Search engine ideas
A search engine in Gemini sounds really fun. As of late November 20 21, there’s only around 200,000 pages across 1200 domains. This is small enough to be manageable, but large enough to be interesting to work on.
- GUS space already exists so there’s some thing to check your work against
- there’s only one active search engine and as the author jokes, it has a bus factor of one! 💀 It would be helpful to the community to have another one
- building a Gemini crawler is cool, especially one that works with the Gemini specific parts of robots.txt
- Gemini files are super easy to parse so the text extraction logic would be really simple as would the link parsing and following logic
- there’s a lot of fun things to think about from full tech search. Using reverse indexes, etc