Quote:
Originally Posted by janusz
Can you elaborate?
|
This is a large and complex subject. However, here are some of the biggest challenges:
- "Accept all domains". What this refers to is the policy followed by many mail services (especially those for corporate users) to not reveal which emails addresses are valid, and which are invalid. This policy is followed so as not to make life easy for spammers.
- "Greylisting". This commonly used technique will always temporarily reject any email from a source it does not recognize. This can be circumvented by retrying checks later. However, it means any check must be submitted with the intention of receiving a reply on whether an email address is valid, perhaps, 30 minutes later. Obviously, the email checker must also have the ability to detect likely greylisting, and retry accordingly.
- "Bounce rate databases".Many free services rely on these. Avoid services that rely on this.They deliver high rates of false negatives.
- "Most of the largest consumer email services (such as those from Google, Yahoo and Microsoft) have their own anti spam measures that create a huge challenge for email checkers. Various techniques (that are service dependent) have been developed to deal with this. Even the best of them basically need to balance some level of false negatives against false positives.
To finish, some of the better services will dramatically reduce their reporting of false positives and negatives by giving an undetermined status when an answer can only be given with a low degree of confidence.