Below are the 2 most recent journal entries recorded in whimsy00's LiveJournal:

Thursday, June 24th, 2004
5:14 pm	Did you know you can integrate a curve by cutting it out of a graph, cutting out a few squares from the gridlines to calibrate your mass to area ratio (and possibly compensate for varying thickness), and weighing everything on an analytical balance? I was talking with my boss the other day about integrating. Him: Well, you can just cut that curve out and weigh it, then cut some squares- Me: [Laughs]. Yeah. Him: What? Me: Just.. use the computer? Him: Oh, well, if you like. Me: Wait, you weren't being facetious? Him: No, it's very accurate. Well, I came in today and had to give it a shot. It turns out its DAMN accurate. Without gloves on I got 90-95% accuracy. Not too impressive. I put on gloves and made everyone else participate in an "integrating by cutting and weighing" (look at some old science journal articles, it is apparently a real technique and name) contest. I came in last, with 99.80% accuracy. The winner...100.0%...accurate to four significant figures, the limit of our balance (.0001 grams). If you work around an analytical balance, try this! If you don't, well, try it with metal or cardboard and see what the best you can do is :). (Comment on this)
Sunday, June 20th, 2004
9:44 pm	I read and discuss a lot of state of technology/social trends/typical tech blog ideas with friends and colleagues. Once in awhile I get an idea or wish I could share the comment with a wider audience. Usually we just banter about them, but once in a great while there's something that I think is worth sharing. The upside to this blog is that I'll commit the occasional good idea to physical media, the downside is that there may be low signal to noise :) Good post about a potential future for Google by Philip Greenspun. Web-based versions of desktop applications have been tried before, though. It's a nice idea, but there are a lot of problems. Not the least of them being browser incompatibilities. They've done a pretty amazing job w/GMail, though . . . a JavaScript applet (200K-ish!) that works in multiple browsers more or less seamlessly is no small feat. Even still, mail is an application that has minimal formatting/file format constraints. Manipulating plaintext is easy enough, and as far as UI goes, most users simply expect an interface much like GMail's simply because Hotmail and Yahoo! Mail broke us in for it. When people edit a document, they expect to see word or something very much like it. WebOffice is a nice idea, but I don't know how viable it is right now. My guess is for now, we'll see interesting applications like Froogle and GMail. A strength Google has now is that it has a lot of data from the search side of things. It also has a decent set of algorithms for taking on large, organically generated datasets. Something to do with bioinformatics or the genome project wouldn't surprise me at all. Search never worked this well, and new methods always give way to new problems we didn't even realize we had yet... (3 Comments |Comment on this)