What if this lecture was held when the semester just started? Would it be helpful for our first assignment? The answer is yes; at least our code would be more organized. But I feel this kind of skill can only be learnt through experience or failure. Once you know the pain, you start to avoid that. When we were reflecting on our fish project, we could see the consequence of not obeying some of the software engineering principles. We did not have a folder structure, so it became very messy when the number of files increased. For the coding part, we did not write comment as a habit, which may be hard for others (yourself as well) to read your code and debug. Another problem was the division of job and workload. Designers could be very busy at the start but had nothing to do near the end. For programmers, each person was working on a certain part i.e. database, facebook api. There was no overlapping of jobs and it became very hard to us to help each other. These were the main problems of our first assignment. I do not know how many of them could be avoided if we learn software engineering in the first hand. One of my friends said that you can never fully appreciate the software engineering unless you dirty your hand. I agree. It might be a good idea to teach us these things after we have some experience. Otherwise the lecture could be forgotten quickly.

So what do we learn? Abstraction, modularity… Apart form the basic rules in coding; we overlap each other’s job to minimize errors and debugging time. Before we start coding, we spent a LOT time in designing modules, database schema, logic flow. It is hard. Now, I can preview the pain of the coming CVWO project. =0 But once everybody is clear about the process, life becomes much easier. =P

For scrum, it is useful for us to estimate the process. But sometimes, it is hard for us to estimate the workload cause we do not know how much time it takes to learn. Different from normal projects that milestones are cleared quickly at the beginning, our group has to spend much effort in learning. So it is slow at first and speed up later.

These are the things we learnt from our experience and the lecture. Still very naïve, but we are trying our best to get smarter. =)

I have done a lot case studies since our group started brainstorming our final project. Several weeks before, I said that there were no any apps like Graffiti in xiaonei and it would be successful if some one translates the app to Chinese. However, I was wrong. There do exist several similar applications, most of which are more technically advanced than Graffiti. One of them can even function as an online version of adobe Photoshop. But nobody use these app. I tried to think hard about the reason. User interface? User interaction? Although they are not as good as Graffiti, none of them could explain the one digit of monthly active user. Later, I was convinced the problem is that developers did not study their user properly. Xiaonei has a very special group of user than facebook. Major xiaonei users are between 18-25 years old. They are either university students or workers. This group of people share some common characteristics: don’t like to think very hard when playing a game, want to have some reward of every mouth-click, eager to show their profile/personality. Understanding the user can help explain some phenomena unique to xiaonei. Some people use xiaonei as frequent as msn and check the news feed every several minutes. Some will go through all the applications every time they login to xiaonei. Some will check their rank among their friends in every application before they sleep and make sure they are among the top five. According to this logic, it is not hard to understand why Graffiti-like application could not succeed. Also, apps like Geo-challenge that require some sort of thinking might not be very popular.

One app that has many users in one platform does not necessary mean it will be welcomed in another platform. Why? Users are different. There are so many SNSs in the Internet. Even in China, several SNSs co-exist and they are all have a large user base. So the great idea of develop one and deploy to all may not always work. =(

Let’s review the GetHelp. For me, version 3 is much better than the first version. The page is neat, color is calm, the smiley face is cute and the form is straightforward. Also the tabs are reduced to three with the graphs replaced by pure text. Compare to the colorful older version, I definite like the new one. However, when I ask my friends to comment about the two versions, some girls said they prefer the first version because they think the newer one is too formal. Here, I am not going to discuss more about the user interface. I just want to point out that there is no way to build an app that suits everybody’s taste. What can be done is to identify the characteristic of the largest user group and make them happy. To some degree, writing an application is very much like lunching a new product. Before that, you have to do all the market research. However, on top of all, the most important thing is the functionality. If the application is really useful, users won’t be too particular about the interface. From this point, GetHelp has improved a lot from the previous version.

Some suggestions:

1. When I fill in the form, I want to choose a category the help message belongs to. After I select a category, the application will generate a list of suggested friends who are good at that area. I just need to select among them and send the message.

2. Add one more parameter to show users’ status:

I am willing to offer help.

I am so sad. I need you.

Sorry, I am too busy to help.

Urgent! Help! (red color)

Sometimes, the task doesn’t require any special skills but take some time to do. And this makes it easier for callers to target their users and get a response more quickly. It is also easy for helpers to filter out the most urgent call.

Some bugs: 1.when I offer help with empty message. The help is still sent successfully.

One last word: Sorry lah Huihui. This blog is finished across three weeks and the logic may be discontinuous. There might be some repeated ideas also. Ask me if you don’t know what I am talking about. =P