Open Source Analytics

By • Sep 19th, 2008 • Category: Projects

My latest project is probably the largest one I’ve taken on single-handedly, but there is quite a bit to gain once it’s complete.

I am currently working on an open system for providing real-time website statistics and analytics.

Yes, there are plenty of free alternatives readily available, such as Google Analytics, Amung.Us, and Woopra. There is even an open source alternative called Piwik, but not one provides the level of privacy and customization a few of my current projects require.

While this isn’t a project I can quickly throw together, I’ve already made a lot of progress that allows me to identify and track visitors real time with and without the need for an external script.

I have been studying the methods some of the other providers are use, and in many cases have already developed a more effective way of doing so. One thing I hate is a lot of unnecessary lines of code so things will be cleaned up quite a bit.

For example:

The tracking code used to identify a visitor in Piwik is called piwik_visitor1, which contains:


Google Analytics uses __utma, which might contain:


And mine is called __oa which contains:


I will share more on my progress, what initiated the need for the new system, and what there is to gain from it next week.

Tagged as: , ,

Email this author | All posts by

One Response »

  1. Hi, I’m the main developer of Piwik. If you have some interesting thoughts, and don’t want to rebuild an open source web analytics framework (we are already trying to do it :)) ; Piwik has been built with a high focus on having a modular architecture, so you can change the functionnality in plugins. What don’t you like exactly in how Piwik does things? Could we patch the current code instead of rebuilding a new tool?


Leave a Reply