Infinitely Extensible Projects

When diving into new language, or radically different framework, it may be a good idea to have a bigger project where you can apply your newfound skills. In my experience, this is typically better than having a lot of smaller ones, because it minimizes the hassle of project’s initial setup. Therefore, it encourages you to experiment more.

To reap the largest benefits of this approach, the project of choice should exhibit two important properties:

• It must be easy to get started. Note that it doesn’t necessarily mean a complete programming newbie should be able to code the initial scaffolding in one session. But getting started must be easy for you specifically, so that you can dabble in interesting stuff almost right away.
  What it means exactly is dependent on your overall coding experience, and also on comparative difficulty of whatever you are trying to learn. For those taking their first steps in programming as a whole, extending their first Hello world program might be appropriate. However, if you are learning your fourth of fifth language you can aim for something a tad more ambitious.
• It should offer practically infinite possibilities of extension. The idea is for the project to grow along with skills and knowledge you acquire, enabling you to try ever more things.
  In normal development, this typically results in feature creep and is best avoided. But experimental, exploratory, educational coding does not really have to be concerned with such notions. Of course, if you can pack your learning experiences into a usable program then double kudos to you.

What types of projects fit into this characteristic? I’d say quite a lot of them.

When I was honing my Python skills, I started programming an IRC bot so that I could cram a few ideas into it rather quickly. They were implemented mostly as commands that users could input and have the bot perform some actions, like searching Wikipedia for any given term.
A similar pattern (collection of mostly independent commands) can be realized in many different scenarios. Aspiring web programmers could come up with something like a YubNub clone (bonus points if it allows users to add their own commands). Complete coding novices would probably have to resort to simple, menu-based programs in terminal instead.

Another option is to attack a problem which is a very broad and/or vague. Text editors, for example, fuel countless discussions (even wars) over what functionality should they contain and how it should be accessible in the UI. Chances are slim that your take on the problem sprouts a new Emacs or Vim, but a home-brewed editor is easy enough to start and obviously extensible, almost without limits. Additionally, editors can fit into pretty much any environment, from terminals to desktop UIs or HTML5 applications.

Some endeavors are a bit more specific, though. In web development, a CMS or blogging engine became something of a timeless classic now. Everyone has written one at some point, and there’s a lot of additional (thought not always useful) functionality that can be added to it. Getting the basics right is also a challenge here, especially from security standpoint.

For mobile app creators, the infamous To-Do list app is an idea exercised ad nauseam. But it’s actually a good playground for toying with various device capabilities (e.g. location-based reminders) or web services (like Google or iCloud calendar).

More?…

I’m pretty sure I’m far from exhausting the list of possibilities here. I cannot really speak for domains I have little-to-no experience with, for example embedded or hardware-oriented programming with equipment such as Arduino.

It should be possible to come up with infinitely extensible projects for almost every environment and platform, though. After all, every program has always one more feature to add ;)

Be Sociable, Share!

• 
• 

Be Sociable, Share!

• 
• 

• Tweet
•