How does that apply to the suggested workaround of the user running the necessary content scraping on their own machine and pushing the results to Power?