Reference
HowToGuides
Manuals
LabAlumni
DataAnalysis
Advice for...
Admin
Octave is a programming environment mostly compatible with MATLAB; thus, it is a “high-level”, interpreted language that lets you get started with programming without having to worry about a lot of the low-level details. An important difference is that, unlike MATLAB, it is free. Octave is used widely, for instance, in Andrew Ng's Coursera Machine Learning course. Anything you learn and implement in Octave will work in MATLAB (with a few rare exceptions) and MATLAB code, if it does not use advanced toolbox functionality, often works in Octave. Thus, Octave is a great environment to become productive quickly.
To get started, a tutorial roadmap for complete beginners is available here.
The MATLAB for Neuroscientists and Signal Processing Using MATLAB books also have tutorials; these are linked from the Data Analysis in Neuroscience homepage.
The full Octave manual (for reference) is here.