A free public nutritional database is a great idea -- needed and useful.
But _not_ one generated by LLMs; at least not LLMs in their current state.
Show HN: OpenNutrition – A free, public nutrition database
121–130 of 160 posts
Re: Show HN: OpenNutrition – A free, public nutrition database
#122Earlier quoted context omitted.
It doesn't matter how accurate the models are, it's not a "data set" (in the scientific sense), it's more of a conclusion set. Maybe the conclusions are spot on. Maybe not. I have no idea.
I envisioned many lines of inquiry from HN but the idea that a compressed TSV of nutritional data is not a "dataset" (definition: a collection of related sets of information that is composed of separate elements but can be manipulated as a unit by a computer) was unexpected.
There are many HN users who are opposed to LLM.
Some of them are fundamentalists, and no amount of reason will reach them (read the comments on the Ghibli-style images to get a sample), others are opposed for very self-interested reasons: "It is difficult to get a man to understand something when his income depends on his not understanding it"
Yesterday, I vibe coded a DNS server in python from scratch in half a day (!) and it works extremely well after spending a few minutes on manually improving a specific edge case for reverse DNS using AAAA records: dig -x requests use the exploded form in the ip6.arpa, while I think it's better for the AAAA entries to keep using the compressed form, and I wanted to generate the reverse algorithmically from AAAA and A records.
Just ignore them, as your approach is sound: I have experience creating, curating and improving datasets with LLMs.
Like vibe coding, it works very well if you know what you are doing: here, you just have to use statistics to leverage the non deterministic aspects of AI in your favor.
Good luck with your app!
Re: Show HN: OpenNutrition – A free, public nutrition database
#123Earlier quoted context omitted.
You can add micro nutrients to those foods, they just don't always have them. Or so I thought.
Hi, Pierre, Open Food Facts NGO co-founder. We have an issue to propose approximation of micro-nutrients from reputable database. Feel free to join the project and contribute your time/coding skills to help us solve this: https://github.com/openfoodfacts/openfoodfacts-server/issues...
Your other comment is too deep in the thread for me to reply, but just wanted to say I appreciate you checking out the project and commenting, and appreciate the many years of effort you've undertaken in this space. How OpenNutrition can work with OpenFoodFacts is something I have thought a lot about (I think MacroFactor set a great example) and it's certainly something I'll consider moving forward.
Re: Show HN: OpenNutrition – A free, public nutrition database
#124Real and open nutritional datasources exist: https://support.cronometer.com/hc/en-us/articles/36001823947...
OpenNutrition uses many of the same open datasources included there, including USDA SR, CNF, AUSNUT, etc. The other datasources are licensed and not open, and I do not use those so that I can deliver a free app with a more generous set of features.
Re: Show HN: OpenNutrition – A free, public nutrition database
#125Are you using Typesense for the search server? If so, are you using hybrid search with embeddings, or just regular search?
Re: Show HN: OpenNutrition – A free, public nutrition database
#126Took a minute to find things that are in the DB, but once I did this is pretty slick. If I can join the endless queue of feature requests, the ability to scale the portion size and update the nutrition facts would be great. I've been tracking my macros daily for about 18 months and for me to hit my protein and stay under my calories I do a lot of what I call speculative food tracking. Sitting on the couch, thinking a…
Thank you for your feedback.
> If I can join the endless queue of feature requests, the ability to scale the portion size and update the nutrition facts would be great
This is all supported in-app if you're in a country with the ability to download it and have iOS (for now). The web product is more of a demo and isn't intended to be used on a day-to-day basis to track your food consumption, but this is a totally reasonable request.
> Also IDK where AI is wrt automated scraping but I've had some success feeding recipes into AI and getting the nutrition facts out. The ability to plop a URL in and get a scraped recipe with a name and nutrition facts would be immense.
I am not doing this for a few reasons, but, you can just screenshot the image of the recipe and use the app to upload that as a meal or recipe and it should parse out the ingredients and portions for you.
Re: Show HN: OpenNutrition – A free, public nutrition database
#127> Final nutritional data is generated by providing a reasoning model with a large corpus of grounding data. The LLM is tasked with creating complete nutritional values, explicitly explaining the rationale behind each value it generates. Outputs undergo rigorous validation steps, including cross-checking with advanced auditing models such as OpenAI’s o1-pro, which has proven especially proficient at performing high-qu…
I agree so much with you. This is not a dataset. This is the vomit of an LLM making stuff up. Like...why couldn't you just collect the data that already exist?? Why do you need an LLM?
Adding an LLM to this just adds a unnecessary layer of complexity, for what benefit? For street cred?
Re: Show HN: OpenNutrition – A free, public nutrition database
#128Are you using Typesense for the search server? If so, are you using hybrid search with embeddings, or just regular search?
I wrote a Go-based search engine that indexes full words and prefixes for the foods and their alternate names because I was dissatisfied with how slow OpenSearch was given the very specific needs I had. I have Cloudflare in front to act as a cache.
Re: Show HN: OpenNutrition – A free, public nutrition database
#129There’s an important caveat to keep in mind when it comes to food databases, especially those relying on branded or restaurant items:
U.S. law does not require food manufacturers to disclose everything that goes into their products. Under the Code of Federal Regulations (21 CFR § 101.100), there are exemptions to ingredient labeling... An example: flavorings, spices, and incidental additives (like processing aids or anti-caking agents) are not always listed explicitly. Also: proprietary blends and "natural flavors" can legally conceal dozens of chemicals (some synthetic), which consumers have no way of identifying.
Micronutrient data is often estimated or missing from labels and restaurant menus, which limits the accuracy of even the best-intentioned databases. Studies show that the nutritional information provided by restaurants and brands is frequently incomplete or inaccurate, especially when it comes to sodium, sugar, and actual serving sizes. (Urban et al. "The Energy Content of Restaurant Foods Without Stated Calorie Information" ; Labuza et al., 2008 and others)
IMO Food databases are only as accurate as the source data allows. Until food labeling laws mandate full disclosure and third-party verification, apps like this can support health awareness. Still, they shouldn't be treated as precise medical or dietary guidance—particularly for people with allergies, sensitivities, or chronic health conditions that require strict tracking.
Re: Show HN: OpenNutrition – A free, public nutrition database
#130Earlier quoted context omitted.
It doesn't matter how accurate the models are, it's not a "data set" (in the scientific sense), it's more of a conclusion set. Maybe the conclusions are spot on. Maybe not. I have no idea.
I envisioned many lines of inquiry from HN but the idea that a compressed TSV of nutritional data is not a "dataset" (definition: a collection of related sets of information that is composed of separate elements but can be manipulated as a unit by a computer) was unexpected.
> a compressed TSV of nutritional data
What is the source of that nutritional data?