Good idea.
I personally love Duolingo. But I fear that the course data, which is mostly user contributed, will eventually disappear if the company goes under...
I personally love Duolingo. But I fear that the course data, which is mostly user contributed, will eventually disappear if the company goes under...
If some of the user-generated content isn't copyrightable, or was contributed by users willing and able to share it with a FOSS project, could only that data be scraped, or would it be too difficult to identify?