Ask HN: Why are Google's search products bad if Google has decades of Know-how?
9 by mrkramer | 2 comments on Hacker News.
I use Google's search products(Google Search, YouTube search etc.) on a daily basis and I see so many reasonable solutions that would significantly improve UX of Google Products but I don't understand that product designers and managers at Google are not seeing the same or similar solutions. Google has decades of know-how and unbounded human, financial and computational resources but still they can not figure out things like: My backlog: Google Search: 1. Crawl, index and rank audio content of the web and add it next to Images and Videos in Google Search e.g. "All Images Videos Audio Maps Books More" so I can find and discover audio on Google. 2. Add tags in Google Search just like you added tags when searching in Google Images so I can find websites in specific category or in similar category or that I can find similar type of web content. 3. Just like GitHub has percentage of languages of source code within source code's project page e.g. C++ 90% C 10% add content type percentage next to each web page or website result in Search Results e.g. www.mybiologyblog.com Text 70% Images 20% Videos 10% Audio 0% YouTube: 1. Standardize both web version and mobile version; in web version on channel's page I can search channel for videos on mobile I can not. 2. Add search option/bar in comment section of videos so I can search for information that is relevant and useful to me. 3. Add search option/bar to playlists so I can search for videos within my playlists. 4. Add Google Lens type search solution to YouTube so I can search "within" YouTube videos frame by frame; e.g. let's say I watch Wild Animals documentary and I want to skip to lions part, machine learning(AI) can analyse frame by frame and point me to that part of the video(Google has similar solution but only for creators called "Chapters"). And use speech recognition so I have transcript and timestamps for YouTube videos or in anther words try to transcribe and timestamp audio of YouTube videos. Frame by frame and transcription solution are complementary. I have other ideas for YouTube and Google Search in particular but this is my short list. For Google Search my focus is linguistic analysis of websites and analyzing websites' content(text, images, videos, audio, links etc.) I would not try to guess what users want like Google does but I would like to mine and analyze the web and then present it in Search Results for users to discover. Users will find good search results if your analysis and presentation of websites and their specifics/qualities is good in Search Results page. Also Google Chrome has some web search and web navigation problems that I would like to solve but that I can explain next time.