Miscellaneous Hacks - NoSpam! - an alternative to CAPTCHA images

This hack is for vBulletin 3.6.x. For 3.5.x, please use the one provided in this thread. If you are using 3.7.x, you must use the one in this thread.

This simple hack is meant as a replacement for the default CAPTCHA system in vBulletin. There are two main reasons one might want to do this: firstly, new technology is constantly being developed to crack CAPTCHA images and make spam accounts anyway, and secondly, the more secure the CAPTCHA, the more difficult it is for genuine users to tell what the numbers in the image are. There is also the issue of visually impaired users, and the fact that not all servers are capable of generating CAPTCHA images.

So what does this hack do instead? It asks a question. Any question you want. That's the best part: YOU make up the questions, which means that every forum is unique, which means that it is impossible for spambots to be simply programmed to bypass it at all forums with the hack installed. You can make one or many questions - if you make many, the hack will pick one at random when a guest attempts to register, search, post or send mail through the Contact Us form. Their input is compared with your specified answer, and voil?, if they get it right they're pretty much guaranteed to be human. If they get it wrong, they're given an error message.

It doesn't have to be a complicated question. Heck, you can just make it "Please type 'blah' into this box." Or you can go with inserting a simple image with HTML and ask what is shown on the image. Or you can ask them to tell you two plus two. It's up to you. In fact, this can also be used as a means of forcing people to read the rules by asking for a certain password found there, or if your forum focuses on a specialized subject, ask a question concerning the subject that all your genuine users will be able to answer, but a random troll or "CAPTCHA-sweatshop-reader" will be scratching their head over.

It's simple to install, too: just one product to import, and that's it.


VERSION HISTORY:

3.0: Added functionality to allow users to specify which pages should use NoSpam!.
2.0: Extended the NoSpam! functionality to guest posts, guest searches and guest "Contact Us" in addition to registration.
1.1: Made template edit automatic (with thanks to Cole2026), added ability to have more than one answer to each question, and made answers case insensitive. To upgrade, I recommend undoing the template edit so you won't have to worry about it anymore (replace $nospamfield in the template register with $imagereg, or if you added $nospamfield above $imagereg, just remove $nospamfield altogether), and then reimport the product through the Admin CP (making sure that Allow Overwrite is set to On).
1.0: Initial release.


INSTALLATION:

Please download NoSpam!.zip, not product-nospam.xml, for the tested version. product.nospam.xml is NoSpam! 4.0 adapted blindly to 3.6, i.e. by editing the XML file for the 3.7 version without actually testing it on a 3.6 board. Theoretically it should work, but I cannot guarantee it at this time. If you download NoSpam!.zip and unzip it, it will contain another file called product.nospam.xml, which you should use.

1. Import product-nospam.xml through the Admin CP product manager.
2. Go to your vBulletin options in the Admin CP and select NoSpam! Settings. Once there, turn the system on and input your questions and answers according to the instructions there.

The system should now be functional and running.


UNINSTALLATION:

Just uninstall the product through the Admin CP.


SUPPORT:

Full support will be given here in this thread. All suggestions are welcome.

Thank you and please click "Mark as Installed" if you like it.


If NoSpam! is not working for you or you are looking for something a little different, you might want to check out Advanced Textual Confirmation.

[schanulleke]

I still have version 2.0 installed and would like to upgrade to 3.0.
Do I need to uninstall the old version before installing the newer one? Or can I install 3.0 without uninstalling 2.0?

[Stifmeister2]

Originally Posted by schanulleke

I still have version 2.0 installed and would like to upgrade to 3.0.
Do I need to uninstall the old version before installing the newer one? Or can I install 3.0 without uninstalling 2.0?

Yes you can. But remember to allow overwrite when you import the 3.0 version from ACP.

[schanulleke]

Thanks!

[OCSupertones]

Going to give it a go right now...we're getting tons of spam at AfterHim.com

Thanks!

[mikewiemholt]

Much thanks for this mod - love it. Absolutely love it.

[RDX1]

Installed, nice to see you have integration for my hack. Just an FYI. I'll be updating it so it works with 3.6, you may need to look over the integration code.

[MissKalunji]

the images are not working for me i get a broken image with a bunch of %%%% to it

[DementedMindz]

what images there are none?

[derfelix]

that just gave me an idea...
wouldn't it be even more difficult for a (future) bot if the text of the question was written in a image?

the bot thinks its a normal image.. and... writes the question text instead of the answer..

so we know its a bot and not a stupid user.. and he can be automatically banned without risking to ban a normal user..

example:
question: HoW much i s two and 2?
written in verdana font on a transparent gif. A user will think it is normal text..
A bot will think its a captcha image... and will enter the text as written in the image ..
so if he writes it exactly (with spelling errors, typos and strange capitalization,
but not respecting the spaces in the sentence)

.. we know its a bot.. (even a "not so bright" user will at least add one space in the sentence)


Felix

[PoetJA-1975]

Originally Posted by derfelix

that just gave me an idea...
wouldn't it be even more difficult for a (future) bot if the text of the question was written in a image?

the bot thinks its a normal image.. and... writes the question text instead of the answer..

so we know its a bot and not a stupid user.. and he can be automatically banned without risking to ban a normal user..

example:
question: HoW much i s two and 2?
written in verdana font on a transparent gif. A user will think it is normal text..
A bot will think its a captcha image... and will enter the text as written in the image ..
so if he writes it exactly (with spelling errors, typos and strange capitalization,
but not respecting the spaces in the sentence)

.. we know its a bot.. (even a "not so bright" user will at least add one space in the sentence)


Felix

Wow! Sounds like an AWESOME idea - I wonder how difficult it would be to implement though? Anyway - perhaps in the next "update"

Jacquii.