These are the old pages from the weblog as they were published at Cornell. Visit for up-to-date entries.

June 30, 2004

More Catching Up

After the WIPS session at Usenix I spent some time with George Candea from Stanford. I like the work he presented on taking the crash-recovery model to a finer grained level. He is looking at ways to control ( = crash) components (J2EE) if they are misbehaving in some sense. He has an interesting model for control over the input side where a sort of software fuse does input checking to avoid erroneous behavior by the component. If bugs or risky behavior are detected that are related to input, the fuse can be updated to avoid problem, while the component, which may be a black box, can be updated at a later time.

While I was ready to go into an evening of quiet seclusion Ellie Young grabbed me to go dinner with Dave Presotto and Clem Cole. Ellie is Usenix's executive director and the heart and soul of the whole organization. I hadn't talked to her since we ran the Windows Symposia and it was great to catch up on all things Usenix. She has done an incredible job on improving the professionalism of the Usenix organization. I was surprised to see that after Rob Pike Google now also managed to snatch up Dave Presotto. I am sure Google will not run Plan 9, so it will be interesting to see what they will be doing for them. I do think it is getting lonely and dark in the Unix room at Bell-labs.

Back at the hotel Jay Lepreau joined us briefly and we talked about how we will be using his Emulab software to get the experiments at the Durip cluster started. There is room for collaboration which I am sure is going to happen.

After Dinner lounging: Dave Presotto and Ellie Young

Posted by Werner Vogels at June 30, 2004 09:09 AM