Facebook app experimenting
I haven’t had any reason to develop a full Facebook app yet, but since I was curious about their API I’ve been playing around with it. The SATS app was originally something I built years ago for my own webpage when I was a member of SATS (the gym). SATS’ search for classes didn’t have the interface I wanted, so I rearranged the search results a bit more to my liking. (It’s all ugly hacking, stripping data from an HTML result.) To my surprise, they haven’t changed their search functionality since then, so I ported the search to a Facebook app as an experiment. The app resides on my server and I rebuilt the whole front-end to return a JSON object with the data to show, then displaying it with FBML. I’d like to implement more of the existing search functionality, but as this has the status of being an ugly hack depending on SATS not rebuilding their search, I don’t think it’s worth the effort. Oh, and I think the app is buggy in IE at the moment.
After writing the SATS app I got curious about Google App Engine and wanted to do some sort of Hello World to test connecting GAE to a Facebook app. I built a very simple function just to display your friend list as an XML file.
The reason for this is that many of my friends complained about not being able to trace the people disappearing from their friend lists. This is also a convenient way to get the user ids for your friends if you’re developing an application and need test data. I’m not allowed to save versions of friend lists due to Facebook API user terms, but this way the users can save versions of their list to compare.
Comments OffFINALLY!
I’ve been wanting to build a couple of really cool social web sites for years. Not very original, right, now everyone are at it. This wasn’t the case a few years back though, so why didn’t I?
1) Lack of time
I had a full time job. I did work with building a community platform in Java, and learned a lot by doing so, but it was never really my ideas that were implemented.
2) Lack of expertise or a team to back me up
I like to think of myself as the one who actually can build anything online, by myself. This is true, to some extent, I can do almost anything. The problem was that what I wanted to do involved some really complex social connections and to my knowledge, databases weren’t really up to that at the time. I needed some real geek power there, which weren’t available at the moment.
But in the last months I’ve started to realize that technology has reached a level where I actually can get stuff done! The ideas I had back then may be obsolete now, but there is so much potential out there right now.
I can use the existing social websites to handle the logins and accounts (Facebook, Google), which automatically creates a great channel straight to the potential users, and I can use Google app engine to host the projects. Cloud services really has helped reduce server costs and eliminated the server config time!
But, what really may be the key I’ve been waiting for is the graph database Neo4J. Pure geek power!
It’s all there now, like someone is constantly tossing stuff at me, yelling “catch!”, and best of all, I have the time to do my projects now.
4 CommentsExperimenting with social web
I’m really excited about all the fun ideas I’ll be able to build once I quit my full time job and start freelancing (it’s only a few weeks left now, weeeee!).
When I was a student a few years back, I had time to build feminetik.se (a feminist community) together with Josefine Alvunger. We didn’t do it for profit, we just had a vision. feminetik.se was our baby, and even if we listened to the users to some extent regarding smaller features, we basically did stuff our way.
Today I wouldn’t choose to work quite like that. feminetik.se never reached any larger amount of members, even if the ones that were there were quite productive (the forum has over 300 000 entries today). There is so much to learn from those years, some things we did right, some could’ve been more optimal. That’s a whole other blog entry (or five) though.
Now I’d like to build a website where the users participate in creating the site from start and a few days back I got an idea from a brilliant Facebook group where users give stuff away to each other. They simply post entries to the wall saying what they want to give away and where the item can be found. And this is not just an easy way to get rid of old stuff, it’s a whole culture of recycling and environmentally aware people.
There are sites doing this on the web already, but they’re mostly simple forums. I want to build on the social aspects of this:
1) Geo-tagging of the objects making them easy to find on a map
2) Not use categorizing like at “Blocket”, instead use tagging. I haven’t solved the issue of being able to browse through objects planlessly yet though.
3) Enable follow ups, for example letting users post pictures and stories of them using the object (”Look here is my daughter playing in the sofa you gave us!”, “Here I am at a party in your old dress!”).
4) Make it easier for people with similar interests to find each other. For example, maybe all latte moms and dads in the southern part of Stockholm would like to have a meetup for swapping old children’s clothes?
5) Use existing technology. Google App Engine, OpenID, Facebook connect etc.
I pitched the idea in the Facebook group, which has 5000 members, and people seemed very positive and enthusiastic. Some even offered to help out! This is still only an idea, but there seems to be an interest for a site like this and it’s perfect for me to use as a playground for a while. We’ll see where this leads, I’ll blog continuously if I decide to start developing this site!
No comments
Networks
