Monday, June 28, 2004

Back at Work
A lot of today was spent catching up on the last week. There was quite a bit of an email backlog to get through, though I'm holding off on the RDF mailing list until I feel ready to tackle it again... probably straight after I post this. If I'm lucky then half of the messages will just be calls for papers to conferences I could never attend.

The first thing I asked TJ this morning was whether or not the RMI reconnection code had worked. Unfortunately it hadn't. That's what you get for trying to implement something in a rush. It turned out that I had kept a remote reference to the Session Factory and was asking it for a new Session when the old one failed. Obviously if I can no longer talk to a remote Session due to a server restart, then I can't talk to the remote Session Factory either. Fortunately it was an easy fix for TJ last Monday morning, and the rest of the code worked fine.

That left me fixing the remote resolver JUnit error that I was left with just before my break. However things change in a week. Several changes were made which had a significant impact on the new resolver code, and yet not all of these changes were put into the resolver classes. Since the resolver code only gets built when explicitly asked for, the missing code got missed in the build process, and the classes were all checked in. These classes were missing methods, contained typos, and had other errors that you might expect from files that had never been compiled. While the changes really belonged to others, I worked on making the required fixes on my own checkout so I could proceed. Until I can make it compile I can't even get back to doing the tests I was unsuccessfully running over a week ago. I had all but 2 files finished by the time I left, but I may not need to do those last couple of files as it looked like AN will have most of it done independently by tomorrow.

Once I get the remote resolver tested and debugged I've been told that I should be able to get back to the inferencing code. I'm looking forward to that.

Today I heard the frustrating news that Apple Australia have no 12" PowerBooks left in stock. They also have no idea when they expect the next shipment.

One reason that this situation has occurred in the past was when Apple were about to release a new PowerBook range. With WWDC2004 being this week then that is definitely a possibility. However, Apple claim to avoid doing that anymore due to decreased sales in the lead up to each WWDC, so a new range of Apples might just be wishful thinking.

I received an email asking where I am with the triplestore I'm trying to build in my own time. Now that I'm back from my break I'd better get some more code written! Still on the file based hashtable, so I'll aim to have it done by end of the week.

Speaking of out-of-hours occupations, this blog held a surprise for me. Over the weekend I looked at the hit tracker for this blog, expecting to see very few hits as I hadn't been posting while I was away. To the contrary, I had more hits than I've had in weeks, and it was still growing. It turned out that it was almost entirely due to Danny Ayers. That was quite a commendation Danny (and very surprising). Thanks!

I was teased today for my newfound whuffie rating (which I'd never heard of before today). :-)

No comments: