My legal last name prevents me from completing my Twitch affiliate onboarding | Hacker News Reader

My legal last name prevents me from completing my Twitch affiliate onboarding

old.reddit.com

68 pointspharrington51 comments

Somebody should really put together a collection of best practices for—a sort of collection of advice for people with this sort of name.

They could call it Null pointers.

Ba dum tish. For real though, the complexities of dealing with human input are a good reason to never accept human input you aren't absolutely required to collect. Once you do, put it in a place that you will never have to touch it or process it again. There's no validation you can do on human inputs, so don't try beyond "are you sure?"

That was exceptional.

GP was just throwing it out there, I'm glad to see you caught it.

This often comes up, but I never understood why.

Is it a case of e.g.

    If "null" == null

With implicit conversion?

What language does this? I think perl might allow it by accident if you don't use strict/warnings pragmas, but it seems weird. Is it a PHPism perhaps?

> What language does this?

SQL injection, for one. For once, not PHP (thankfully).

But an SQL injection is an exploit. In SQL ‘Null‘ != Null

Right, but a ton of well-meaning but cure-is-worse-than-disease "Web Application Firewall" tools still block it

PHP understands this properly:

    var_dump( "null" == null );
    // bool(false)

I have seen before that null is encoded as "null" when an encoding doesn't support null itself, e.g. multipart/form-data. Of course, then the server will treat any "null" as null

This can happen in random circumstances. A long time ago Amazon did not allow me to ship to my home address. By trial and error I found out that if I changed a specific o in my address to something else, it suddenly worked. There are no special reserved or offensive words in the address, I've never found out why this was. It had been fixed since.

Also I own a domain name with a catch all email address. If I sign up for websites, I often use the company name in the email adress to track leaks or spam. They sometimes don't allow this.

company@example.com is too obvious, so I've gotten into using company_jei68@example.com, where jei68 is just me mashing on the keyboard and different every time. Very handy if you have to prove that a particular database got popped.

Have you ever been successful in proving though? I found out no one ever listens to me being right about their leak.

no but you can go on the Internet to let their other customers know when they don't respond.

I only recently switched off the catch-all email configuration. Until then, I did what you did. But over the years, I got an increasingly annoying amount of misdirected email (possibly legitimate business, marketing, and spam) sent to all sorts of usernames, and only landing in my inbox because of the catch-all.

I now just generate masked email addresses with Fastmail. Much less noise now.

Classic: https://www.kalzumeus.com/2010/06/17/falsehoods-programmers-...

> People’s names are all mapped in Unicode code points

Curious how the author recommends we avoid this one

Actually, there is a case of this in China where (previously) both Unicode and the Chinese standard (GBK) was unable to encode Ma Cheng's (马𩧢) name. If you have Android, you might even see this in action (since that Noto still does not encode this one).

Currently, the solution going forward is to restrict what characters are acceptable as names.

Safari on iOS 18.1 doesn’t render the second character properly either. (Unless it’s a 口 :)

There are lots of Chinese people whose last names use rare characters or variants that are not mapped in Unicode. Taiwan, which retains traditional characters and hasn't forced people to standardize as much as mainland China and Japan, is particularly notorious for this. There's also the whole Han unification debacle, where similar but not always identical characters used in Chinese/Japanese/Korean have been mushed together.

Support for some Indic scripts also remains quite patchy: https://modelviewculture.com/pieces/i-can-text-you-a-pile-of...

I'm guessing the solution is to forget Unicode sequences as an identifier, and assign a hashed integer account number instead. Treat Unicode as non-textual data format for human use only, like how account pictures are. I think many web systems for mainly CJK users do so.

Similar documentation from the W3 https://www.w3.org/International/questions/qa-personal-names

This list should contain the case of "the name is not a keyword in your programming language" which is what this thread is about.

It sort of does but there could be one more specific to this case:

> 31. I can safely assume that this dictionary of bad words contains no people’s names in it.

And I’m immediately reminded of that family I knew when I was a kid whose last name was Cummings. We also knew someone who went by their first and middle initials of BJ.

Programmers don't care about things like "the real person using this program". They just want to complete some crappy function, mark a task as "done", and go on to the next one. If it annoys or hurts regular people, no big deal.

This is like government policies that hurt a minority of people. The bureaucrats don't care if it hurts a few people. And the minority is small enough that nobody else cares. They can't threaten the status quo, so they will just keep on suffering, and now that's just "how things are".

I think this type of "they don't care" attitude on any type of engineering is too simplistic.

Are there engineers who don't care? Probably. Engineers solve the problem given to them within the specifications also given to them. Those ultimately can't come from another engineer but the person implementing the thing implements it to specification. If that specification doesn't include edge cases then it's literally not their job to implement them.

The author also lives in a town called Scunthorpe. Can't catch a break.

Particularly after moving from Penistone.

I hear the cost of living there is more reasonable than in Cockfosters.

For the curious, somewhat disappointingly, it is pronounced PEN-nus-stun (all closed short vowels)

There’re few people on LinkedIn named “Swapnil Dev”! Amazingly all work in IT! That’s swap, nil and dev in one’s name. Such a cool name!

Even the error message is a sign of bad error handling: Last name can't contain "Null".

Contain? Is this is really what the if-clause did test for?

I'm only being half-facetious, it's hackable. Use homoglyphs. For example: https://gist.github.com/StevenACoffman/a5f6f682d94e38ed80418...

Just say your name is NuII with two capital "i"s.

Also recently:

When your last name is Null, nothing works (wsj.com) 272 points by impish9208 16 days ago | hide | past | favorite | 295 comments https://news.ycombinator.com/item?id=43113997

This seems to come up pretty regularly; here's some random articles about it:

https://www.wired.com/2015/11/null/

https://www.bbc.com/future/article/20160325-the-names-that-b...

I blame the popularity of languages that encourage "stringly-typed" code; which means this problem is unfortunately unlikely to disappear anytime soon.

JavaScript or python? Js is notorious for having problems with equality and casting.

Using a name of "null" needs to be in all unit test suites apparently.

I worked with a guy called 'Com' which is a reserved word in Active Directory. It causes chaos so apparently his AD name is always modified to be less offensive to the Microsoft gods.

Is Prn and Nul also reserved? That goes back to DOS device names.

Often repeated, but people still need to see it:

Falsehoods programmers believe about names

https://www.kalzumeus.com/2010/06/17/falsehoods-programmers-...

https://xkcd.com/327/

Reply on news.ycombinator.com