User:Sulfonilklorida/Signpost Opinion1

Optional: Give a short WP:LEAD-like introduction statement here.

In the early age of Internet, IP block is probably one of the most effective way to ban somebody from accessing some website's feature. Back then, the Internet users are still quite scarce. ISP is giving everyone static IP because there are plenty of them.

But now, as the number of global Internet users skyrocketed, the IP address stock is getting more scarce. ISP is resorting to dynamic IP and NAT to solve this problem. By using this technique, a single IP address could be used and reused by plenty of users at once.

The direct consequence is, we cant really use IP address as a mean to uniquely identify someone on the Internet anymore.

Please don't bite the newcomers
In the age of dynamic IP and NATs, IP address is actually poor way to identify someone on Internet. I think the important problem that need to be addressed right now is "how to reduce IP blocking false positive".

I've seen a lot of new users, particularly those who interested in joining Wikimedia Movement due to competition event held by local chapter, complaining on why they cant sign up. Their IP address (mostly dynamic IP address behind ISP's NAT) is being blocked due to past's abuse. They complained it directly to the local chapter official, and the official sometimes cant really solve the problem. "It's beyond our scope" they often said. We're this close to "recruit" some new Wikimedia contributors, but then we're kicking them away by this harsh treatment.

Fingerprinting as a solution
Instead of identifying user by their IP, i think we should look the other, better way. For example : browser fingerprinting. By using browser fingerprint as an identification method, we can solve two problem at once. First, no more collateral damage due to IP blocking. Second, the fingerprint could work as IP masking implementation, since storing and publicly displaying someone's IP could have some legal consequences.

Further research is needed on how to implement our own fingerprinting algorithm. We can always use 3rd party off-the-shelf algorithm, but perhaps we need to design our own algorithm that suits our own unique need. For example, we could improve the algorithm by using AI technique to identify user based on their wiki-editing behavior.