Experiments
-
- Flavor Social
-
Social media is transient. It is not (at time of writing) easy to archive and explore your past social media history. Flavor Social was originally born out of a desire for displaying Twitter posts on my website without worrying about Twitter being down or slow and causing an annoying page slowdown or error (as it used to quite often in 2006-07). As I began to experiment with caching and repurposing my Twitter feed, I began experimenting with other APIs to see what data I could cache for later use. While this app could be highly useful to archive and aggregate anyone's social media presence, I have to this point only used my own accounts in the app.
The app currently has support for multiple accounts of each type and presently supports the following APIs (but needs updating to support new features of some of the newer features of the APIs): Twitter, Tumblr, Flickr, Digg, Last.fm, Delicious, and formerly Pownce.
In addition to the base app, I have developed a "TweetEnhance" class that takes the content of any given tweet, wraps links with appropriate HTML links (includes an option to remove URL shortening where applicable), hashtags, username mentions, stocktwits mentions, and automatic retrieval and display of certain image/video services (currently includes twitpic and yfrog).
An invite-only beta is expected during 2010 for a refined, hosted version of the service with a do-it-yourself production install available shortly after that. Flavor Social is what powers my Firehose section of this site. Similar to my Forecaster.php experiment, one of my intentions is to do some visualizations and data analysis with the resultant data after a prolonged period of time.
- Technologies Used
- PHP, MySQL, XML, CRON, Twitter API, Tumblr API, Flickr API, Digg API, Last.fm API, Delicious API, Pownce API (R.I.P.)
- Status
- Active — Under Development
-
- Forecaster.php
-
A relatively simple National Oceanic and Atmospheric (NOAA) Weather Data Caching Experiment, Forecaster.php, either run manually or via CRON is a script that interacts with the NOAA Weather API to retrieve data, sort it as needed, and store the data. It takes a variety of parameters such as location data, API request variables, and output method either set as options in the file (for CRON) or through optional command-line input if executed manually.
Currently, I have been caching hourly data via CRON for the town I where I lived 2008-2010 related to Forecasted Weather Conditions, Expected Cloud Cover, Extreme Weather Conditions, Temperature Values, and more for just over six thousand hours. At the one year mark (8,760 hours), I will be using this data to provide a comparison between forecasted data and actual data over the last year as well as create some interesting visualizations regarding how forecasts shift over time. It should be noted that although Archived Forecast data is readily available, it is rarely is publicly available at this granular level targeted on a specific location without a personal weather station (which simply records data without forecasts), thus the purpose for this experiment.
- Technologies Used
- PHP, MySQL, XML, CRON, NOAA API
- Status
- Active — Data Collection Stage
— Steve Wozniak, founder of Apple and engineerAll the best things that I did at Apple came from (a) not having money and (b) not having done it before, ever. Every single thing that we came out with that was really great, I'd never once done that thing in my life.
