A Retrospective on Cleveland qua æsthetic

Cleveland is a funny place. Like Wisconsin, it’s significantly greener than Illinois, and I find myself surprisingly aflush at the sight of all the verdure, but Cleveland is like a rusted, hulking monolith overgrown by jungle. The juxtaposition is frightening: while lost, we drove along so many leafy parkways lined with large embellished houses, but our destination was inevitably in the sprawl of the city itself, hemmed together with slummy, apartmented streets and the blackened faux-marble of the botanical gardens, the public library, and the enormous Presbyterian church on Euclid Ave. One building along a main drag called Chester St. appeared to have two dark bronze statues of skeleton warriors, as though someone had stitched together twisted detritus from a plane wreck to make this thanatotic duo, their sharp ribs prominent, their wielded swords even moreso.

It rained all three days we were there, in random spurts of thunderstorms. Cleveland’s drainage—at least downtown—isn’t the best, as even a relatively gentle storm left standing pools of water on the roadsides and muddy ponds in the grassy medians as late as the next morning. I’ve already said that Cleveland is organic, but add near-constant construction and rain to the list, and the city itself is a violent storm of confusion, parkways ceding to slums, trees to effaced apartment complexes, prestigious university to vacant, overgrown gas stations.

I realize that I saw but a fraction of the city, but I would expect that the university-dominated section would be a thriving little community of its own. Perhaps things are different during the school year: I saw few college students at all; only middle-aged conference attendees and the occasional gaggle of misguided tourists—why Cleveland?—gaping at the squalor.

A Retrospective on Ben qua Conference Attendee, as Well as a Discussion of Code

But enough sentiment. My own journey to get here, in a plain grey van, speeding down I-480, was just as mixed. I woke at 7:09 this morning and —my understanding being that only my boss and his boss were going to the presentation at 10:30—went back to bed, to be awoken exactly four hours later at 11:09 by the sound of little kids (?) playing soccer on the athletic field next door. I was alone—all four of my companions had gone to the presentation, had even called for me through the door of my sleeping area, but to no apparent avail. The neck-ache I awoke with had nothing on the painful feeling of being such a schlub.

I mentioned last night that the recurring theme of the conference was that Oracle’s portal is utter shit. To that end, I find myself excited not by the opportunity to work with it, my loving ministrations being severely limited by its obstinate nature, but the opportunity to work with Tim, our DBA and resident Java guru, on his separate-but-not webpiece that will tie into the portal. Having seen yesterday’s presentation on Web 2.0 and AJAX, he and I decided that it would be a good idea to integrate it into his webpiece. Since the logic that drives the overhead menu requires database calls, we thought that using AJAX to load content into the specified area would reduce overhead and feel more responsive to the user, broken back-button or no. So, during the afternoon session on Thursday, I found a little AJAX script and made a working prototype with static HTML. Tim came back from his afternoon session with news of OS3Grid, a Javascript-based table system that allows for dynamic reordering, editing, and effects that wouldn’t be possible in IE with pure CSS. The only problem is that they don’t want to work together: the function to render the grid is run on page load, which doesn’t work if you are loading the grid’s host page with AJAX. I hacked up a fix by adding a new line to the AJAX script’s pageload function that reevaluates OS3Grid’s table-rendering function, forcing the grid to be rendered when the page was loaded with AJAX. The only problem is that each grid has a specific line that has to include the name of the parent div that hosts the rendered table, and since we’re guaranteed to have pages with multiple tables, we can’t simple hard code a standard id into the function and let it go. I will need to research other methods of forcing the script to run.

These are both neat Javascript tricks, but the real question is how well we can get them to work with the database. The AJAX function for navigation is entirely client-side: there’s not yet AJAX to submit any forms or grab any data from the database—it merely calls a JSP, which in turns talks to the database. One of the interesting features of OS3Grid, however, is the ability to edit table contents, and I’d like to make it so that an edited table cell submits—via AJAX, of course—back to the the database the changed contents. For that, I’ll need Tim’s help in using AJAX to talk to Java to talk to the database. It’s possible, of course, but my experience with Java is so limited that the slightest problem will sending me running to Google. Still, I think that this has the potential to be a really great interface, and I really like working with Tim because he’s truly excited about what he’s doing, and he asks things of me that push my limits and force me to improve my knowledge set.