I agree with everything you said. To address the lack of recipes I'm offering to import recipes from the blogs of food bloggers who register. Doing it by hand

I agree with everything you said. To address the lack of recipes I'm offering to import recipes from the blogs of food bloggers who register. Doing it by hand for now, want to automate it eventually. Automating is a more difficult problem than I thought at first, so it's on hold. I hadn't thought of your second point before. I don't think it would effect the search right now. The way it works right now is entirely text-based. It checks for an exact ingredient match (by name) and assigns a high relevance for each one. Then it looks for partial ingredient matches and assigns a much smaller relevance. Then it aggregates the relevances to order. A planned feature is to allow high reputation users to set ingredient relations. So they could mark two ingredients as substitutes for each other. I suppose I could have some ingredients be marked as common and automatically assign relevance for them as if they were included in the search. I've also been thinking about how to do a sort of general ingredient relationship system and how to utilize that for search. So far I'm ponder parent child - so ground beef would be a child of beef and ground meat. Beef would be a child of meat, etc. Then assign a relevance to ingredients that are found some number of steps out on that graph. The relationships could be set by high reputation users. As to the issue of quality, I'm hoping to source food blogs. A lot of food blogs have high quality recipes in my experience. I'll offer links with high SEO value and on site reputation and rewards in return for them posting their recipes. A further filter will be the on-site voting and reputation. I'll work on these for the next iterations. Thanks!