Advanced Search

Please click here to take a brief survey

Sarah Rich, 18 Apr 06

barcode-copy.jpg Ever stand in a grocery aisle contemplating the unknown (and potentially unappealing) story about where a product comes from and what it contains? Given the deceptive nature of advertising and packaging, the best way to get a straight answer about a product is to aggregate the opinions of its users. Enter GreenScanner, a public database of consumer opinions about the environmental accountability of over 600,000 products.

GreenScanner was developed by Bill Tomlinson, a researcher at UC Irvine's California Institute for Telecommunications and Information Technology. He's created a simple page, with a text box where you can enter a UPC code to find other people's ratings of a product. It's designed for use with network-enabled mobile devices, meaning that when you're standing in the grocery aisle fretting over whether to buy something, you can whip out your blackberry and find out on the spot. You can also add your own ratings and commentary after you've tried something.

Now that Tomlinson has set this system up, though, it's up to us to build up the arsenal of consumer reports. The service will become more effective the more we contribute our own opinions to it. Collaborative networked technology for choosing better groceries! We'll give that a Worldchanging stamp of approval.

Bookmark and Share


I have been incubating an idea for something very similar and what Bill has done is a great first step! The working title for my version of it is "vector pricing with affinity scoring". Basically, the idea is to use an index like the UPC code (as Bill does) to access a database of publicly-contributed ratings or "scores" on that product. For each product, there would be a vector of scores. A secretariat-like organization would act as a clearinghouse to assign a section of the vector to the organization contributing the scores. The requirements would be that 1) the scores fall in the range of 0-100, and 2) the organization must publish on its website (in some agree-upon standard format) information about what each of its scores means.

For example, the Forest Stewardship Council may want to contribute ratings on certain products about whether they were made with sustainably grown hardwoods. Greenpeace may want to rate various manufacturers based on their environmental records. Fairtrade Coffee may contribute information rating various coffees. Child labor groups may want me to know how Nike and other overseas manufacturers are doing. Etc., etc. The secretariat says, "OK, Greenpeace, you get the first 20 numbers in the database, FSC you get the next 10, Fairtrade you get the next 5." Actually, any and all comers would be given as many slots for their scores as they want. And they could come back later and add more. The secretariat would just act as traffic cop to say, you park yours there, and you park your over there.

The database then becomes a huge growing matrix where each row is a product, and the columns are the scores given by various groups. (Note that not all columns would necessarily need scores filled in for each product -- in fact, most wouldn't.) Conceptually, it looks like this:

UPC-code-1 - GP-score-1,1 GP-score-1,2 FSC-score-1,1 ... X-score-1,m
UPC-code-n - GP-score-n,1 GP-score-n,2 FSC-score-n,1 ... ... X-score-n,m

Now, say I want to buy a new piece of lawn furniture and I want to make a purchase in line with my values. I have a range of choices and I use an aggregator webtool to pull together all of the UPC codes for my choices and enter them into the user interface for the vector price database. The various scores for those products then get dumped back to me. I can either:

1) Hand-code my own weighting scheme to combine those scores into an aggregate rating. E.g., I can say I'm only interested in knowing what percentage child-labor was used to make the various lawn chairs, and I can additionally say I'm willing to pay a 10% price premium for those that rank low on that score. Whatever tool I'm using could give me the weighted scores, or red, yellow, green lights, or prepar a Consumer Reports -like table showing me all the various factors I have chosen so that I would then know which purchase would be most in line with my values; or

2) I could say, "I'm a Greenpeace member -- please use the standard weighting scheme developed by Greenpeace for these kind of products," and then I could get scores, lights, and/or tables as above; or

3) It shouldn't be too hard for the software to figure out from how the scoring database has been populated just what factors people think are important, and then to generate an appropriate weighted score, lights, and/or table.

The idea is that many, many groups want you to know about how they rate various companies and products, and this would provide a way to aggregate that information so that it is more readily available to you when you want to make a purchase. I'm still undecided about whether groups that we may not agree with should be allowed to contribute their scores, too. In the spirit of openness and fairness, I'm leaning towards "yes".

On the flip side, if people were willing to (anonymously) allow their purchases to be tracked, then you begin to build a dataset of people's purchasing preferences. You can then go to a manufacturer and say, there is a demand for a product that meets these criteria, and you would be able to put some quantity numbers on just how great that demand is. Manufacturers already pay an arm and a leg for product testing and advice on whether their products will succeed or fail. This could be a powerful tool for steering the economy towards producing products more in line with our values (if used properly).

Email me if you want to talk about the above and some of the thorny issues that might arise in implementing it.

-- Philip B. / Washington, DC

Posted by: Philip Bogdonoff on 19 Apr 06

This should give many companies a push, once it's possible for people to find out exactly how green their products are. Too bad for those who don't have such mobile devices, though. It would be a great idea for an online database of various products' green rating to be available.

Posted by: Lynn on 19 Apr 06

Reminder: the really green one is the one you don't buy at all.

Posted by: David Foley on 19 Apr 06

Actually, you could download a subset of the ratings to your PDA for whatever purchases you anticipate making. If you were going to the grocery store, you could download your typical purchase list. If your PDA/cellphone has a camera, it can be trained to scan the UPC barcodes. Then, if you encounter a new brand or flavor of macaroni, you could scan its code and see how it compares environmentally to the brand you usually buy. And perhaps you only would need to comparison shop in this way 1 or 2 or 4 times a year to update your usual shopping list with the preferred items.

-- Philip B. / Washington, DC / pbogdonoff at igc dot org

Posted by: Philip Bogdonoff on 19 Apr 06

Thanks for posting this Sarah. Dovetails with a lot of our work at Grass Commons, whichs works on all sorts of ways to get information about social and environmental impacts to consumers at the point of purchase.

One of our project designs is for the Handwich, which will let users with cell phone cameras take pictures of bar codes -- the image scanning is an excellent candidate for an open source project.

Currently we're focusing more on wiki components, but we've also got a lot of data networking design in place that would address many of Phil's interests. I was unaware of the GreenScanner project, though, so I'm grateful to WorldChanging for introducing me to yet another great project.

Posted by: Ethan McCutchen on 19 Apr 06



MESSAGE (optional):

Search Worldchanging

Worldchanging Newsletter Get good news for a change —
Click here to sign up!


Website Design by Eben Design | Logo Design by Egg Hosting | Hosted by Amazon AWS | Problems with the site? Send email to tech /at/
Architecture for Humanity - all rights reserved except where otherwise indicated.

Find_us_on_facebook_badge.gif twitter-logo.jpg