5 ms·
Surely we can come up with a better technical solution for this! Currently pharmaceutical companies try to prove authenticity by e.g. printing little holograms
by vilhelm_s 15y ago
Surely we can come up with a better technical solution for this!
Currently pharmaceutical companies try to prove authenticity by e.g. printing little holograms on the pillboxes, which is not very effective since pirates can print holograms too. How about this instead:
Each pillbox comes with a little paper which contains a random-looking string of characters and a trademark. It is folded-over and sealed so the contents are not readable. When an end-consumer wants to know if their drug is authentic or not, they tear open the paper, go to a website operated by (say) the US Patent and Trademark Office, and type in the trademark and character string (or scan a QR code). The website says "authentic"/"inauthentic"/"this string was already used".
On the production side, you can set it up so that the factory needs to communicate with a central company server to allocate codes, so you get an accurate count of how many codes are produced and if it matches the amount of drugs manufactured. This way, if an insider starts leaking codes to pirates it can be noticed quickly.
- VMG 15y agoThat would be a useful additional method. But still, selling pillboxes under the name of "Foo" if your company in fact isn't called "Foo" should constitute some kind of fraud in my opinion.
- rmc 15y agoYour talking about people who buy pills of the internet from spam emails. What makes you think "Remember to type in the letters on the government's site to check it" is something that would happen at all? Fake pills is not a technical problem, so a technical solution is not required. You're just shifting blame to the customer.
- vilhelm_s 15y agoWell, I wasn't primarily thinking about the internet pills. In many developing countries (particularly in south-east asia), something like 5-30% of all drugs are counterfeits without proper active ingredients (see e.g. http://www.bmj.com/content/324/7341/800.full http://www.bmj.com/content/324/7341/800.full). There are many skilled and motivated users (doctors) failing to detect the fakes because the packaging is identical. Similarly, the rock climbers mentioned earlier in the thread probably would also be motivated to check authenticity? I find the "a technical solution is not required" comment a bit odd. Spam is not a technical problem either, but we still use spam filters.