This question was covered under the security site where it was determined that there is no security implications for allowing + signs in e-mail addresses, so I would like to ask from the UX point of view: why do sites insist on ruining the user experience by disabling part of a well-known and well-documented standard?

as for examples, here is a list

3 Answers 3


I've been apart of numerous projects, as a UX professional, where the project is led by the tech team, and one big fault in all of this, is that the tech leads believe they know everything about UX practices and start saying, "I do it this way, therefore..."

That's where it all begins...

I also had a meeting where a tech lead says,"we should do it this way because personally I think..."

The moment that phrase is said, without even finishing the sentence, we stop her and say, "listen, it's not about personal preference, it's about the data." Personal preference varies, but the data never lies.

Programmers need to start realizing that there is more to an application than it's bare minimum functionality. I come across this problem now where I have to fight against horrible UX experiences because either the programmers think it's a great idea or the higher ups believe it's a better idea.

What we need to do as UX professionals (which is what I'm doing now) is show them that it works so they would finally just listen to what we have to say by doing usability tests or actually having the research to prove it. Once you gain a reputation of being a good UX guy (even though you are, you just have to prove it), then they will finally understand that you know what you're talking about.

It's annoying, I know, but you have to go through it at a place that knows little to nothing about UX.

  • You would think that user complaints would lead to fixes in this scenario then.. Aug 18, 2014 at 17:42
  • Here's a problem though: they don't see the complaints directly on a web application. So it's really irrelevant to them. The only way they can "see the complaints" is through either a drop in sales (but that doesn't really prove anything in terms of the application) or when you finally do a usability test to show that what they did was just awful.
    – Majo0od
    Aug 18, 2014 at 17:42
  • While I always prefer data over personal preference, it should be noted that data can lie as much as personal preference can. (lies, damn lies and statistics...)
    – DA01
    Aug 18, 2014 at 18:50
  • Hah! It could, that's why you have to do it properly like everything else :)
    – Majo0od
    Aug 18, 2014 at 19:27
  • 1
    @DA01 I see what you mean. I mean that could also happen with usability testing where the moderator is skewing the data with biased wording. That's why it has to be done either with the UX professional with keeping an open mind, or commissioning third party testers
    – Majo0od
    Aug 18, 2014 at 20:20

Because an extremely large proportion of programmers out there have no idea that there IS a standard, let alone what that standard might contain. When they need to validate, they just make something up based on their (absurdly) limited experience and call it a day. Since such characters are relatively rare in e-mail addresses, they don't get called on their BS quickly enough to actually get the problem fixed before it goes live.

  • So then this is something that slipped past UAT? Aug 18, 2014 at 17:37
  • @user2813274 - Yes. Basically, the number of people who put punctuation marks in their e-mail addresses is pretty darn small, and therefore the number of developers who realize such a thing is legal is small. This begets testing (if it exists at all) that doesn't account for the condition. And here we are. I suspect this WILL get cleaned up in the future, as a variety of unicode code points become common in e-mail addresses, and these web sites get forced by relatively normal users to get their act together. Aug 18, 2014 at 17:44

The reason is typically because they either don't work with a UX team, or ignore the UX team.

Bad validation typically comes from inheriting bad validation and bad validation habits. They've been cutting-and-pasting default field validations for the past decade and just keep on doing the same with little thought put towards determining if the validation even makes sense.

In regards to email validation: http://davidcel.is/blog/2012/09/06/stop-validating-email-addresses-with-regex/

Not the answer you're looking for? Browse other questions tagged or ask your own question.