The fan on my old Samsung Q1 gave out and I had to find a replacement. After a lot of digging, I had to hit eBay and get something close.

The sticker on the original fan states MCF-4008AM05-S, DC5V 120mA, Toshiba Home Technology Corporation. If you peel that off you’ll see MCF-TS4009M05, DC5V 200mA CODE GDM610000156". Several searches on Yahoo! later I ended up with a result stating that this is close to a Toshiba Portege M200/M205 small fan.

I haven’t measured whether there truly is a difference between the two fans (120mA vs. 200mA), but I did have to change to connector on the fan since that was different.

Time to hook up some stuff to an Arduino Mega, plug it into the Q1 and install that as a CarPC.

The Keyspan ER-RF1 USB remote is designed to work with Macs, but I just plugged it into my Vista machine and it showed up as a standard HID compliant device just fine. The keys, as far as I’ve had time to figure them out, are:

Now all I have to do is get some steering wheel bezels and suitable buttons to hook that up to the CarPC.

If you get the error “Another program has already opened a LocalConnection with id ‘_flexcover’. No coverage data will be recorded.”, do the following:

1. Open Task Manager and kill any instances of CoverageViewer.exe (or your Mac equivalent).
2. Close all browsers.
3. Close FlexBuilder.
4. Close Yahoo! Messenger (or other client that hosts the Flash Player).

#3 did it for me.

I was using FlexCover 0.81.

If you’ve done P90X for some time, you’ve probably noticed that the timings (or counts) for exercises aren’t exactly the same for the right side versus the left side (Yoga knee to forehead, anyone?). This is understandable as Tony Horton usually explains the move/stretch the first time around.

The bad thing about this is that you might be favoring one side more than the other, usually the right. The easy fix for this is to do the movements in the opposite order every other time. This will balance things out for you in the long run.

Tuesday being Plyo-day, I thought I’d drop a short tip: When you’re recovering between moves, make sure you relax, especially when breathing out.

Make a mental note and check next time. I bet you you’re constricting your throat somewhat. You’ll notice it when breathing out. Relax and you’ll recover faster.

Obviously this applies to other exercises, too.

I know I’ve seen the solution to this one somewhere, but if you’re getting the following error on a 64-bit platform, the solution is to set your application to compile as “x86” and not “Any CPU”:

ERROR: 'Microsoft.Jet.OLEDB.4.0' provider is not registered on the local machine.

There have been several people requesting help using the Yahoo! Address Book web services in the .NET Developer Forum. People are getting 403 or 404 error codes back.

I did some digging into the issue this morning and the cause for the 404 error turns out to be an invalid URL for the DOCTYPE declaration for the returned XML data. In short the URL http://l.yimg.com/a/lib/pim/r/abook/xml/2/pheasant.dtd set as the DOCTYPE for the resulting successful web service call XML data is invalid.

This causes problems when using the Yahoo.Authentication class methods, such as GetAuthenticatedDataSet(), resulting in a 404 when the .NET XmlReader attempts to validate the data against the DTD.

The way to work around this issue is to use your own XmlReader with custom XmlReaderSettings specifying to ignore the DTD.

Some mighty trippy default images seem to be in store for Windows 7.

See them all here.

The plug got pulled on OAuth this week in several places, including Twitter and Yahoo!, with Twitter taking a lot of heat on the subject.

Yahoo! is now allowing users and developers to authorize applications, but with some additional messaging letting users know that their data may be at risk if they continue. The main idea here is obviously that developers can still continue their work while the protocol level problem is resolved.

Additional information

Yahoo!’s OAuth Update #2

OAuth acknowledgement of issue

Details on the issue from the advisory

Kent posed the question whether the NetGear GS108T would solve his stuttering problem with HD and the XBOX and whether the SRW2008 I bought required any settings to be modified.

The GS108T might actually do the trick. The determining factor is said to be the size of the buffer memory in the device. The GS108T is listed as having “128kB per unit”. The GS108 I was using before lists “192kB on-chip buffering”.

Now the problem in comparing these two numbers is that the first might mean the device has a total of 1MB (8 ports x 128kB) and the second might mean a total of 192kB (Amazon lists 32kB per port, but the math there doesn’t add up with the specs (8 ports x 32kB = 256kB). I just don’t know if that’s the case, I can’t find solid info on that. But it doesn’t make sense that the cheaper non-smart model would have more memory.

As for the SRW2008, Linksys doesn’t make it easy to find the buffer memory size. I actually found it here:
http://www.smallnetbuilder.com/content/view/29915/51/1/9/

In the table at the end, the SRW2008 is listed as having 1MB of buffer memory. So if that means it’s a total of 1MB, then it has the same amount as the GS108T you are looking at (1024kB / 8 ports = 128kB per port). In short, I don’t have a simple answer to your question. If you do go with the GS108T, please let me know if it works.

One other thing to take into consideration with smart switches is that many of them have fans in them, often noisy. This is simply because these devices are meant for businesses and a living room. This shouldn’t be a problem with the GS108T, it’s listed with “Acoustic noise: 0dB”, so it shouldn’t have any fans.

As for configuring the Linksys SRW2008, I didn’t have to fiddle with the settings. All I did was set an IP and password for the device and left everything else at their defaults.

UPDATE: Kent bought a NetGear GS108T and it solved the stuttering he was seeing. Anyone else have switch models that have resolved the issue?