Single mistakes will happen always as it is crowdsorcing project and it is not possible to verify all prices before their adding to the DB and to ask waste too much time from users for every price without damage to project mission, because they won’t be ready to waste their time, if adding proccess will be hard. IMHO, for Open Prices collecting of huge amount of data (more prices from more users) is more important than trying to find and fix every little single mistake. So this basic reporting mechanism is good enough for them and it just should be implemented in the app.
By default we trust a contributor and believe most of added prices by this user will be correct.
In the same time, I believe it should be important to improve moderator tools for finding more systematic mistakes like vandalism, incorrect understanding of adding process by new contributors and similar, because these kinds hurt data quality much more than single mistakes made of human factor.
In future, after getting more price data, it should be possible to build tools for analyzing new added prices, marking them as suspicious and send them to moderators automatically in cases like:
- We have some product X with 200 prices from 10$ to 20$. And if suddenly DB gets a price where this X product costs 100$, it can be marked “suspicious” by the system and sended automatically to moderators, and only moderator can clear/remove “suspicious” tag.
- When we get a first price for some product without any price until now and know its category, we can try to compare this first price with prices of other products in the same category.
-
- Product X is deodorant. It doesn’t have any price.
- We get a price for this product for the first time. Let it be 150$
- We check prices for other products of deodorants “category”, find it is basically have prices between 5-40$ (just random values for illustrating).
- We recognize it is suspicious this price is 150$, mark it “suspicious” and send to additional moderation check.
- etc
While these ideas are interesting, I don’t think their implementation should be started right now, because OpenPrices DB is still small and perhaps has not enough data for building and using these tools effectively, which is crucial because team resources are limited and it is not a great idea to waste them for building ineffective tools. So I guess we should wait until price DB will grow and will collect more representative and massive data.