Making an organization-independent implementation

[nickslevine]

I'm wondering how hard it would be to make an implementation of this that would not require initially setting IP ranges. Rather, the bot would look at every anonymous Wikipedia edit, try to link the IP address of the edit with an organization on the fly using reverse DNS, and, if successful, tweet a link to the edit with the name of the organization just found.

[webmaven]

A general purpose de-anonymizer would be dominated by consumer ISPs. The list of IP ranges you would therefore have to exclude for it to be useful would be huge.

[joellimberg]

As a sidenote — for a smaller Wikipedia the number might not be so huge. For example, the Estonian Wikipedia only has about 100 anonymous edits per day.

[remagio]

• WHOIS have specific policies, same reason why conf are no more stored on repo. And you have to query the DB of the ORGANIZATION's records of all WHOIS services. Not only network or domain, and it's regulated by additional policies to limit scams (every single query are monitored and stored to avoid misuse).
• About automated queries: How can anon parse the ORG whois's results to understand when it's your target on a specific language and country?
  I tried anyway thinking it could help new countries. Not mine, but it's a mess. At the end it require human reviews. Something too much complex for a deployer of a very simple apps like it's anon.

[joellimberg]

(I was thinking about doing reverse DNS queries for IP addresses and getting some sort of ownership info / domain names from there.)