Experience in building Facebook Web applications within 3 days

Recently, a friend on Weibo recommended a Facebook creative application Crush Notifier (Love notification ). The basic rule is: 1. Submit the opposite sex you like (the other party will not receive it); 2. If the other party also lists you as your favorite object, the two of you will receive a notification at the same time. 3. Each user can only be free twice. This application is booming on Facebook like Breakup Notifier, a product previously developed by the app.

Next I have translated an article about Facebook web application development experience by adore. ly, an application that is very similar to this application. I hope it can help and inspire application developers.

------

Address: http://tilomitra.com/what-building-and-releasing-a-web-app-in-3-days-taught-me/

Author: Tilo Mitra

Jiang Yujie, http://blog.csdn.net/hfahe)

 

 

Recently, two of my friends and I have released adore. ly website, a Facebook-based web application, allows you to "love" others, but your identity will not be exposed unless they also "love" you ". This application has been launched for several weeks. Now I have a chance to think about and accumulate what I have done, what I have done right, and what I have done is not good enough. This article records these ideas and shares some experience with other students who are preparing to gain some experience in web applications.

Experience 1: Quick Start

The Facebook hackathon (hackathon refers to a developer who takes a few days or a week to focus on a specific purpose through collaboration) at the University of Washington to build most of our application features. After 24 hours of coding and brainstorming, we completed a rough version and presented it to users. In the next week, we began polishing our applications and spent more time on the front-end. We basically built adore. ly within 7 days.

Experience 2: Release and repair

It is easy to find the defects in our applications, and the temptation to delay the release date for a long time is always in front of us. However, releasing it immediately is the most correct thing we have ever done. After we know that users are accessing this site, we can fix defects and iterate faster.

A small warning: we launched it in February 7 and published it on Hacker News (a news channel under Y Combinator, A Digg product) and Reddit (also a Digg website) in February 8, has a good response. I realized later that it may be better to wait until Valentine's Day in February 14.

Experience 3: not only code-but also ideas

I think we are not doing as well as expected. We previously thought that receiving anonymous love information from friends on Facebook should not be a disgusting thing, because they are people you are familiar. We only guessed 60%. Boys support us, but some girls do not. As a result, these girls do not want to send people they love, or do not respond to the information they receive.

The key point I want to point out here is to understand users' psychology and whether your applications are within their acceptable range.

Experience 4: Understand your market

Having a market and a target user is one thing. This is something that every application must consider-Otherwise what is the purpose of building it? For example, the idea behind adore. ly is that the virus can be sexually transmitted in high school or in more conservative countries (such as India), but we cannot know exactly how to influence these users. Generally, most high school students do not read Hacker News, TechCrunch, or tech blogs. They are Facebook and Twitter, but people of this age are not in our circle of friends. I think luck may be more important next.

Experience 5: Have an all-powerful Group

We did a good job. We have two programmers: Victor and I, and Jin Fei, who are responsible for business and marketing. Victor is the owner of the background, and Python encoding is quite fast. I am mainly responsible for the entire front-end. It is useful for everyone to have a clear role, because we know exactly what we need to do. At the same time, we are small enough to make fast decisions.

 

 

Data about adore. ly: more than 50 pairs of adore have been matched successfully.