Below are the 5 most recent journal entries recorded in nick_aragua's LiveJournal:

Monday, April 18th, 2005
4:13 pm	To date, I've gotten the camera to work separately, and the RFID reader to work separately. Today, I'm working on getting the RFID Reader API that my previous group wrote to send RFID events back to our main book application. I've got the code changes done that we need to get this stuff working, but none of the hardware wants to work. Since nobody else showed up today (and I showed up later than I said I would), we'll continue this tomorrow. P.S. Dear Eclipse. You know that "public static void main(String[] args)" thing? That's a "main method". The one you keep saying I don't have. (Comment on this)
Saturday, February 26th, 2005
1:00 pm	Off-by-one errors Apparently, the API we wrote this week couldn't handle reading data from more than one tag at a time. The reason: an off-by-one error when looping through all the tags returned by the inventory command. So now, you can throw in as many tags as the reader can physically handle (my guess: not too many). Also, keep in mind that waving tags around rapidly within about a foot or so of the reader is not advisable, as it will behave inappropriately (and also cause our API to throw "RFIDExceptions"). The front face is the green side, which can read tags within about a foot or so. From the back, it has a range of about an inch. So, if you want a bigger fridge, get a better reader (also, write your on reader communications API if it's of a different brand). (Comment on this)
Wednesday, February 23rd, 2005
12:49 pm	Ok, note to self (and everyone else): We'll all be better off if we stick to using the Texas Instruments RFID tags that we've got in the lab, until I engineer the RFID API to be a little more tolerant of differences between RFID tags. As a reminder, important differences include: Storage capacity Data block size / Max # of data blocks per data upload Supported command sets (Comment on this)
Tuesday, February 22nd, 2005
9:34 pm	RFID tag rigamarole To do list for tomorrow: - Establish which brand of RFID tags are being used. For "data write" and "data read" operations, "Block Size" (DB-Size) and "Maximum Block Count" (DB-N) might be different for different types of RFID tags. Additionally, different types of tags support different command sets. Finally, different types of tags can store different amounts of data. Reference section 8 (page 69) of the serial protocol document: http://www.cs.tufts.edu/~apokro01/rfidspecs.pdf - Write startScanner method. The general idea is to start a loop that goes something like this: while(not signaled to stop) { reset(); inventory(); compare list of responding tags to current list of present tags. for each tag that's in the current list but not in the responding list, throw out a 'disappear event' clean the tag out of 'current list'. for each tag that's not in the current list but in the responding list, throw out an 'appear event' add the tag to the 'current list'. sleep for some waiting period } - by contrast, stopScanner() will simply set some signal flag so that on the next iteration of the scan-loop, it'll catch the flag and stop processing. - Atomic interactions (such as 'read data', 'write data', 'inventory', 'reset', etc) will consist of both sending the command to the RFID reader and waiting for the response. - Also note that due to the whole 'threading' thing, an atomic interaction will need to acquire a lock on the serial port object before starting its communication and release said lock after getting the full response. (1 Comment |Comment on this)
Thursday, February 17th, 2005
7:55 pm	Tangible User Interfaces This is where I will put anything that I think is interesting, as related to the Comp150-TUI course at Tufts. This post is also a test. (1 Comment |Comment on this)