Spam Filters

While the design of Webonizer makes it difficult for spam bots to send emails to your Webonizer account or post in Webonizer guestbooks, there is no way to accurately block all possible spam attacks. At the moment, Webonizer blocks automated systems-- but that doesn't mean an unscrupulous person might not physically submit spam to your mail, guestbooks or members.

To combat direct human spammers, you will have to start implementing Spam Filters. The spam filters will allow you to create rules for handling words and phrases.

When logged in, your control panel will have a link to Spam Filters or Anti-Spam Filters. Click this link to go to your spam filter page.

The way your spam filters work is like this: you can assign a weighted value to a word or phrase that determines how likely the word or phrase is to being spam--so the higher the weighted value, the more likely that word or phrase is spam. With each rule, you can decide where the rule is applicable and what action to take. For example, a rule applied to email will be applied whenever someone attempts to send an email to you through the site's email form; likewise, if guestbook is checked, the rule will be applied on guestbook posts.

Below is an explanation of the Spam Filter Form.

Apply To Flags:

Server-Wide -- this flag is only available to site administrators. If this flag is checked for a filter rule, that rule will be applied to every place on the site where spam checks are made, regardless of whose account the target recipient is for; this flag means that the rule applies in every member's account.
Emails -- this flag will apply to all emails to the owner of the rule. The rule will be calculated when a visitor sends the member an email through the site's email form.
Guestbooks -- this flag will force the rule to be applied in guestbook signatures for the member who made the rule.
IP Address -- this flag is currently not functional. When ready, the site will calculate the rule on all posts originating from the chosen IP address.
Domain -- this flag is currently not functional. When Webonizer has enabled this flag, it will block posts that originate from the specified domain.

Spam Text

This field determines the word or phrase that triggers the rule. For example, if the phrase is "spammerking" then the rule will apply to all posts that have "spammerking" in the content of the post--and where the situation matches. If the rule for "spammerking" applies only to Guestbooks, the rule will be triggered in guestbook posts but not in email posts. This text is NOT case-sensitive. The text does not use regular expressions and must match exactly--though matches within words will trigger the rule: if a guestbook post used the word "spammerkingsong" the rule above would match. Note that a rule will find any exact matches in substrings, including non-alphanumeric characters.

Weight Value

This must be a numeric entry from 1-100. The higher the weight, the more spam-like the word or phrase is. If a word is likely to appear in non-spam messages, you should give its weight a value of 1 or simply not use the word--find other phrases that will block spam.

Actions

Actions are the results of matches to this filter. There are three possible actions that will determine what to do on matches.

Spam Warning -- this is the only rule that will actually use the Weight Value explained above. The spam warning will force the spam checker to add this rule's spam value to the total spam threshold of a post. The way this works is this: every time a post matches a spam rule with Spam Warning checked, the spam checker adds the value of the weight to the a value that is compared against the recipient member's Anti-Spam Level (as defined in that member's account settings). If the total ever reaches the member's Anti-Spam Level, the post is labeled as spam and will not be processed. A spam notice is given to the visitor.
Spam Block -- this flag will trigger a spam block immediately. When triggered, the post fails and the post will fail. A spam notice is given to the visitor.
Keyword Block -- this flag will force the post to fail, but will not give a spam notice; instead, the visitor will see a notice saying that the language was inappropriate. Use this flag to block vulgarity.

Currently, a Webonizer site will automatically block a visitor from using any email/guestbook forms after getting three spam warnings. The block is temporary. Future upgrades to Webonizer will allow each site to control how many spam warnings will warrant a total email ban and the length of time such bans will persist.