What is the difference between a trained and untrained domain/shill bidding investigator? Are there certified standards? I assume trained means been taught a detection method, given proprietary resources/data, and then using experience, conducts an audit?
Untrained investigators do not know how to determine based off of the "Matched accounts" data if the person is a real match or using the same IP address for another reason (
examples listed here). If someone is not properly trained or experienced, then they can't detect, lookup, or investigate whether that was the case (Not the same person and may not even know each other) or it was a real match (Same person) at any dependable degree of accuracy.
Some have called the bidding on your own names at NJ thread a witch hunt. Would you refer to that as a community-based investigation, and if so, would you classify it as a witch hunt?
I'm not going to take the bait. Everyone can form their own opinion on these things.
What's to stop a community-based investigation from becoming a witch pursuit thing?
If you know how to stop a community investigation from turning into a witchhunt, do share.
why do you think community lead investigations occur?
There are many reasons but the only one we care about is when they want to help. We greatly appreciate that help when it comes to our service. This is an open community and it works best when we all look out for one another.
Community-based investigations are most valuable to NamePros when they detect things that we've missed in our marketplace and help alert us to them. The problem is when you give an untrained investigator a weak signal (Multiple accounts) that they misinterpret to be suspicious (
Which it is not in and of itself) and then they search for any small detail to confirm their bias that they're searching for (Lookup "confirmation bias"). That leads to issues where an outdated whois record was enough for an untrained investigator to think they had caught something and make an accusation based on it. Far more often, "matched account" would be plenty for people to start making incorrect accusations and causing far more harm than good because they don't know what they're doing and they don't know what the information actually means (It doesn't mean they're the same person, it just means they have used the same IP. There are many reasons for two people who don't know each other to have used the same IP).
In reality, it is highly unlikely that someone would create multiple Business Accounts to commit any sort of fraud. They would be giving us more information about themselves by doing so than if they just created free accounts. This is all basic stuff that a trained investigator understands, and while we could try to train the community on it, there will always be a large portion of users who will not read it and will jump to conclusions of their own from seeing "Matched account" and their conclusions will be wrong almost every time unfortunately.
Could it be that some are unclear if machine based detection systems are doing their job?
Machines aren't going to replace human eyes anytime soon. Nothing beats a community of people investigating. However, that doesn't mean they should be armed with information that they don't understand how to use. That wouldn't lead to anything helpful, most of the time.
Does this mean we get to have a discussion about best practices for shill detection, auction integrity, and community-based investigations?
Ya, but a new thread should be started for that so that the topic of the thread is clear and others that want to contribute can do so. Members aren't going to expect that to be the topic of this thread and that means it needs to be in another thread.
Thanks,