Text redering in experiments
Hello there,
I'm making an experiment with a letter probe task. The most convenient way to program it is by just placing a variable text input in the sketchpad, which seems to work fine. Also, the final output is an image rather than actual text (i.e., the text can't be selected when running the experiment), which is great. The question I have is whether this image will always be exactly the same. In other words, how is that text being rendered, and is that rendering the same across all operating systems/browsers?
The rendering of actual text on the web is quite complicated, with huge differences between operating systems and even browsers, so I just want to make sure that this is not the case for text within an online experiment. I hope I'm making sense here ;)
Cheers,
Jasper
Comments
I'm tagging @sebastiaan for help here. :)
Did you like my answer? Feel free to
Hi @JasperdeWaard and @lvanderlinden ,
The text for OSWeb is rendered by PixiJS. There are slight (but not huge) changes in appearance depending at least on the browser, and possibly also on the operating system; for example, I once made an experiment with a smiley/ frowney emoji as feedback, and this was rendered as a colorful emoji on Chrome and as a black-and-white emoji on Firefox.
If this is important, then I would simply try out a few common browsers to see whether the result is acceptably similar. If not, then you can resort to using actual images.
— Sebastiaan
Check out SigmundAI.eu for our OpenSesame AI assistant!
Hello @lvanderlinden @sebastiaan
I'll have a look at PixiJS.
My suspicion is that emoji's might look different (they are not part of the font one specifies, but are drawn from a kind of system-font), but text will look identical (i.e. same font = same rendering).
Anyway, thanks for your help!