On Friday, though, I very nearly had a toothpaste disaster of my own.  Well, disaster might be too strong a word, but definitely a fiasco.  I had just started using a new tube of toothpaste two days earlier. and when I used it again on Friday I noticed that I’d somehow gotten some toothpaste on the end of the tube, and wondering how I’d managed to do that.  Upon closer inspection, it became apparent that the end of the tube was not sealed at all.  It had been presssed flat, but when I’d squeezed the tube, the toothpaste started squeezing out the end.  If I’d squeezed more vigorously, or if I squeezed the middle of the tube, I would have gotten toothpaste everywhere.

I use Colgate Total Whitening Gel, with which I’ve been quite happy.  I formerly used Aquafresh, and every time I went to the dentist for a cleaning, he or the hygienist would complain about how much plaque I had and how hard it was.  Finally a few years ago the dentist asked if I was using an anti-plaque toothpaste.  I said I was, but he said to switch to Colgate.  Since switching, I’ve had no further problems with plaque.  I’ve very happy with Colgate, other than this unsealed tub, which presumably is the result of a malfunction on the production line.

Before I went to bed last night, I told iTunes to download a bunch of the iPhone development videos. After the first one had finished downloading, I watched it while others continued downloading.

I just checked on it, and about halfway through the download queue it stopped downloading. It consecutively reported four errors on each one, including three “Unable to check for purchases. An unknown error occurred (1001).” messages and one “There was a problem downloading ‘iPhone Development Introduction / iPhone Tech Talks / Apple Developer Connection’. An unknown error occurred (8003). Please check that the connection to the network is active and try again.”

The network is working fine, as can be seen by my ability to post this blog entry from Safari on the Mac. However, when I asked iTunes to download the videos again, it just reported the same four errors for each file, one at a time, so I had to click “OK” about 32 times, even though I didn’t think it was OK.

I tried logging in to the iTunes store again, but that didn’t fix it, and I had to click through 32 error dialogs again. Next I tried quitting the iTunes application and restarting it. It forgot what videos I’d asked it to download, so I had to select them again, and then it got the same cryptic download errors again. So finally I rebooted the Mac, logged into the iTunes store again, selected the videos, and started downloading them again, and of course got to click through 32 error dialogs yet again.

So, Mac, I thought you said that these kind of problems only happened with PC?

Only after I got home did I discover that Apple overcharged me $16.50. The receipt shows the price of the phone as $299.00 and the sales tax rate as 8.25%, but the amount of sales tax charged was actually $41.17, for a total charge to my credit card of $340.17. The correct amount of sales tax at 8.25% is obviously $24.67, and should have resulted in a total charge of $323.67.

In the store, I didn’t really see what they were charging me. They had me sign for the credit card transaction using a stylus on a handheld point-of-sale terminal, and it might have shown the amount on the LCD, but it was very hard to read, so I (foolishly) assumed that they were competent and honest.

I wonder how many people have fallen victim to this without noticing. Are they actually remitting the full amount of tax they’ve charged to the state of California, or are they pocketing the difference?

Update: As Dan pointed out in a comment, this actually is legal in California.  If a cell phone is sold at a discount, it has to be taxed at full price: State Board of Equalization Publication 120. I think Apple should have indicated on the receipt that the tax was based on a phone price of $499.

I’m not willing to pay today for the privilege of waiting two weeks for the phone to arrive.

Given the feeding frenzy that occurred at the introduction of the original iPhone, I’m surprised that Apple didn’t do a better job of stocking the stores this time around. Perhaps they think that the limited supply will add to the hype and eventually generate more sales. Or perhaps they simply don’t believe that a customer that wants to buy a phone today might choose a different brand. Certainly the AT&T salespeople were quick to recommend other phones that they actually have in stock.

On the way home, I remembered having seen a sign for an urgent care clinic at 500 Calaveras Blvd. in  Milpitas, so I went there.  The sign was still in front of the building.  I parked behind the building, where another sign said the clinic was in suite 100.  There was also an “urgent care” sign right outside suite 100.  I entered and went to the desk to sign in.  The receptionist asked if I had an appointment, which of course I didn’t.  She seemed surprised, and said that occasionally people drop in thinking that is an urgent care clinic, but it isn’t.  It’s a doctor’s office for cardiology and internal medicine.

Geez, what do you suppose might cause people to drop in without appointments expecting urgent care?  Maybe all the signs saying that it is an urgent care clinic?

She said that she thought there was an urgent care clinic somewhere on Main St., but when I got home and did a Google search I didn’t turn up anything.  It appears that the clinic in Cupertino still exists but has moved a few miles from its old location.

And they wonder why so many people go to the emergency room for non-emergency care.

Usually this works pretty well. The employee or cashier puts the badge up against a reader. Today I was behind someone who’d forgotten his badge, and had a temporary badge. He gave the cashier a piece of paper with a long string of numbers or some such, and she keyed it into the POS terminal, and it didn’t take. She asked him his name, and it was some long hard-to-spell name. She tried to enter it using an on-screen keyboard on the touchscreen display several times, then let him try it, and he had trouble with it too.

As I was waiting there, several people abandoned the line to find another one. Eventually I gave up too.

I think if I’d forgotten my badge I’d either pay cash (there’s an ATM in the cafeteria), or go home and get the badge.

When I vacation in Denver, I stay at a friend’s house. It’s very generous of him to put me up, and I’m not at all inclined to complain, but the one thing that could be better is that he doesn’t have a broadband internet connection. He uses a 56K modem, though he does have his Windows machine set up with “Internet Connection Sharing”, so I can take advantage of the dialup link.

I offered to pay for a DSL line for him, but he’s happy with his ISP and doesn’t want to mess with it. I’ll readily agree that “if it ain’t broke, don’t fix it.”

Back when the Metricom Ricochet service was a going concern, I tried that, but it usually didn’t work very well in the places I needed it. Although Ricochet service is available in Denver (from Terabeam), it isn’t available anywhere else I go. And the performance isn’t really any better than a 56K modem anyhow.

On the other hand, “broadband wireless” is now available from cellular carriers. Verizon and Sprint offer EV-DO, and Cingular offers HSDPA, with T-Mobile to follow. I currently use Cingular for my cellular telephone service, but after comparing the service offerings in terms of price, coverage areas, and terms of service, I decided to sign up for the Sprint EV-DO service.

A friend who has done some work in the cellular industry warned me not to get my hopes up about the performance, but as I stated earlier, for my single sample point it works great. I think that may be because Sprint has deployed “Rev A” service here, which is supposed to offer much better latency than the original “Rev 0″ service. It’s necessary to have a “Rev A” compatible modem to take advantage of that.

I chose the Novatel U720 modem, which has a USB interface, rather than the S720 PCMCIA card. My laptop has an ExpressCard slot rather than PCMCIA, so the latter wouldn’t help me. I also want to use the modem with a Shuttle small form factor PC as a router, and USB was more suited for that.

I installed the supplied Windows software to get the service going. It went fairly smoothly. The “Sprint PCS Connection Manager” software downloaded a software update, installed a new PRL in the modem, and updated the modem’s firmware. It worked fine with Windows, but what I really needed was for it to work with Linux.

I’m running Fedora Core 6. I had to manually issue a modprobe command for the usbserial driver, because the driver does not yet know that it should claim vendor 0×1410 product 0×2110. The command is

/sbin/modprobe usbserial vendor=0×1410 product=0×2110

Once that was done, the modem appeared as /dev/ttyUSB0 and /dev/ttyUSB1. I think the latter is a management and diagnostic interface; the former is used for the PPP connection. I’ll submit a patch for usbserial so that future releases will work automatically.
I used the Fedora network administration tool to configure a PPP interface on ttyUSB0. It did not work the first time I tried it, so I had to screw around with it a bit. Eventually I manually changed the “serial” bit rate in the configuration files from 430800 bps to 961600 bps, and it started working. I’m not sure if that was the real problem. The Fedora network administration tool does not offer 961600 as a choice, so I will file a bug report.

Now I can start the ppp connection with the command

/sbin/ifup sprint

Sometimes the first attempt gives an error message “RTNETLINK answers: No such process”. I’m not sure what that’s all about, but giving the command a second time usually seems to work OK.

The V450 mouse is about the same size as the V200, but with more concave sides. The receiver slides into a slot at the bottom rear of the mouse. The laser gives higher resolution. As with the V200, the V450 works fine with Linux and Windows without installing any special drivers, although Windows drivers are provided and give additional capabilities such as low-battery reporting.

Unfortunately, pointing sticks seem to be losing popularity to trackpads, and my latest laptop, an HP Pavilion dv5210us, has a touchpad.  Rather than continue to fight that, I purchased a Logitech V200 Cordless Notebook Mouse.

This is an optical wheel mouse that is slightly smaller than a normal desktop mouse, and uses a very small receiver “stick” that plugs into a USB port. Rather than being rechargeable, it uses a pair of ordinary AA alkaline cells. When not in use, the receiver stick can be attached to the mouse, which will press a switch on the mouse that powers it down.

Overall I am fairly happy with this mouse, but the receiver doesn’t actually tend to stay attached to the mouse when stored in the accessory compartment in my laptop carrying case. This will result in the batteries draining faster, but I can carry around spare batteries. The annoying thing is that I always have to dig around in the bag to find the receiver. Perhaps on a future version of the product they can come up with a way to make the receiver stay attached better; perhaps it could actually insert into a recess in the mouse.

Even when using the mouse, I still have the problem of the touchpad sometimes erroneously thinking that I’ve touched it. I could try to come up with a way to automatically disable the touchpad while the mouse receiver is connected, but this is probably actually forcing me to use better typing hand posture, which is actually a good thing.