I have an all-core MediaWiki website; no addons, no forms besides a login form (not even a contact us form); the MediaWiki installation is continuously upgraded and everything is quite minimal and secured.

Say my domain is example.com and I bought an email address such as [email protected] from my domain registrar and I want to publish that email address as a text/link (certainly not as an image) in my website's "contact us" webpage.

What steps should I take, if at all, to protect from spammers in that regard?

  • You don't need to buy an e-mail, when you own a domain. Just set up the MX record.
    – Marcel
    Commented Apr 11, 2022 at 5:59

3 Answers 3


There is not a thing you can do from the website in these circumstances. Some basic modifications might help (only display it after solving a Captcha for instance), but it sounds like you don't want to do anything like that. If your goal is to have it available as plain text on a page that is linked to from the main site, without making further changes, then it can obviously be found by spammers.

Honestly though, it doesn't really matter. Spam is just a fact of life, and if your email address is something like [email protected], or [email protected], it will get spam anyway because that is an obvious email on top of a public domain registration.

Other options

So what to do? Just make sure you have a good spam blocker or use a provider that has strong anit-spam measures. Educate yourself (and everyone else you work with) about the dangers of phishing and downloading attachments from people you don't know. No matter what you do, you'll get spam anyway.


You can write the email in such a way that normal bots won't be easily able to decode it as being an e-mail.

For example <span>lserni</span><span>@</span><span>gmail.com</span>. You can also play several more complex CSS tricks to hide parts of the email from humans, or have them only appear in the browser.

  • Any modern spambot will be able to find that, the most advanced ones will even process javascript
    – Ángel
    Commented Feb 15, 2021 at 0:57
  • @Ángel well, of course. You can also get a HVS-optimized screenshot of the page and run OCR on it, so that images will also get harvested for addresses. But it's a race between performance and yield. I could place those spans in a different order in the HTML, maybe even mixed with another, invalid but plausible, email address, then straighten them out via CSS. This is a classic "five hunters and three lions" race; I feel that it's not yet necessary to resort to extreme measures (albeit they do exist) to achieve second place.
    – LSerni
    Commented Feb 15, 2021 at 11:33

you can also use the html code and unicode equivalents for each character

  • 1
    Hi, could you provide an example and proof that bots will be tricked by that?
    – Lou_is
    Commented Nov 12, 2021 at 6:50
  • link[link] Commented Nov 12, 2021 at 14:22
  • unfortunately no, I could give you an example in html but it wouldn't be real. and I don't know how to generate bots. link[link] Commented Nov 12, 2021 at 14:23

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .