It's been a while, but at one time, I was working on the AntiSpam version 2.0-dev.
I added the following features to 2.0-dev.

  1. Added service Mollom and BlogSpam.net in addition to current 3 services.
  2. Make it possible to select and use multiple services at the same time.
  3. Determine if it's spam or ham depending on the rule in case of multiple services are selected (majority rules, weight rules, all or nothing rules, etc...)

I stopped development during the beta testing of this 2.0-dev since I got too busy.
I would like to know if this would be a good feature to have or not. I would like to get some feedback from users and then decide what to do.

Also, I would like to hear if there's any idea and suggestions to make this module better.

Thanks,

PS: I have not forgotton to work on existing open issues and pending patches too. :-)

Comments

idflorin’s picture

1. Usefully feature. I used Mollom and it does the job done , I don't know about BlogSpam.net.
2. I see this as very usefully. I'm thinking for example to use Mollom as backup for akismet (when akismet timeouts)
3. Oh yee.

mdcockfi’s picture

Component: User interface » Miscellaneous

Do you have any plans to enable this for the Contact form?

pixture’s picture

Assigned: Unassigned » pixture

@mcockfi

AntiSpam module can add spam checking to any forms that uses standard form API. However, the questions is when and how it checks the content.

When> when the form data is validated or when the form data is submitted
How> what it's supposed to do based on the spam check result (possible results are SPAM, HAM or ERROR).

More detail specification is needed.

susgeek’s picture

I also would like to see the contact form added to the node options.

mdcockfi’s picture

I'm not sure what I can add in the way of specification although that could very well be due to my adolescent status in the Drupal learning curve. I am interested in having antispam protection on the core "Contact" module. Am I correct in assuming that you are suggesting I would need to modify the core module to enable this functionality?

pixture’s picture

@mcockfi,

No. I am not. I wanted to get some input from users first what they want in order to think about the implementation method of the feature. Of course, it's possible to add spam checking to the standard contact form (done by Core contact module) without modifying/patching core contact module. That's not a problem.

For example, let's say that the AntiSpam module has added a capability to check spam in the contact form content. Then, what it is supposed to do when the AntiSpam module find the content a spam?.... Should it throw away the data and pretends as if nothing happened? or should it go back to the contact form without sending the message? or should it send the message to the recipient anyway with some spam mark added? or?????? there are many ways to deal with the situations and I am not sure what is the best way.

In addition, this feature needs to be determined with the fact that spam check may fail (false positive/false negative). Therefore, you do not want any extreme measure (such as throwing away the message). If we design AntiSpam module to return to the contact form without sending message when the content is determined to be a spam, then a user would be stuck if the content is not spam. They can never send the message to the recipient.

Ideal way to implement this feature is the approach that Mollom module uses. If the content is suspicious, then it displays Captcha and ask users to do the captcha to determine if the message is sent by a actual human or automatic program. Unfortunately, this may be a way too much for the AntiSpam module since it does not have such secondary method build in.

Anyway, to add support for contact form spam checking, there are many things we need to think and decide, and I want to get people's opinions on this. What do you think it's supposed to do?

spacereactor’s picture

Hope by the time version 2 can port over to drupal 7. Today drupal release alpha 2, which is much improvement but still got some bugs need to fixed but should be getting close to beta release.

mdcockfi’s picture

I guess I had not considered the subtleties involved especially an over-aggressive approach to false positive. Obviously summarily dismissing a legitimate contact would not be a good thing. Unfortunately Captcha has become fallible as well. So I suppose anything suspicious should be sent on marked as suspected spam.

Everett Zufelt’s picture

Version: 6.x-1.x-dev » 7.x-1.x-dev
Assigned: pixture » Unassigned

I would love to hear ideas for a 7.x-2.x branch.

tike012’s picture

Idea: implement a testing feature that will allow developers to post comments and get the response they want (spam, ham, unsure, etc.) This could be implemented as either a setting (turn TEST mode on/off), or by making the module look for certain message strings (submit message containing "antispam-[ham]" for a HAM response, etc.).