HootCourse lets instructors create communities where students and teachers can interact via status updates and tweets.
Users sign up by logging in via Facebook Connect or Twitter's OAuth implementation. Once in a course, they can push their tweet sized messages back to their respective social network with the click of a checkbox. If 140 characters isn't enough, users can connect to their Wordpress, Xanga, Blogger, or Posterous blog and post to it directly from the HootCourse interface.
HootCourse also features an embeddable mini-client, allowing instructors to put their class conversation wherever it's relevant (e.g., course management systems and course blogs).
Since May 2010, over 4,000 users have signed in and we’re now approaching 800 unique visitors per day.
Presskit'n helps organizations of all sizes easily build social media newsrooms.
Users can customize their room to match their organization's branding by changing the color scheme, adding a logo, and specifying links to various social networks. If an organization has multiple representatives, they can be easily added and given full or limited access to the newsroom's configuration and content.
The release editor enables representatives to share and collaboratively edit press releases (before and after they're published). It also allows the attachment of videos and photos, with cropping support, previews, and captions.
Once a press release is published, the site's realtime tracking features allow organizations to monitor how many hits their releases receive.
Simply put, Pyposterous is a Python API wrapper for Posterous with 100% coverage. It's also the coolest thing ever.
The library dynamically builds methods based on an IDL that lists all of the various functions provided by the API and their input requirements. Posterous' XML responses are converted into objects, with special cleanup methods called on individual attributes when appropriate. This means that if the Posterous team makes a change to an existing API feature or announces new API methods, I don't need to scramble to change a bunch of functions and classes in my codebase. All I need to do is update the IDL, and Pyposterous should be good to go.
I'm currently using this library as part of HootCourse's blog integration interface and she works like a charm.
Twitter + Mad libs = @libs. Yep. I’ve climbed that mountain. The rest of humanity is officially spared the trouble.
Twitter and Facebook users can sign into @libs and create their very own puzzle to challenge their friends. Alternatively, they can play one of the @libs that others have created. The hilarious and often inappropriate results can easily be tweeted or pushed to Facebook for the world to see.
You know you want to give it a try. Go ahead. I’ll still be here when you get back.
HAPPO (Help A PR Pro Out) connects public relations job seekers with public relations jobs.
It’s supposed to be a quarterly event, but there are always people seeking jobs and PR firms seeking employees. As a result, related tweets tend to roll in all the time with a variety of regional hashtags. The team at Match Strike saw an opportunity to organize these tweets into one easy-to-browse web app.
Before RedBeard, the online course development department at the nation's third largest university had no unified view of the data related to the courses they were charged with supporting and developing. This often led to unacceptable situations where low level support representatives had to escalate issues simply because they had no way of accessing the (nonsensitive) data required to answer basic questions from students and instructors.
To make matters worse, there was an internal database of contact information and course metadata that was being updated manually, even though IT had access to much of that same data as it was updated elsewhere in the institution via PeopleSoft.
It was a nightmare.
RedBeard, powered by a very bad PHP framework that I wrote myself, pulled in data from 6 different sources (WebCT, Blackboard 8.1, PeopleSoft, ORION, EIS, and Basecamp) and cobbled together a unified view from that information. This alone saved the institution countless person-hours and greatly improved the quality and responsiveness of our technical support.
It's a website. It's big. People use it.
I managed the team of developers that wrote the CSS, JavaSript, and HTML for ucf.edu. That meant extensive accessibility testing, cross browser compatibility checks, and occasionally having harrowing implementation debates with the project manager (who is a really nice guy and didn't deserve any of it).
When my team was done with their work, our static templates went off to another team that was responsible for incorporating our output into a deployment of a Java based content management system. As you can imagine, this wasn’t exactly an ideal strategy for building a website.
In many ways, our development process was dictated by various bureaucratic intricacies and not by established best practices. However, we persevered and the site turned out pretty good in spite of these challenges.
Need real time activity updates for the 10,000 Twitter users that use your app? Sitebucket is the answer.
It's a Python based, threaded site stream consumer that strictly adheres to Twitter's requirements for the site streams endpoint. All you need to do is pass it your OAuth credentials and a list of user IDs to follow. Sitebucket will figure out the smallest number of connections required and follow those users.
Additional accounts can be added to the follow list without restarting too. Sitebucket will simply create new connections for the new follows and then consolidate those connections later on if necessary.
Sitebucket is fully documented with Sphinx and lovingly tested with a healthy mix of doctest and unittest.