oTree Architecture¶
What is oTree?¶
oTree is a platform/software package to run online/lab experiments.
Participants interact with the experimenters and with each other by using their browsers
oTree runs on/as webserver
It may run on your computer for development (local installation)
It runs on a physical server (Server setup)
It runs on cloud services (Heroku, oTree hub, AWS, DigitalOcean etc.)
It has two modes:
Manual programming (Dominantly Python)
Point-and-click interface (oTree Studio)
In this book, we teach and refer to the first mode.
Point-and-click interface is available here: oTree Studio (Free for small/medium-sized projects, paid subsription of the rest)
oTree and z-Tree comparision¶
This comparison is for those who already use z-Tree :cite:`fischbacher2007z`. If you don’t know what z-Tree is or have never used it, please feel free to skip it.
z-Tree
is click & run, oTree
needs a server¶

Fig. 1 z-Tree: Click and Run¶

Fig. 2 oTree: Server¶
If you use z-Tree, you know that, for a basic setup, all you need to do is to run ztree.exe
on your server machine and zleaf.exe
on your clients with few clicks and that’s it. oTree is a little complicated than that. It should be running on a server.
z-Tree
needs client software, oTree
doesn’t.¶
With
z-Tree
, clients require software to run on their computer. WithoTree
they connect to the experiment with a web browser (Firefox, Chromium, Google Chrome, etc.).