Ask YC: How would you teach a Facebook Apps course? | Hacker News Reader

Ask YC: How would you teach a Facebook Apps course?

6 pointsbkrausz12 comments

A while back I was given permission to run a course at Carnegie Mellon entitled "Creating Facebook Applications". I have free reign to do what I want with the course, but from conversations with people in the industry, I have narrowed down a few specifics. Here's the info:

* The course will have 20-30 students in each section and will meet once a week for 80 minutes

* Students will work in groups of up to 3, and will develop 3 applications over the course of the semester (one sample "learn the API" application and two legit applications)

* There will be a product fair towards the end of the semester showing off the apps.

Since I'm sure most people here are pretty familiar with the past/present/future of the Facebook API, I figured I'd ask what you would like to see if you were taking such a course. Specifically:

- What topics would you like to see covered besides virality, monetization, and scaling?

- How should I measure success of an app?

- How much should I guide the application concepts, should I restrict one of the 3 apps to "education themed", or should I let the students run wild?

- Any speakers I should get in touch with to see if they'd like to give a talk (I have a few in mind already)?

- Is it worth inviting VCs and other influential people to the product fair? It's much further for most of them to travel than going to, for example, Stanford...

I'll be spending the summer preparing for the course, so any input to help direct my lectures would be greatly appreciated.

Why make them create 3 apps? You want 30 super-extra-cool poke apps?

Why not make them make 1 app that actually does something? That means, no pokes, gifts, quizzes, walls, etc.

Haven't you noticed that people HATE these apps? And that there are literally tens of thousands of them? That's about all you're going to be able to create in 1/3 of a semester, as a 1 day a week class, having no facebook experience.

Also, the api is not so difficult that they need to make a throw away app. You can figure it out as you go along. It could all be covered in 80 minutes of lecture.

Is it worth inviting VCs? No. If they make ridiculously useful apps, they'll get attention without a VC circle jerk.

"virality, monetization, and scaling" How about you teach scaling, but only when someone actually needs it. Monetization, I'd hope, is taught in a business course, and virality could be covered with about a 10 minute powerpoint. Instead, I'd talk to them about how to leverage the data and social connections of their users, and how to use that to solve practical, every day problems.

While I agree that I hate those apps, the good ones do have a massive userbase. While I do intend to discourage stupid pointless apps, I won't explicitly ban them. Though 3 may be a little high, I may switch it to 2.

Keeping in mind that the course is meant to give a taste of something that wouldn't normally be taught in an average course (forgot to mention that), I think giving a taste of everything is a fair gameplan. Though talking about leveraging the "social graph" is a good idea.

Honestly, I don't think this is the type of course people should be taking in college. At most it should be a component of a web technologies class.

It's a relatively light course, worth 1/3 the units of a regular course. It's meant to give people a taste of something, and isn't a full-fledged course.

You should get in touch with BJ Fogg at Stanford. He taught a class recently that was similar.

http://www.readwriteweb.com/archives/what_stanford_learned_b...

Already spoke to him a few weeks ago, he gave some great advice.

I'd broaden the scope of the class to tackle the more general topic of scalable viral applications and how to develop, integrate, and monetize them.

1. Development. Introduce microframeworks designed for rapid development of small footprint apps. Tackle initial product design issues with viral applications.

2. Integration. Show the students how to (1) integrate their apps with Facebook and other services and (2) bridge their small apps to larger legacy apps and other web-based services.

3. Monetization. Here you can cover metrics, user feedback, virality, etc, etc.

I don’t want to sound negative here. I personally don’t think Facebook itself has any cool feature or technical challenging stuff for college student. I was a grad student 15 years ago, I don’t think then it would have been a kick for me.

I guess anything around the concept of Facebook and how it impact users experience or usability stats; in short anything related to the frenzy around Facebook will be much useful and educative to student.

To the question how much can you achieve with Facebook API? Very little technically

Some thoughts:

You should let the students run wild on at least one of the projects. Don't limit their creativity. While it would be great if Facebook users were interested in education themed apps, they are not. Just have them make apps based on things they wish they could do on Facebook. Crazy and stupid ideas can actually work, so mind as well not focus them in the wrong direction. The more times they get to try these ideas, the more they will learn what works and what does not.

Successful apps so far are all about connecting people in different ways and allowing self-expression, so that should be emphasized. Multi-player games are also getting big. From what I've seen, tools have not been so successful.

Since presumably not all the users will be adept at design, giving them a run through of various interface/design elements of Facebook's design would help in making better apps that users become comfortable with faster.

Facebook's platform has a number of different bugs and quirks (like its crazy implementation of Javascript) which merit discussion.

Don't forget to talk about other services like OpenSocial, even if it's a Facebook class.

One subject you could explore is how the metrics given by Facebook affects the perception of a "successful" app. The primary metric for applications is daily active users, while Facebook itself defines an "active" user as someone who logs in at least once a month - I always thought that distinction is interesting. The secondary metric are total installs. Things like repeat users, page views, pages/visit aren't as used. This has the strange effect that large quiz apps - which tend to rise and fall in a matter of about a month - are usually seen as more "successful" than applications with a steady returning user base.

I think success is easy to measure - the number of users for that app.

Regardless of how they do it, what it is, that is the 'success' of a facebook app.

I wouldn't put any rules like no poke apps, if they can create a super-duper-cool-elite-poke app that still takes off now, they are doing something right on the marketing/viral side.

Measuring success that way is really just turning the class into a seminar on viral marketing in facebook. If the class were for credit and one were the competitive sort, it would be "easy" enough to send out spam mails and to advertise the app as doing something amazing that it doesn't really do.

Rather than look for "success" metrics, I think the students might appreciate a qualitative assessment of their efforts from bkrausz.

Reply on news.ycombinator.com