3 ms·
Another nifty example of how search engines and digitization are improving research. How would anyone have ever discovered that one super-obscure Japanese trave
by gwern 9y ago
Another nifty example of how search engines and digitization are improving research. How would anyone have ever discovered that one super-obscure Japanese travel book happened to include one old black-white photo? This reminds me of another Japanese-related example of search-engine-driven historical research, the castaway Australian sailors: https://www.theguardian.com/australia-news/2017/may/28/australian-convict-pirates-in-japan-evidence-of-1830-voyage-unearthed https://www.theguardian.com/australia-news/2017/may/28/austr... Or more literarily, the recently-discovered Walt Whitman novel https://www.nytimes.com/2017/02/20/arts/in-a-walt-whitman-novel-lost-for-165-years-clues-to-leaves-of-grass.html https://www.nytimes.com/2017/02/20/arts/in-a-walt-whitman-no...
- badosu 9y agoIndeed. Moving to the opposite direction though I wonder when will we discover the first pseudo-historical artifacts generated by AI (speeches, art, photos, and even videos). This is the greatest weapon ever, the ministry of truth would never dream of something like this!
- badosu 9y agoNot on mobile anymore, some examples: - https://github.com/junyanz/CycleGAN https://github.com/junyanz/CycleGAN (drawings to photorealistic images and perhaps videos) - https://lyrebird.ai/demo https://lyrebird.ai/demo (inflamatory or incriminatory speeches) - https://deepart.io/ https://deepart.io/ (forgotten works of art)