In early June, I wrote about some problems Susan and I were having and how Susan wasn't sure whether she had broken up a while before that point. We are still dating. We are making good progress on the problems we are having, but we keep finding new issues. The only thing I can really say at this point is that signs point to complexity.
Back in February, I wrote about concerns that I am stagnating at MIT. Well, part of that entry is no longer true. Kerberos has become much more interesting than it was then; future design challenges promise to be even more interesting. It looks like we may get chances to solve some fairly long-standing design limitations in Unix credentials and in how cross-organization authentication is handled. I'm enjoying what I'm doing. But I still feel that sticking with this job long-would be a mistake. I wouldn't mind doing the same job in a context where I could work on building the right team. However I suspect I'll have to decide between a chance at an ideal work environment and continuing on what I consider to be a good project.
The new laptop is working better than when I last wrote about it. I have sound working using the ALSA driver. The i810_audio driver has better power management and seems to be a better driver, but it deals poorly with short audio input like a single key name being spoken at high speed. I have wireless working. Thanks to tytso, I have my speaker beeping again. ACPI sort of half works; I can suspend to memory and even resume. However, minutes after I do so, the sound driver gets confused and keeps repeating the DMA buffer. This seems true even if I unload the sound driver before suspending. More experiments are needed. Neither pmdisk nor the other module (swsusp) for suspending to disk in the 2.6.6 kernel actually seem to work. I really like the hardware though; IBM continues to do an impressive job of laptop construction.