Hi Chris + Community,

I was wondering whether it is possible to create a series of screenshots (or the like) of the templates for the technical abstract. The idea is to provide some material to reviewers etc. who do not want to run the software but just want to see the frontend (especially the measures and manipulations).

I am not sure whether this is more efficient than just taking screenshots (and it may depend on the number of conditions) but I thought that one could maybe write some command when programming bots or so..?

Has one of you guys tried and found a good solution?

Best
Hauke

Maybe there is a extension for Chrome or Firefox that automatically takes screenshots of each page. Then you could use that together with browser bots.

I would strongly discourage making "screenshots". They have too few DPI. Instead, print PDFs from your experiment and crop them (e.g., automatically using `pdfcrop`). Those should then be included in your appendix. This gives you vector graphics, and thus conforms to the highest quality standards that could be imposed by journals.

I can send you an example if you email me.

I am sure all of this could be automated, e.g., using Selenium in combination with oTree bots, as mentioned by Chris.

I generally agree with gr0ssmann that it is better to create PDFs of your experiment’s pages than screenshots. An advantage beyond resolution independence of the text elements is that text remains searchable.

One caveat is, however, that the page layout may change or that certain elements or background colors may go missing when creating a PDF, even if you activate “Print backgrounds.” This depends on the definitions of the CSS classes. (For instance, when I try to create a PDF of this oTree Forum page, the dark navigation bar at the top is omitted. This does not seem to apply to standard elements in oTree but happens if you, say, embed Tweets in your experiment.)

Moreover, one might argue that it is helpful to see as a reader exactly what participants saw. This is better reflected in a screenshot than in a PDF — at least in lab settings where you know the resolution of the displays that participants used and can take screenshots at that very resolution.