Our assignment for the week was to take our idea for an ITP-oriented SMS application and execute it. After class, I spoke to Gordie about the waitlist notification idea, and while he liked it, he said he needed to be in contact with people individually because every waitlist rise has to correspond with a class drop and thus he has to find out what each person wants to do given new choices. Makes sense.
So then I thought I’d go ahead with the class list in relation to the current courses…but I think this idea is better suited to a web page, especially since there isn’t an easy unique ID or code that people can text to get results for individual students or classes, and of course the text message that would be sent back would be truncated due to character limitations so long lists of names or courses wouldn’t make it through. Even to do a web page would be prohibitive in that I can’t get an official list of student enrollment in classes to work with, so I’d have to solicit this from everyone myself, which would be a pain.
I had other ideas that involved lists of students – for instance, if you text a Zodiac sign like “Libra,” you could find out all the people who have that sign, or you could text “PC” to find out which students have PCs instead of Macs – but once again, this information would have to be gathered from people individually and there would be the same limitation of names being sent out.
For assigments like this, it’s best to keep things simple, so I decided to build off the class example, which queried the Yahoo weather RSS feed with a zip code to return the corresponding temperature via text. I sat with Cindy and David, and together we troubleshooted the PHP code to get the example working with our own databases on the NYU ITP server and with our TextMarks accounts. The idea I came up with that night was to pull headlines from the RSS feeds of geek-related news sites like Cranky Geeks. I just had to find unique tags or words (by viewing the source code) around the latest headline text that I wanted and make them the values of the $stringStart and $stringEnd variables. I also cleaned up the code and folded the function into the “if” statement relating to the input codeword. I got this to work on my mobile by texting “cranky” to TextMark’s 41411 number, after also texting my account keyword. I got back the latest headline (and also tested the other returns specified in the “else” sections of the code). I had planned to expand this idea to include four other RSS feeds from other sites but decided to try something else.
The next day I sat with Matt G, and we came up with the idea to get data from the ITPedia site that would be relevant to ITPers. Matt focused on the TNO page and I chose the DriveBy one. Both involved stripping excess code between the start and end strings – we looked up the appropriate function on the PHP reference site, which is strip_tags($text). It also involved replacing some words in Matt’s case, which meant using the str_replace function. The DriveBy page lists events by date, so I was thinking of creating an automatic way to determine the next DriveBy instead of finding it manually. I looked at the getdate function in PHP, which returns an associative array, and contemplated calculations that could compare the current date with the dates listed on the DriveBy page. I started to work on this but ran out of time and also decided that it was uncessarily complicated to execute for this assignment. Will ITPers use this? Probably not, since it’s easier to check the wiki than remember the 41411 number, but it does give people another option to remind themselves of the DriveBy topic if they’re out and about, have the 41411 number stored on their mobile phone, and if they don’t have an internet connection or the time to check online. I plan to finesse it and send it out to the list to test.