We're sorry but this page doesn't work properly without JavaScript enabled. Please enable it to continue.
Feedback

Open Food Facts: The Wikipedia of food products to improve the food supply for all

00:00

Formal Metadata

Title
Open Food Facts: The Wikipedia of food products to improve the food supply for all
Alternative Title
What's in my food ? Open Food Facts, the Wikipedia of Food: Mixing mobile crowdsourcing, ai, opensource and opendata to improve food transparency
Title of Series
Number of Parts
490
Author
License
CC Attribution 2.0 Belgium:
You are free to use, adapt and copy, distribute and transmit the work or content in adapted or unchanged form for any legal purpose as long as the work is attributed to the author in the manner specified by the author or licensor.
Identifiers
Publisher
Release Date
Language

Content Metadata

Subject Area
Genre
Abstract
Open Food Facts is a collaborative and crowdsourced database of food products from the whole planet, licensed under the Open Database License (ODBL). It was launched in 2012, and today it is powered by 27000 contributors who have collected data and images for over 1 million products in 178 countries (and growing strong…) This is the opportunity to learn more about Open Food Facts, and the latest developments of the project. Scan a product using your phone, take a picture, and you're already part of the Open Food Facts revolution ! In this talk we'll show how Open Food Facts leverages open source technologies such as Perl, Python, TensorFlow, MongoDB, Java, Swift, React and Flutter as well as the great power of communities to open data of public interest for health & science, as well as unforeseen applications in your daily life. We will also introduce you to Open Beauty Facts, for freeing your cosmetic cupboard: shampoos, toothpastes, lipsticks, etc. How does it work? Using our Android or iPhone app, you can easily scan the barcode of products from your home or local store. You can either check them out (thanks to the decoding and comparison tools) or contribute pictures of their labels, assisted by our label-reading AI. The same can also be done from the website, where additional tools are available to fill in the product details from the labels, navigate or vizualise the database based in various ways, or access the APIs and raw data to make your own tools and analysis. Open Food Facts is developed and managed by a community of open source, open data and food enthusiasts and is organised as a non-profit association. All its creations are open: - the collected data is published as Open Data, - the software running the server(s) is open source and reusable (it was also used to create the Open Beauty Facts database), - the mobile applications are open source as well.
33
35
Thumbnail
23:38
52
Thumbnail
30:38
53
Thumbnail
16:18
65
71
Thumbnail
14:24
72
Thumbnail
18:02
75
Thumbnail
19:35
101
Thumbnail
12:59
106
123
Thumbnail
25:58
146
Thumbnail
47:36
157
Thumbnail
51:32
166
172
Thumbnail
22:49
182
Thumbnail
25:44
186
Thumbnail
40:18
190
195
225
Thumbnail
23:41
273
281
284
Thumbnail
09:08
285
289
Thumbnail
26:03
290
297
Thumbnail
19:29
328
Thumbnail
24:11
379
Thumbnail
20:10
385
Thumbnail
28:37
393
Thumbnail
09:10
430
438
Open sourceOpen setProduct (business)Beat (acoustics)Goodness of fitOpen setBitTable (information)Computer animation
InformationCustomer relationship managementFigurate numberAxiom of choiceComputer animation
Computer animation
CausalityWordLevel (video gaming)
Open setInformationProduct (business)Level (video gaming)Open setWordProcess (computing)Point (geometry)WebsiteComputer animation
Product (business)Open setOpen setWebsiteMobile appProduct (business)Group actionComputer animation
Open setProduct (business)Level (video gaming)Transformation (genetics)Mechanism designMobile appProduct (business)Computer animation
Product (business)Digital photographyComputer animation
Performance appraisalLevel (video gaming)AdditionDeciphermentNumberProduct (business)Computer animation
Product (business)WebsiteProduct (business)Mobile appData storage deviceAxiom of choiceDivisorComputer animation
Axiom of choiceDigital photographyObservational studyRight angle40 (number)Total S.A.Computer animation
DatabaseOpen setFreewareOpen setDatabaseUML
Open setColor managementService (economics)Mobile appWebsiteDatabaseProduct (business)Image registrationProjective planeArithmetic progressionMereologyTransformation (genetics)Point (geometry)Network topologyOpen setAdditionService (economics)DatabaseComputer animation
Term (mathematics)Multiplication signFreewareInstance (computer science)Graphics softwareMaizeOpen setComputer animation
Data storage deviceProduct (business)WordComputer animation
Mobile appOpen setProduct (business)InferenceComputer animation
Open sourceMobile WebOpen setMobile appAxiom of choiceDefault (computer science)Mobile appNeuroinformatikOpen sourceData storage device1 (number)Computer animation
Open sourceOpen setMobile WebMobile appMathematical analysisPairwise comparisonProduct (business)Axiom of choiceDefault (computer science)Similarity (geometry)AverageAxiom of choiceComputing platformDefault (computer science)System callInstance (computer science)Product (business)Core dumpSelf-organizationComputer animation
Latin squareOpen setMobile appComputer animation
Instance (computer science)QuicksortDigital photographyField (computer science)WordComputer animation
Field (computer science)Projective planeDigital photographyRippingCross-platformComputer animation
Android (robot)Computing platformAeroelasticityDatabaseOpen setGame theoryPlanningArtificial neural networkMobile appCross-platformoutputAeroelasticityProjective planeDatabaseFront and back endsAndroid (robot)RoboticsOpen setDebuggerInstance (computer science)NeuroinformatikComputer animation
Online helpFormal languageCategory of beingOnline helpComputer animation
Factory (trading post)Instance (computer science)Level (video gaming)Online helpAdditionComputer animation
Product (business)Interface (computing)WordHypermediaTranslation (relic)Product (business)Arithmetic progressionSoftware bugProper mapHacker (term)Open setGoogolInstance (computer science)WebsiteComputer animation
Programmer (hardware)WikiProduct (business)Coefficient of determinationComputer programmingWeightMultiplication signInformation privacyAxiom of choiceMereologyCodeProjective planeGoogol
Student's t-testCodeGoogolCodeProjective planeTouch typingSelf-organizationCompact spaceComputer animation
TwitterSelf-organizationComputer animation
Data modelBitMobile appWell-formed formulaOpen setGradientPoint (geometry)Instance (computer science)Goodness of fitOpen sourceMenu (computing)CurvatureFlow separationCausalityCore dumpDemosceneFile formatRight angleLevel (video gaming)Multiplication signData structureMedical imagingProduct (business)Computer animation
FacebookPoint cloudOpen source
Transcript: English(auto-generated)
Good morning, everyone, and thank you all for being here. It's great to talk at Fosdam. So, I'm going to present to you today Open Food Fact, which is a bit like Wikipedia of food.
So, Open Food Fact is all about this problem that many of us have in the supermarket about choosing which pack should I pick for my children. So, should I flip over each and every pack and read the nutrition table? Personally, I've never managed to understand it.
It's a lot of figures, so you have to do that for each and every single pack if you want to make an informed choice on food. Also, as you might have heard, the impact of food on public health is massive. It's the third cause of issues with humans,
and it has many preventable causes, many levers that you can pull to improve the situation.
And we wondered if citizens have managed to create with Wikipedia the best encyclopedia in the world. If citizens have managed to create the most accurate map in the world, why can't we do it with food products? And that's what Wikipedia, what Open Food Fact is set out to do.
So, we have gathered 1.1 million products from all over the world. So, that's the website of Open Food Fact. And the intent is to decipher labels, which you can also do with the Open Food Fact app. The mechanism is simple.
You just take a product, you look for the barcode, and you scan it with the app. And you will get valuable information, such as the nutrient score, about nutritional quality, the NOVA group, which informs you about the level of transformation of the product,
and finally, in France, the carbon impact, which tells you if a product is not only good for you, but also good for the planet. And it's collaborative, much like Wikipedia. If a product is not there yet, you just take three photos, front, ingredients, and nutrition,
and in a minute, you can get those scores for any product in the world. We also decipher those e-numbers that you can see on the products, those additives, using EFSA data scientific evaluations. So, you get, depending on your age, the level of overexposure for each and every single additive.
You can also compare products. When you are in a store and you don't know which zero package you want to pick, you can basically weave the app, compare many products side to side, and make the better choice based on the rational factors.
You can do that, of course. This is the photo that we saw earlier. This photo has data. So, you can basically look at all the breakfast cereals and pick the right one. This is the one I haven't been eating, and this was the one I've been eating as a child.
Smacks, you have to imagine, but half of the size is pure sugar. Forty-three percent of it. Of course, it's data of public interest, so it has to be open. So, our data is under the Open Database License, which requires you to be free for all and all use, to attribute, and finally, to share alike.
And we've managed to have quite a lot of impact. We're a French project, but we know the community is international, mostly European. So, we have one million products that have been contributed by volunteers like you, me, and producers alike.
We have 182 countries where we have products. If you're going to North Korea, we still have no North Korean products in the database.
We have 1.5 million, so far, users benefiting from the NutriScore and the Nova. And as it's open, as we have an open API and open data export, there are more than 200 apps and services using the data to do various things. What's more interesting, even to us, is the scientific papers that use OpenFoodFact to make science progress.
So, whether they want to study additives or the impact of transformation of food, that's possible using open data. And finally, we have 25,000 people who have registered. I mean, we don't ask for your personal data. Registration is optional.
But 25,000 people who have decided to sign up and to be an active part of food transparency. And we can do many, many things with open data. For instance, this is Fanta in Italy. It has 12% fruit. This is Fanta in Serbia. It has 3% fruit.
This is Fanta in Portugal. It has 8% fruit plus sugar or high fructose corn syrup. And in Reunion, which is a French island, it has zero percent fruit. So, that's the kind of thing you can do when you have open data about your food.
That's another thing. In terms of impact, you may have heard about the NutriScore. That's a picture that one of the contributors did using an image editor.
And he put the NutriScore on each and every single product. Fast forward a couple of years later, that's what we are seeing in French stores. It's not just applications anymore. It's the real world.
So, you can see NutriScore that is still not compulsory in France. But producers think, no think, it's good marketing to display it on their product. So, this shows the kind of impact open data and apps can have. Our stores are still citizen crowdsourcing, of course.
So, using the app, you can basically create data. The third-party apps that also bring data. But no, producers are taking note and providing data as well. So, you might recognize some of those brands. Many of them are French, but they are beginning to have some international ones.
And we want to help manufacturers make healthy the default choice. Just like a computer, we want the default choice to be healthier. And that's why we are providing, thanks to the French prevention organism,
the National French Santé Public Organism, we are making a platform for them to find opportunities to improve their products. So, for instance, if they have a beef preparation, they can find opportunities to improve the NutriScore or to reduce fat.
Things that are possible to do for them, without altering too much the taste, and that can benefit everyone, even people who are not using OpenFoodFact. And we are trying to extend the impact beyond food. We have started to do cosmetics with OpenBeautyFact.
So, same, pretty simple. You scan a barcode. You have those latent ingredients. I don't know if you've tried reading your bottle of shampoo, but you can't basically understand what's on it. So, we tried to decipher all that to detect 17 problematic substances and show them to you in the OpenBeautyFact app.
So, great, that's very cool. How can I contribute? The thing is, as you've seen, you can have impact for all. People who are blind, for instance, can't see the labels. So, being able to read the labels without having in full autonomy is important. That's the kind of impact you can have by contributing to OpenFoodFact.
And we're not focused on France or Belgium. We want to have this impact on the whole world. So, basically, this is a photo in Vietnam. You see they have a large grocery aisle, but ultra-processed food is gaining field even with this fresh food culture.
So, you can have an impact on many people and in many countries. And we have many exciting projects that you can contribute to. So, we have the native apps for Android and iOS. So, if you're an Android or iOS developer, you can help with that. We have the new cross-platform Flutter app that will be easier to use
for an even broader impact that we're starting planning. We have the OpenFoodFact artificial intelligence called Robotov, which is able to read labels. So, if you're a Python or a TensorFlow addict, you can help with that.
We have the OpenFoodFact database and backend. So, if you code in Perl, JavaScript, Python, you're very welcome as well. And then, on top of that, gamification projects in JavaScript and design and frontend if you're more of a CSS person.
So, this is very quickly some kind of stuff that we do using artificial intelligence. For instance, we detect the interest score. And also, if you live in a country, in any country, actually, you can help. And even if you don't have any particular technical skills,
you can help improve the support for your countries. There are so many food labels, so many food categories, that we need to help translating them into your language and expanding the knowledge of OpenFoodFact. Similarly, you can help improving gradient detection in your country so that we can compute the NOVA and detect additives.
And you can help with our growing environmental effort to basically understand better if the food has an impact on the planet. So, this is, for instance, a map of all the European food factories that we made thanks to the data. So, many ways to contribute.
Scanning new products and adding them to OpenFoodFact. Translating OpenFoodFact into your language, there's a lot to translate. Spreading the word in your countries, among your friends, in the media. Designing, making the interfaces better, we have much room for progress there.
Hacking, of course. And fixing bugs that are in OpenFoodFact. And with this voluntary effort, we've been able to achieve many great things in the past year. For instance, we have a brand new API documentation.
Thanks also to Google's Season of Dog program. So, it's easier to basically manipulate the data and play with food. I also would like to thank NLNet for its support. We are going to build a privacy-first recommendation engine that will let you make better choices based on your preference
and the data won't leave your cell phone. So basically, you will be able to find the best as NLNet spreads. I mean, Nutella is obviously E. But you will be able to find better solutions on your device based, tailor-made to your, if you're a vegan, if you have allergies,
that will be able to take that into account. So, thank you NLNet. And finally, we are part of, this year, we are a candidate to Google Summer of Code. So, if you want to get involved either as a student or as a mentor,
you're very welcome to do so. We have, as you've seen, many exciting projects to work on. So, let's keep in touch. Either on the website, contact, we also have a chat room. And you can also donate because OpenFoodFact is a non-profit. It's a volunteer organization.
And finally, to conclude, come play with food. Thank you.
Thanks for the talk, first. I tried the app from Asteroid the other day and I was a bit surprised that it's only with scanning, so no textual input, is that right? Can you... All the things I could to contribute is with scanning images, so taking pictures instead of entering something manual-like,
for example, with OpenStreetMap, is that correct? So, you can actually... The first thing that the app will ask you is take picture because that's the source of the data and if you don't have time, others can finish the work for you. But if you click on the edit button, you can enter structured data very much like OpenStreetMap, for instance.
That's a very good question and I'm glad you asked.
We're not scientists, so we... Repeat the question. Oh, sorry. So, the question is how do we grade, for instance, NutriScore, Nova, how do we grade products? So, we are not scientists, first point. So, we rely on scientific work. So, NutriScore is an open formula by Professor Serge
on nutritional... So, it takes into account fat, salt, sugar, and for the Nova, it's a scientific work by Professor Montero. So, the data is open. We ask the methodology to be open as well.
If any of you have more questions, I'll be just next door.