconds over 11 minutes. I need to cut that down to about 9:00 or 9:15 to allow room for O'Reilly to stick in intro video and closing video with URLs and so on. Either that, or break it into two 5+ minute segments.

I'm sure the next video will be easier to do, although probably just as time-consuming. The software and my system are fast enough. The problem is that I have to play DV files over and over to decide what I want to use, where to set start and end points, and so on. The next time I shoot a video, I'm going to keep a written timeline, with numbered takes, the start- and end-time for each take, and so on. That should eliminate much of the drudgery of hacking up the segments to get what I need.


Tuesday, 5 August 2008
[Daynotes Forums]    [Last Week]   [Mon]  [Tue]  [Wed]  [Thu]  [Fri]  [Sat]  [Sun]   [Next Week]    [HardwareGuys Forums]

08:52 - Barbara leaves tomorrow for a bus tour to Chicago with her parents and sister, so until Sunday it'll be just me, the dogs, and any wild women I can find.

We'll probably spend as much time as possible indoors. The high temperature today is to be near 100 F (38 C), with high humidity. Malcolm turns 9 next month. Duncan turns 14 on New Years Day, which is pretty old for a 75-pound Border Collie. Both of them start panting as soon as we get out the door, and I'm afraid to have them out for very long. Come to think of it, I'm 55, and I'm afraid to have me out for very long.

I'll probably do a Firefly marathon while Barbara is gone. Barbara didn't mind Firefly in small doses, but she doesn't want to watch it again. When I really like a series, I watch it over and over, getting more out of it each time.

Speaking of good series and wild women, I've been watching Californication. Barbara watched half of the first episode and gave up. She's no prude, but she found the constant foul language, bouncing boobs, and simulated sex offensive. Although the series is certainly rude and crude, it's also top-notch television and, at times, laugh-out-loud funny. The writing is first-rate, and there's not a weak member in the cast. Quality-wise, I'd put it on a par with, say, Veronica Mars.


Wednesday, 6 August 2008
[Daynotes Forums]    [Last Week]   [Mon]  [Tue]  [Wed]  [Thu]  [Fri]  [Sat]  [Sun]   [Next Week]    [HardwareGuys Forums]

07:58 - Barbara is safely off. The tour bus makes a 7:00 a.m. stop in Winston-Salem to pick up people who'd rather not drive the 30 miles or so from Winston-Salem to Randleman, where the tour bus company is headquartered.

Unfortunately, the pick-up point is a K-Mart parking lot on the other side of town. Barbara would have had to leave here by about 6:00 to meet the bus. She'd also have had to leave her truck in that parking lot. Mary Chervenak had agreed to give me a ride over to pick up Barbara's truck and then to deliver it back on Sunday when Barbara returns (there would be five people in her truck--Barbara, her parents, her sister, and her sister's friend Marcie--and Barbara's truck seats only five, so I'd need to ride back with Mary in any event.)

Fortunately, the bus makes a scheduled breakfast stop at a cafeteria that's only a few miles from our house, so Barbara made arrangements to meet the bus at the breakfast stop. That means she needed only to pick up Marcie, who lives on our side of town. I rode over with them and brought Barbara's truck back. Sunday, I'll pick up Barbara and Marcie at the K-Mart parking lot.

I'm working on the forensic drug testing chapter right now, specifically a lab session drug-testing currency. I'll do that in two parts, one merely illustrative, but the other as real as can be.

Because of the type of paper it uses and the intaglio printing, U.S. paper currency picks up drug traces very easily. Most US notes have detectable traces of controlled substances, particularly cocaine. The older the bill and the higher its denomination, the more likely that traces of drugs will be present.

Unfortunately, tests that are practical in a home lab may not be good enough to detect drugs on a single bill, or even the combined residue from several bills. I'm using a modified Scott Test, which has a threshold limit of about 65 micrograms for detecting cocaine. Any particular US note may have cocaine traces from less than one microgram--a level that is undetectable outside a formal forensics lab--to more than 1,000 micrograms (1 milligram). That means the Scott Test may or may not work for readers. So, just to make sure they can see what a positive test looks like, I'm going to have them contaminate a dollar bill with the contents of an acetaminophen capsule, and then test the dollar bill with ferric chloride reagent.


Thursday, 7 August 2008
[Daynotes Forums]    [Last Week]   [Mon]  [Tue]  [Wed]  [Thu]  [Fri]  [Sat]  [Sun]   [Next Week]    [HardwareGuys Forums]

09:00 - We've all survived the first 24 hours without Barbara, although no wild women have shown up yet.

Duncan has gotten very bad about taking his pills. It used to be that both dogs would sit in the kitchen waiting for their pills. Barbara wrapped the pills in a small piece of lunch meat, and the dogs willingly took them. A couple of years ago, Duncan started balking at taking his pills, as do all of our Border Collies as they get older. I think as they get older they decide that Barbara is trying to poison them.

In would go the pill, wrapped in lunch meat. Duncan would eat the lunchmeat and spit out the pill, sometimes after a delay of up to 30 seconds. A year or so ago, Duncan stopped accepting lunch meat, even if there was no pill embedded. Barbara tried every workaround we could think of, including buying special pill pockets, soft dog treats with a hole in the middle. She'd insert the pill, squash the pill pocket, and Duncan would eat it. Then he figured out that there were pills concealed in the pill pockets, so he started disassembling them in his mouth, eating the pill pocket, and spitting out the pill. Then he stopped accepting pill pockets entirely.

Duncan is now very suspicious of any food we offer him. If it's flat, like a potato chip, he snarfs it up. But if it's lumpy, such as a piece of meat, he carefully examines it for adulterants before eating it. Once he decides it may be okay, he chews it carefully to make sure he's not accidentally swallowing a pill.

Lately, Barbara has had to start giving him the pills by force, shoving them into his mouth, holding his snout closed, and stroking his throat until he finally swallows the pill. Malcolm, conversely, happily eats plain pills. We can just hand him the pill. He eats it and begs for more. If Duncan spits a pill out, Malcolm tries to grab and eat it, and sometimes succeeds before Barbara can stop him.

Barbara has been Duncan's human since he adopted her when he was only six weeks old. He'll let her get away with things that he won't let me get away with. One of those is prying open his mouth to stuff pills in. He may be nearly 14 years old, but he still has a dangerous set of fangs, so I'm very careful giving him his pills. As I'm doing it, he gives me a baleful stare, and I can just tell what he's thinking: "I'm not allowed to bite humans, and I would never bite Barbara, but for you I may make an exception. Go ahead, make my day."

I got kind of sidetracked yesterday. Instead of working on the thing with using Scott Reagent to test dollar bills for cocaine, I ended up working on a lab session on general presumptive drug testing. This one resembles the similar lab in the chemistry book, but instead of using Marquis reagent, nitric acid, and ferric chloride reagent, I'm using Marquis reagent and Mandelin reagent.


Friday, 8 August 2008
[Daynotes Forums]    [Last Week]   [Mon]  [Tue]  [Wed]  [Thu]  [Fri]  [Sat]  [Sun]   [Next Week]    [HardwareGuys Forums]

07:47 - Early start this morning. Someone was walking a dog at 6:35. The dog barked, and that set Malcolm off. I took our guys out and picked up the newspaper. When we came back in, both dogs immediately headed back to the bedroom to go back to sleep, leaving me wide awake.

There was an interesting article in the newspaper yesterday. Apparently, the Hispanic population of Winston-Salem is declining, as many illegal immigrants are leaving Winston-Salem to return to Mexico. The slowdown in the economy, particularly the precipitous decline in new housing starts, is affecting them badly. There's no work here for them, so they're leaving.

Let's hope this trend continues until they're all gone. We have more than enough unskilled workers of our own without importing unskilled foreign workers to take jobs away from US citizens. And, yes, I know all of the bogus arguments about illegal immigrants taking only jobs that US citizens aren't willing to do. To the extent that's true, it's because those jobs don't pay enough to make them worth doing. If those illegal immigrants are gone, those employers will have no choice but to pay a living wage. Or to mechanize jobs that have historically been done by low-wage, low-skill people.

If that means the price of agricultural commodities rises, that's fine with me. What few people consider is that in exchange for cheap vegetables harvested by illegal immigrants, we pay through the nose in indirect costs that are directly attributable to the presence of those millions of illegal immigrants. In effect, by allowing those illegal immigrants to live and work here, we're subsidizing the costs of the Mexican social welfare system, in addition to greatly increasing the cost of our our own social welfare system. The net benefit is to Mexico and Mexicans, with us carrying the costs in the form of higher taxes and higher unemployment among low-skill US workers.

Enough is enough.

Which brings up another interesting phenomenon that apparently no one considered. As the cost of energy continues to climb, the cost of transporting goods increases. Already, some goods produced on the Pacific Rim are being priced out of the market, as locally-produced goods become more price competitive because it costs much less to get them to local markets. For now, the effect is most pronounced for high-bulk, low-value goods such as bulk commodities, but I expect this trend to continue and expand.

It now costs enough to transport a refrigerator or a sofa from a Pacific Rim factory to the US that many companies must be considering moving production of such items back to the US. In effect, high energy costs have begun functioning as the equivalent of significant import tariffs for such items. As energy costs continue to increase, it wouldn't surprise me to see some of the shuttered furniture factories in North Carolina brought back into production.

The sooner that happens, the better. I'm so tired of the flood of cheap, shoddy Chinese goods that Wal*Mart flogs by the megaton. I want the stuff I buy to be made in the US, Canada, or another first-world country. When Barbara and I got married, nearly 25 years ago, the underwear, t-shirts, jeans, and boots she bought for me were all made here in the US, much of it right here in North Carolina. Nowadays, most of that stuff is made in China, Mexico, Romania, or wherever. Places I don't care about and don't want my money going.

So, to the extent that they encourage the migration of production back to the US, high energy prices are not entirely a bad thing.


Saturday, 9 August 2008
[Daynotes Forums]    [Last Week]   [Mon]  [Tue]  [Wed]  [Thu]  [Fri]  [Sat]  [Sun]   [Next Week]    [HardwareGuys Forums]

08:18 - Barbara returns tomorrow evening. The dogs and I will do our happy dance. They're not happy with Barbara gone.

This morning is pretty typical. I got up about 7:00 and took them out for a few minutes. When we came back inside, I sat down at my computer and they went back to the bedroom for a nap. Duncan hasn't been eating much, and Malcolm is just moping around. Last night, I tried to play some ball with Malcolm. He chased it down the hall a few times and then just took the ball and left. Ordinarily, he'd be willing to chase the ball as many times as I was willing to throw it, all evening if necessary.

We'll all be happy when Barbara returns and things get back to normal.

My two Kingston 4 GB flash drives showed up from NewEgg yesterday. Apparently, DHL and the USPS have some sort of arrangement. NewEgg shipped the drives from the west coast by DHL, which delivered them to a USPS center here on the east coast. USPS then actually delivered the drives.

All of the other flash drives I've had have little plastic caps to cover the USB plug. These drives have no cap. Instead, there's a little knurled slider that moves the USB plug in and out of the drive body. It's a better arrangement in that there's no little cap to lose, but I wonder about the USB plug accumulating pocket lint. We'll see.

I'm deep into Chapter 12, forensic drug testing. Here's the current chapter line-up.

Laboratory 12.1: Presumptive Drug Tests
Laboratory 12.2: Detect Cocaine and Methamphetamine on Paper Currency
Laboratory 12.3: Analysis of Drugs by Chromatography
Laboratory 12.4: Observing Microcrystalline Structures
Laboratory 12.5: Assay Vitamin C in Urine
Laboratory 12.6: Analyze Ethanol Content in Exhaled Air

I've finished designing and writing the first two lab sessions, and I'm well into the third. I hope to have it and perhaps the fourth completed by the time Barbara returns tomorrow.

14:18 - I never thought I'd say this, but Paris Hilton may just be playing dumb. She apparently took umbrage at this McCain campaign commercial that used footage of her.

And here's her response.


Sunday, 10 August 2008
[Daynotes Forums]    [Last Week]   [Mon]  [Tue]  [Wed]  [Thu]  [Fri]  [Sat]  [Sun]   [Next Week]    [HardwareGuys Forums]

00:00 -


Copyright 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008 by Robert Bruce Thompson. All Rights Reserved.