Featured App: BetterDate.Me
Our fourth featured AppHarbor application is BetterDate.Me, built by Wesley Tansey.
What is BetterDate.me?
BetterDate.Me is an app that analyzes your online dating profile and tells you how people are likely to feel after reading it. I built it as a demo of an NLP algorithm we've developed at EffectCheck that can measure the subliminal emotional impact of a body of text.
It turns out a lot of words have an emotional impact on us that we don't even realize. Studies have shown, for example, that if you're walking down the road and you see the word "ELDERLY" written on a sign, you'll actually walk slower without realizing it.
BetterDate.me measures these subconscious effects of your dating profile across six emotions: anxiety, hostility, depression, confidence, compassion, and happiness. Each emotion is scored on a simple scale of "very low" to "very high" and each profile is summarized using a catchy summary metaphor/icon at the top of the results.
The goal is to help people understand and potentially improve their profile to be more attractive to prospective dates.
How many users do you have?
We've just soft-launched to a few friends right now to get feedback. We're preparing to launch a marketing campaign for the site in a few days.
Why did you choose AppHarbor?
I have used Heroku before, but I'm not a huge Ruby fan, so when AppHarbor launched I was really excited. Thus far I haven't been let down.
Which technologies is BetterDate.me built on?
It's built in ASP.NET MVC 3 (Razor engine) and Fluent NHibernate.
How has AppHarbor worked for you so far?
Flawlessly! The workflow is super smooth, the dashboard UX is straight-forward, and support has been excellent.
What did you like the best? Where could we improve?
So far I haven't had any complaints. The database we use for BetterDate.me is around a million rows. When I was having trouble uploading that much data, Michael wrote a great utility to help me out.
I think the next important addition for me will be background tasks. Most of my projects involve some sort of machine learning algorithm, so it'd be great if I could launch a background task to run the algorithm and update the user when the algorithm is finished.
Anything else?
I really think AppHarbor is on to something. I'm planning on using it to build the full EffectCheck site, including our developer API. We're looking to find a few developers interested in scoring emotional impact in their own applications. If anyone would like an invite to the closed beta, just email me: wes.tansey@gmail.com.
Building on AppHarbor has been a blast. I'm looking forward to using you guys for a long time! :)
