Did a bit at the weekend, and finished off this morning. I now have a four-page handout on using the Image Markup Tool for this afternoon's class.
Set up, participated in, debugged and debriefed a video conference with CANJAS nodes. Tim Miles from PAS connected his iBook to the system and we had a bit of a practice for two upcoming events on the 13th and 15th of Feb.
Problems we had:
1) The data display that is supposed to display incoming data (from remote nodes) didn't want to work. I tried a CRT but that didn't work either. I plugged the CRT directly in to the RGB connection on the codec (the bigger of the two Sony boxes) and it worked. I'll need to test this with the new TV before the 13th.
2) Video was very bad. So much lag time made two-way conversations impossible. This is not normal in our experience so far. It was the same for all nodes, so I don't think that it was our network connection causing the problem on our end.
After installing and testing IMT in Lab C, and testing the projector (the monitor res was too low on the teacher station, so upped it to 1280x1024), did the first half of a handout for the class. Found an image we can work with (John had promised to provide one but didn't), and got half way through the handout. I'll probably have to finish this at the weekend, because of interruptions during the day.
With Greg, tested a simple way of installing one of our apps into a lab, using the /VERYSILENT switch that all our Windows installers have courtesy of InnoSetup. You create a task that pushes out the file to the temp folder, then executes it with the argument VERYSILENT (no slash). Works a treat. We can now look at the other command options in Inno to figure out if we make make this more sophisticated.
It appears our first shot at this was based on a false assumption about the database structure and data. After another look at the tables, we diagnosed the problem and ran another query against the db, retrieved the results, and formatted them as before. These results appear to be right.
Tested running the IMT installer with the /VERYSILENT switch, and it ran just fine with no prompts or dialogs, so it looks as though this will be easy to install it in lab C tomorrow, by pushing out the installer then running it remotely as administrator on each of the machines, with this switch. The app is needed for John Lutz's class on Monday.
Ray Siemens needed some SQL work (confidential). Logged into the mySQL db, figured out the structure, and (with help from David and Stew) wrote and SQL query to retrieve the data. Saved it as XML, and then wrote a little XSLT file to spit out the list in an appropriate format.
The course listings etc. on the history dept site rely on a javascript array in a file called hist.js. The dept has had a number of requests to change the course titles, descriptions etc. to keep them in accord with what is on the calendar site. I took code from the philosophy site and modified it so that when a specific php page is called from the browser, it goes to the uvic calendar site, grabs the appropriate information, transforms that into a properly formatted etc. javascript array and then writes it into the file hist.js.
Only thing to do manually is once a year the Year variable has to changed to the current academic year. As always, the list of which courses are currently offered (kept in a separate js file in the same folder) must be edited manually.
This code could easily be re-used for any other department wanting similar capabilities.