Simulation doesn't persist

[selfthinker]

When you choose a simulation, it stops working as soon as you click on a link or open a new website. That makes the simulation less useful if you wanted to test a whole part of a website or a user journey (which should be the most common use case) and not just a single page.
The next step would be to make it work in the whole browser and not just in the tab where the simulation was activated. That could be controlled via a setting as it might not always be desired.

[andreasnylin]

Hello,

Thanks for your feedback!

This is something we played around with during development. The goal was that the simulation should run until the user clicked the cancel button or closed the browser. But we ran in to a problem: if the user closed the browser and then returned to the web site at a later point the simulation would still be running which might be confusing. We could not solve that the simulation should keep running on navigation to a new web page but be canceled if the browser was closed. That's why we decided to stop the simulation on navigation. We hope to solve this in a future version.

With "whole browser" do you mean all tabs? Or do you mean the browser addressbar, toolbar, etc?

[selfthinker]

I mean all tabs when I say "the whole browser".
Thanks for your explanation. If things persist after a browser restart that would best for us. I understand that it's probably an unwanted behaviour for most people, though. Which is why I think it would be good to add it behind a setting (switched off by default).

I can explain a bit more about my use case...

I have created 7 different login profiles for our 7 accessibility personas so that just by logging in as a persona, you will get a) a simulation of their condition and b) use the tool(s) they would use. That way it's relatively easy and effective to experience the web as a specific persona with specific access needs.

I'm describing the setup in a new repo I'm currently putting together. It's not live yet, but will be in hopefully less than a week: https://github.com/alphagov/accessibility-personas

It would be great if we could use your "concentration" simulation for our Pawel persona. (We're using Funkify's Hyperactive Henny so far. Although we can keep on using that for our own setup, we cannot tell others to use it as the main purpose of the setup is that it's easy and free. I have written my own simulations for most of the other personas.)

[selfthinker]

FYI, I have now extracted two of your scripts to be used in our project as userscripts, e.g. https://github.com/alphagov/accessibility-personas/blob/master/source/javascripts/wobbly-mouse.user.js
Userscripts are persisting (used e.g. with Tampermonkey), so that works well for our use case.