HJCotton.net

I had heard that the Tweet-A-Watt could now be hooked up to Google Power Meter, so I decided to give it a go. It’s actually pretty easy – I just followed the instructions here and had it up and running in a few minutes.

I have actually been trying to track down the cause of a recent increase in my electricity usage, and it’s more than likely to be the fault of the dehumidifier in my basement (I have a strong dislike for these things already, necessary though they are). This little guy actually draws a LOT of electricity, but I already knew that. Still, it will be really useful to see the power usage trends over time and find out what it really costs me to run. Given an average cost per kilowatt hour,  Power Meter will give you some useful figures back about the cost to run a device. Based on my latest bill from PP&L (is it worth switching?), my average cost has been 13 cents / kW h since the rate caps expired in PA.

This is the confirmation dialog that appears when you install the Speed Tracer extension in Google Chrome.

“This extension needs access to: All data on your computer and the websites you visit.”

If only spyware installations were so up front about their intentions!

My CPRuby presentation from Thursday, March 18 2010: “Physical Computing with Ruby and Arduino”

RAD-CPRuby03182010.zip

I’ve been eying up the Tweet-A-Watt project for a few weeks now. I think the idea is pretty cool, taking an off-the-shelf power monitoring device (a Kill-A-Watt P3) and tying it to a wireless transmitter in order to get the data from the device to a PC. The tweeting part of the project doesn’t really interest me, but storing power usage data in a database would let me see some long term trends in electricity usage. I might even be able to quell the feeling I have that the “Reduce exterior moisture / Save energy” switch inside of my refrigerator doesn’t do anything.

What I tackled this past weekend was the construction of the wireless reciever (which plugs into my PC via a USB cable) and the transmitter that gets hooked up to the Kill-A-Watt. The entire parts list and assembly instructions are available from ladyada.net. I opted to get the kit out of simplicity. It’s a little pricey, but includes everything you need for the project.

The instructions are pretty good. Provided you are alright with a soldering iron, you can manage just fine – at least up until attaching the power and data wires to four individual pins on the tiny chip found on newer Kill-A-Watts. That part is a little trying. A helping-hands type device is a must have for a project like this and I know that it made things much easier for me. So did a new, super-tiny tip for my soldering iron.

I put the whole thing together according to the instructions and plugged it in. And… nothing happened! I got no display on the LCD and the XBee activity light didn’t blink. The super capacitor takes a while to charge up, but no matter how long I waited, I got no info out of it except for the fact that it would make the standard beeps when the buttons on the front of the unit were pressed. I took it back apart to double check the connections to the tiny chip underneath the LCD display and they seemed fine. I couldn’t see any possible short circuits, but I did read on the forums that it would be important to cut a pathway halfway through the foam on the underside of the display to let the wires out. I also filed a notch in the side of the PCB so that the wires would not be pinched. The LCD has to sit flat in the right location for it to work at all. I double checked the transmitting XBee and discovered that it was not working. I figured I must have fried it as X-CTU couldn’t even communicate with it when plugged into the USB cable. There are instructions about how to revive an XBee that has been put into sleep mode on the forums at ladyada.net. I had to perform the process a few times and it eventually started working again. Upon reassembly, I found that the LCD worked – but there was still no activity light. The LED flashes once when the Kill-A-Watt is plugged in, but that’s it. It works though, so I’m not complaining.

For the moment, the wattcher.py script available from ladyada.net is the extent of the logging I’m doing. By default, it logs to a csv file every five minutes. There is also code for connecting to Twitter and to Google App Engine. Eventually, I’d like to log to MySQL and display the power usage right on my blog. Someday…

Anyway, Tweet-A-Watt is a pretty cool and possibly useful project. It’s on the expensive side if you go the kit route, but no matter how you build it, you end up with parts that can be used on other projects (mainly, the XBee chips and their adapters) if you ultimately decide that this project is not what it should be.

Did you know that the Amazon MP3 Downloader runs on Linux? I didn’t. Generally, when I see a note stating that additional software is required on a website, I head for the hills figuring there’s no chance it will run in Ubuntu. Today, for some reason, I clicked to find out more about the “Requires Amazon MP3 Downloader” button next to the album I was looking at. Much to my surprise, they offer Linux support. Not the “here’s the source, good luck compiling it” kind of support, but rather the “here’s a .deb package, ready to install” kind of support. I was mighty impressed… for about a minute – the 32bit package is not a straight-forward install for 64bit users. Fortunately, it’s pretty easy to make the 32bit package work.

Of course, the fact that extra software is required to download purchased audio tracks doesn’t exactly thrill me, but it sure beats trying to run iTunes in Wine. The fact of the matter is that this is bare-bones, working software that actually runs on Linux:

Whether or not it turns out that tracks you have downloaded can be spontaneously deleted remains to be seen, of course

How much privacy are users of social networking sites willing to give up in order to use the service? Sites like Myspace and Facebook have a veritable wealth of information about the users of its services. Everything from addresses, schools attended and companies worked for to a group of supposed friends and a list of hobbies. People have put up with targeted advertising for quite some time on the web. I know that my Myspace profile is being parsed for keywords because it has suddenly decided I have a dire need to buy custom envelopes and find Bob Dylan memorabilia when I log in because of a Kurt Vonnegut quote and an artist’s name I have in it, respectively. That’s the kind of thing you expect though and I don’t really have a problem with it. I don’t have anything there that I don’t already have on this website – it’s essentially public knowledge at this point. Facebook, for a time, seemed a big improvement over Myspace simply because it wasn’t completely overrun with comment-spamming bots and doesn’t allow users to destroy their pages with horrible styling that makes them unreadable. Granted, being able to customize your page is a big deal. Some years ago, and shortly after discovering how to upload an HTML file to some free webspace and having my own website at a URL that took up nearly the entire address bar in my browser, I learned that just because you can have 60 images, music playing in the background and red text on a blue background on a page doesn’t mean that you should. But that’s another story.

Facebook opened up its API to external developers who can create applications for the users of the site – which is really cool and I see that as being a positive thing. I’m even using one that gives me a Kurt Vonnegut quote at the click of a button – now that’s convenient! It’s also something that never would have come about had Facebook not opened up their system to allow outside developers access. What they have also recently done is roll out a service known as “Beacon”, which has been quite controversial. This service tracks a Facebook users movements on participating sites, of which this article) states there are more than 40 of. Actions such as signing up on one of these sites, adding an item to a wish list or purchasing something directs details of that action back to Facebook, where it is then broadcast to your friends list via their news feed. These sites apparently present users with the ability to opt out of having this information sent back to Facebook, but in the article linked to above, data was passed back to Facebook whether the user was logged into Facebook or not and even when opting out of having it sent on the third party site. Clearly, this is a privacy concern. What’s more is that Facebook was making it impossible to opt out of using the service entirely, instead only giving the option to opt out from individual sites. It wasn’t until today that they announced a universal opt-out feature for Beacon. I made sure to log in and select that all important check mark today and I’m sure many others did as well. Maybe this is the kind of thing some people don’t mind, or maybe even find valuable. When a website starts tracking my web surfing actions after I have left their site, warning bells go off for me though. It makes me feel like I need to quarantine Facebook in Opera, do my regular surfing in Firefox, and use IE7 (running under Wine of course – I’ve been on Ubuntu Linux since June of this year) when viewing any of the sites Facebook lists as participating in Beacon. And that’s just not the way it should be.

I have upgraded WordPress and Gallery in an effort to get people to stop posting links to gay porn. Everything should work just as it did befire, and I can cross off “Stop people posting links to gay porn” off of my list of things to do before I die… if only for a few short weeks.