Stories of Coincidental Electricity

The annual Tanner Electronics Robot Show was on Saturday, April 14. The DPRG held their annual robot talent contest concurrently. So, not suprisingly, I was planning on working late the preceding Friday to get my new little robot, Robozoa, into shape. This mostly involved finishing some hardware-related things like wiring from the H-Bridges to the motors and from the motor encoders to the microcontroller. This sort of work is better done at the DPRG Lab where there are plenty of tools and test equipment to make it easy.

The weather prediction was for rain in the evening, so my plan was to head up to the DPRG immediately after work. Not suprisingly, a last-minute work-related emergency held me up for a couple of hours. By the time I was finally able to leave, a torrential rain had started. When a break in the rain materialized, I ran out to my car; only to get a phone call before I was out of the parking lot. The call was from Susan, who was holed up at home in a bathroom with the three cats because the TV had just announced a tornado was headed her way. She said the tornado watch area extended to the downtown area where I was, so I decided I’d be better off inside the office than in my car until things calmed down.

I ran back through the now heavy rain into the office. As I dried off, I clicked up a few weather radar sites. Sure enough, there were some nasty looking thunderstorms headed my way. They passed over Irving, where Susan was, without any serious damage resulting (it’s now unclear whether the reported tornado really touched down or not). The worst of storms were now north of Dallas in the Garland area, where the DPRG Lab is located. I decided to settle in and do what work I could on the robot at the office. I finally left about 1am by which time the rain had stopped. I was a little annoyed that this series of events had kept me from making it to the DPRG where I could have worked more efficiently.

The next morning, I showed up at the Tanner’s event. The previous night’s storm had brought with it a freak, one-day cold front. Despite the cold, a fair number of humans and robots showed up to participate. But, more interestingly, several people said they’d seen the DPRG’s building in Garland on the news. There were firetrucks in the parking lot. Apparently it was hit by lightning. Eric Sumner, Ed Paradis, and I decided to drive up to Garland and check out the damage.

From what we could tell, the lightning hit the transformer immediately behind the DPRG building. It largely destroyed the power line between the transformer and the building, reducing it to a series of short fragments. The power meter was completely destroyed. The charred metal casing of the meter was still on the wall, surrounded by blackened bricks. The transparent housing and meter electronics, or the remains of them, were found on the ground. The meter had contained several boards with surface mount components. The lightning blast had desoldered all the components and completely vaporized many of them. Inside the building, the main breaker box was also a charred mess but it appears the breakers vaporized so quickly that it limited the damage to the downstream breaker boxes.

By Tuesday power had been restored and we were able to evaluate the damage. Remarkably, the only losses discovered were a single surge protector and one very old dot matrix printer. Aside from those two casualties, test equipment, networking gear, computers, all seemed to have survived no worse for the wear. All thing considered, I’m glad I wasn’t around Friday night when it hit.

Random Free Software Stuff


One of my longer term ToDo items made it to the top of the list this week. I’ve been setting up a Subversion repository for the Dallas Personal Robotics Group. The DPRG has a number of programming projects in various stages of completion ranging from working code to idle talk. Having a repository like Subversion will make project development easier, particularly for projects with multiple programmers. All projects hosted in the new Subversion repository will be Free Software and/or Open Source.

I used Subversion v1.3.2 for the initial setup. ViewVC v1.0.3 was added to provide a web-based interface to the repository. Last, I added Highlight v2.4 to provide some nice code highlighting for the ViewVC code browser. Everything is running on a CentOS Enterprise Linux box that also hosts the DPRG website.

At this point, everything seems to be working but I still need to customize the ViewVC templates to tie the look-and-feel in with the main DPRG site.

In addition to DPRG projects, I’m going to keep the mod_virgule codebase there too. Mod_virgule is the code used on and Advogato. Getting mod_virgule into Subversion was a good learning experience. I used the trunk, branches, tags layout recommended by the GPL’d O’Reilly Book, Version Control with Subversion. I used the last 2004 release of Raph’s codebase, version 1.41 as the initial commit. I created a separate branch for Raph’s code and also tagged it as release 1.41. Then for each of my releases since 2004, I committed them and tagged them as a release.

Ubuntu and Proprietary Drivers

Seems like everyone has been talking about Ubuntu’s decision to start including proprietary graphics drivers in the Distro to support flashier eye candy on the desktop. There’s been a lot of discussion and some flame wars over the issue. Rather than joining in flame wars, how about an an alternative? Why not put that energy into making sure there are free drivers for ATI and nVidia?

There are currently free ATI drivers with DRI support for 3D acceleration. Maybe someone could find out why the free driver is not suitable for the Ubuntu folks? Not fast enough? Missing a critical feature? Maybe it can be improved enough that the Ubuntu developers would reconsider their decision.

The free nVidia driver doesn’t support DRI so there is no free alternative yet for nVidia cards. The nouveau project is working on the problem. They’re making fast progress but they could use your help. They’ve developed a tool called called REnouveau to assist with reverse engineering the nVidia hardware without violating the license on the proprietary driver. If you have an nVidia card, you can help by downloading the proprietary driver and using this tool to generate dumps of test data for your card.

For a few more ideas on how you can improve Free Software support for 3D accelerated graphic cards in general, visit the Free 3D wiki.

Robots and Linux

What free time this month wasn’t sucked up by the Advogato migration was spent working on Tankbot GTR, the current DPRG group robot project. We now have the Mini-ITX mother board that Via donated mounted on the robot. We were doing our initial testing with an old 800MB laptop hard drive but it really sucked the batteries dry quickly. So I picked up an IDE to CF adapter and Martin donated a 1GB CF card. For the moment, I just used dd to move the entire hard disk content to the CF card. This is working surpisingly well considering we were running an old Redhat 9 distro intended for the desktop.

While most distros offer bootable CD images of one sort or another, almost none offer bootable CF card images. Many provide overly complex instruction on how to get their distro to boot from a CF card but few provide something as easy to use as a simple image file that you can copy and boot. Once exception is Flash Puppy, so I’ll probably be experimenting with that later this week. I’m begining to think there might be a real need for an embedded linux distro targeted at robot applications. And one that’s as easy to install as copying an image to CF card, sticking it in a motherboard, and booting.

Art, Cars, Cows, and Robots

As I stepped out of the Deep Ellum Subway where I ate lunch today, I saw a wondrously strange vehicle drive past. It was a Chrysler covertible covered in bovine-themed mosaic tile, with a giant, longhorn and barbed wire hood ornament thing, and driven by a beautiful girl. We briefly made eye contact as she passed. As she drove away, I noticed the words “Cow Goddess” emblazoned on the back of the cow car. Moo.

Seeing an unexpected art car reminded me that I recently posted a few photos from the 2006 ARTFEST event in Addison and the 2006 Deep Ellum Arts Festival.

Are robots art? I think so. The 2006 Tanner Robot Show was held recently and I shot a few photos there too (if you want more, Tanner’s posted some official photos of the event). I’m particularly fond of Ron’s pink bunny-laden robot, though Frank’s Zombarbie also stands out.

I’m still enjoying my Fuji S5200. While I miss some of the flexibility and image quality of my good ol’ Canon T90 film camera, I don’t miss the cost of film processing.

Storms, Crustaceans, and PQFP Chips

There’s a major thunderstorm passing overhead right now. The noise from the rain hitting the roof of our building sounds like loud applause. Ten minutes ago it was warm and dry outside but that’s Texas weather for you. I’m working late again tonight and I’ve been nervously watching the lightning strikes all around the area and hoping one of them doesn’t take out our power. I replaced the batteries in two of our UPS units this month, so we’re probably ready for it if it happens.

Susan is working late too. We took a little break earlier and tried another of the new restaurants that’s opened here in Deep Ellum. There seem to be new ones opening all the time. Tonight we tried Crustaceans, a creole/cajun place. It’s actually the reincarnation of a New Orleans restaurant that was destroyed by hurricane Katrina. The owner, chef, and some of the employees ended up in Dallas and a local organization helped them re-open their restaurant here in Deep Ellum. It’s the real thing. If you’re in the area and like that sort of thing, definitely check it out. In the past week we’ve also tried Tarantino’s (Italian food, a bit expensive but very good) and Kim’s Cafe (breakfast, burgers and sandwiches – so so).

The website design business is booming lately, leaving me not very much time for more fun things like robotics. The latest group robot project at the DPRG is proceeding nicely without me. I’ve been working on my own robot off and on. Just as I finally felt like I’d wrapped my head around PID algorithms and got some working code, I managed to toast the 68332 on my MRM board. I’m guessing it was a static discharge. It’s a surface mounted 132 pin PQFP chip; not exactly something you can replace with common household tools. I ordered a few new chips, though, and a friend with a hot air rework station is going to swap out the bad one for me.

Robonexus 2005 Debriefing

I’m finally getting caught up on things since returning from Robonexus and one of the last things on my ToDo list is posting something here about the trip.

Before I get to that, I should mention that Susan and I went to the State Fair of Texas this year just prior to my Robonexus trip. Nothing new and amazing to report there but I used the Fair to test out a new camera, the Fuji Finepix A345. The A345 is an inexpensive 4.1 Megapixel pocket camera. I still shoot a lot of 35mm film but didn’t want to lug my Canon T90 and associated gear around Robonexus, so I picked up the Fuji to try out as an alternative. If you’re curious, take a look at a few of the 2005 Texas State Fair photos or the Robonexus photos I shot with it.

Okay, so on to the Robonexus trip itself. Like most trips, it started with the hassle of airport security and placing all my belongings into little plastic bins. I filled one bin with cell phone, camera, keys, change, belt and shoes. It took another for my laptop and a third bin for the laptop case. This was followed by some hopping around on one foot while trying to put on shoes and simulatneously hang to my other stuff so it wouldn’t get stolen.

I’ve been to a lot of places in California but this was my first trip to San Jose. Random San Jose info: It doesn’t look that different from the Dallas area. They’re still using incandescent traffic lights instead of LED lights. The cross walks emit all sorts of weird sounds that are presumably intended to assist the blind. We’ve got nothing like that here in Dallas. The Kinko’s in San Jose have service as bad as those in Dallas.

I ended up in Hotel Montgomery which is within easy walking distance of the convention center so I didn’t need a car. I got in on the first day and picked up two sets of credentials: a press pass as editor of and an exhibit staff pass for the Dallas Personal Robotics Group. Because the DPRG didn’t really have enough advance notice to prepare anything interesting, we ended up just placing some flyers on our table that described the group. I had the flyers printed down the street at a Kinko’s (see above comment on Kinko’s).

I only had time to sample a few of the talks including Matt Mason’s overview of robotics and AI research at CMU, Michael S. Chester on launching a robotics company, Max Chandler on robotic art, and Stewart Tansley on Microsoft’s plans in the field of robotics. I showed up to hear Phillip Torrone of MAKE but he was a no-show. Most of the talks were interesting but primarily technical in nature. One non-technical talk, Joanne Pransky on The Frankenstein Complex and Its Impact on Robotics, brought up more philosophical subjects such the human reaction to robots with intelligence, emotions, and sentience. The talk left a number of people, including myself, involved in debates with other audience members over the subject. Some people still seem to be made profoundly uncomfortable with the idea that machines made of metal and silicon may one day be as good as us meat-based machines at thinking and feeling. No mob of angry peasants with torches appeared however, so perhaps things have improved since Frankenstein’s time?

Along with several other Dallas attendees, I escaped the conference for a few hours on Friday and drove up to Stanford. Sanjay Dastoor, a DPRG member who is now a student at Stanford, arranged a tour of some of the robot labs for us. We checked out the Stanford quadruped and also got to see some of the Stanford Sprawl robots in action (they’re fast!). After the robot labs, we took a few drive-by photos of the Frank Lloyd Wright Hanna House which is on the Stanford Campus. I also got a chance to check out the San Jose Museum of Art since it was just across the street from the convention center, though I had to do that one by myself. It was worth the visit and I enjoyed the Sandow Birk exhibit in particular. Overall I’d say it rates higher than the Dallas art museums but not as high as the museums in Ft. Worth.

Back at Robonexus, I spent a lot of my time networking. It was good to meet so many people in person who I deal with online regularly including most of the fine folks at Servo Magazine. The conference includes an interesting mix of high-end commercial robotics companies such as iRobot, commercial hobby robotics suppliers such as HiTec, and non-profit organizations such as the Robotics Society of America. There were also demos of many robot contest formats including NATCAR, Botball, FIRST, Robo-Magellan, and the Trinity Fire-Fighting competition. One suprise this year was the arrival of Lindz Lawlor and his Electric Giraffe. The Electric Giraffe is a huge mechanical quadruped equipped with more lights and audio amps than your average dance club. It walked around belting out dance-velocity Kraftwerk tunes. The Electric Giraffe was created for Burning Man, which seems to be a venue increasingly used by robot builders to demonstrate their work. After talking to Lindz and others who’ve been there, I started thinking it sounded like a lot of fun.

I returned to Dallas from Robonexus with two general feelings. The first was that the DPRG needs to go to Robonexus 2006 with more people and plenty of hardware to show off. The second was that the DPRG needs to think about creating something worthy of Burning Man. Interestingly, I’ve discovered that when you mention attending Burning Man, people will either react by saying something like, “What?! Burning Man is jest a bunch of nekkid hippies runnin’ around!” or they’ll say, “I’ve always thought Burning Man looked like fun but I’ve never had a good reason to go”. The DPRG members seem equally divided but I wouldn’t be suprised if we manage to make it out there in the coming years.