I’m designing an online experiment but I’ve found the builder to be very clunky to use. My ideal pipeline would be to code the experiment from scratch using PsychoJS and an IDE such as vscode. However, it seems as though it’s not possible to host psychoJS html files in a local server - this makes debugging really annoying, as I have to push my code to Pavlovia and run it within Pavlovia every time I make changes to the code.
Is there any way around this? Feeling really disappointed by how difficult it feels to use the powerful PsychoPy / PsychoJS framework without having to jump through all the hoops of using the builder.
To be honest I would have to say that whatever downsides you might see in PsychoPy Builder are more than worth the upsides, especially when you want to put the experiment online. I use a lot of code components and find that Builder helps add structure rather than complexity.
If you go down the direct PsychoJS coding route then you won’t be able to get much support on this forum or from the Open Science Tools team. If you prefer a coding approach then JSpsych could work for you. I don’t know anything about it myself, but there are at least a body of users who might be able to support you.
To be honest, I’m still on the fence about using the builder. One source of irritability is that when I define variables in python, e.g. n=10, the auto js converter keeps the syntax the same instead of defining with the ‘var’ keyword, even if it is the first time I have used these variables. I’m not sure if this is a bug but at the end of the day if I’m going to have to manually code an experiment twice over it seems more feasible for me to use js directly.
Another possibility is to sacrifice online functionality and just code using the Python libraries - is this a more common approach? Would there be support for this?