Ask HN: Why is barcode data so hard to find?

5 points by richardw ↗ HN
I had a concept that seemed easy enough, and step 1 was identifying products from their barcode. I tried it on a couple items (Bose speaker, wifi router) in my house and the scanning apps couldn't identify them. These aren't once-off products made by a home inventor, or a bag of apples from a local producer. It's Bose.

Googling open barcode databases hasn't been useful. What's the blocker here? Do people get sued? Not enough motivation? Not enough momentum? Does the effort force people to charge for this?

There don't seem to be that many e.g. electronic products in the world, but obviously my intuition is broken on this. What's up with barcodes, HN?

10 comments

[ 1.8 ms ] story [ 34.6 ms ] thread
You’re searching for the wrong things.

https://en.wikipedia.org/wiki/Universal_Product_Code#Numberi... https://en.wikipedia.org/wiki/Global_Trade_Item_Number

IIRC, you have to purchase access to the GTIN database from GS1.

Thanks. I've seen the GTIN, but...no OpenStreetMap-like initiatives, for something so obviously useful? My question is just why nobody has persisted with this.
I’ve no idea what your idea is, which means that it’s obviousness is clear only to you. It also seems to me that GS1 has persisted with this since the seventies, so they have a pretty good track record.
I'd like to recognise the QR code on my Bose speaker, which comes back with a serial number that GS1 doesn't recognise. I scanned the barcode on my TP-Link mesh router and GS1 doesn't recognise that either, because both have alphanumeric values and it wants numeric. Maybe it wants a special option selected on the site, shrug.

I scanned a box of tissues and it came up with "Company with GLN 40 8850000 000 9 requested data privacy" and no item identification.

My intuition says these should be pretty easy to identify, but after trying a few apps and services I still haven't found one that just works.

If it has letters in it then it isn’t a UPC code. There’s no guarantee that it is even globally unique; it could be something made up entirely by Bose. Or by a retailer.
It’s probably Bose as it’s printed directly on the rubber at the bottom. No other barcodes of any kind. Oh well, thanks for taking the time to reply.
Barcodes and QR codes are just a way to display data. The assumption that every one of them identifies the object is invalid. Some do, for sure in particular in retail stores - others could be serial numbers or other such info that aren't going to match with any specific database. This is also why on many non-grocery items you'll see multiple barcodes... the one that the store uses for identification, and the rest that represent other things.
GP, correct me if I’m wrong, but I believe the idea is an open source GTIN dataset.

I have no answer, but am also curious now.

No, I just assumed they would exist. I’ve seen things like this:

https://product.okfn.org/ But the product open data link is dead. So there is both 1) a need and 2) a reason it’s not ongoing.

The idea is a lot smaller. Something I could hopefully put together in a month if the above existed. It’s not important enough to build out a big open DB.

If I'm not mistaken, Grocy has a nice database for household products. I remember trying that and scanning most of the stuff I had in the kitchen, and not only it found the products but also the price. I don't know their sources though.