Post Creativity and Cognition

Posted: July 8, 2013 at 4:38 pm

I am back in town for almost a week and still struggling with jet-lag. The conference went well, and I will post my paper as soon as the proceedings are on the ACM website. The papers in the conference leaned quite a lot to the HCI side, so I felt like my work was quite a bit at odds with the community. Still it was a good opportunity to meet people and see Australia for the first time.

The exhibition of work in progress on the dreaming machine did not go as well as the presentation. The system crashed every single day of the conference, and was not even running during the opening. I ended up pointing the camera out the window at the Sydney sky-line, due to poor lighting in the exhibition room. Even if I had restarted it for the opening, it would have just shown black anyhow. I only had one chance to run the code in a debugger because I could not run it full-screen and it took somewhere between 8 and 24 hours to crash. I did notice what looked like an URL error for the code reading the image from the IP camera. I changed Ethernet cables to the camera and tried again, and it acted exactly the same way. There was no network, just Ethernet from the computer to the PoE injector and Ethernet to the camera. One of those two cables was the one I brought and did 4 day tests on at home. Starting Friday afternoon I ran the (same!) code in the debugger, this time on my local network, and it ran all weekend without problems. It was working fine until it crashed when I tried to saving the percepts late on Sunday night. In short, I have no idea what caused those daily crashes in Sydney. Two bad Ethernet cables seems implausible, as does some issue with 220V power… In this latest test at home I even tried disconnecting the camera while the system was running, which would cause the segmentation thread to stall, waiting for a connection, and then only crashed when the timeout was exceeded, which was reported using a different message than previously. For long-term, I need to look at making that code robust so that it would just wait indefinitely for a connection, rather than segfault.

Following is the plot of this recent test, still showing the memory leak, but at least running a few days: 147748_frames_1500bgc_500fgc_DEBUG