I've solved this problem ages ago. Just refer to people by their social security numbers and the problem solves itself. Seriously, though, the title is a really unnecessary exaggeration of the...
I've solved this problem ages ago. Just refer to people by their social security numbers and the problem solves itself.
Seriously, though, the title is a really unnecessary exaggeration of the problem. Just because there are some poorly made applications doesn't mean that computer science has no answer to the problem.
One of the things I do when I wear my database design hat is to add an auto-incrementing ID column to a table as a primary key for every table I design. Sure, it's wrongTM, but it solves so many problems in such an easy way. I find it hard to believe that any database of users - especially ones designed to span millions or billions of them - would ever use a person's name as a reference to an account? We already know there's going to be more than one John Smith, so much so that the name is literally used as a stand in for "a common name".
I think more than anything that this is just something that isn't put as a self-service option simply because it's not a priority for most people. Yes, name changes do happen, but they are fairly rare events, so it's not something developers consider to be a priority.
I really liked the top comment, though:
Cis men are the group generally least affected [by name changes] in our society right now, and this is one of those problems which would likely be handled much more smoothly if they weren't so overpowered in computer decision making.
You'd think. But a bug in a travel company's process accidentally charged my credit card to a trip booked by someone who had the same first and last name as me but different middle names. So...
You'd think. But a bug in a travel company's process accidentally charged my credit card to a trip booked by someone who had the same first and last name as me but different middle names. So people are out there doing it and for a time I had a $700 flight on my credit card to prove it.
I'd argue that if it solves problems, it's not wrong. As long as you're aware of the issues that this raises, and think about ways to mitigate them, then having serial, enumerated IDs is super...
database design hat is to add an auto-incrementing ID column to a table as a primary key for every table I design. Sure, it's wrongTM, but it solves so many problems in such an easy way.
I'd argue that if it solves problems, it's not wrong. As long as you're aware of the issues that this raises, and think about ways to mitigate them, then having serial, enumerated IDs is super useful (which is why people have been doing it for a long time). Alternately, having a UUID as your key is also a decent way to achieve what you want, which is a unique identifier that is unrelated to name, email, telephone, etc.; an identifier that you assign to an account that is immutable and is not something that the user has control over.
That's the key to solving this issue, which as you've pointed out is certainly solved, or you and I are geniuses that should collect some kind of prize money. Maybe we should write a paper.
A better title for this would have been "Changing your name is an easy, solved problem, but 90% of programmers suck."
Edit: i successfully prevented myself from tangentially listing why Serial IDs are wrong, and then arguing against that, but if anyone is interested, I haven't held forth on something all decade, and I'd be happy to put on my pedagogy hat.
What about women who change their last name when they get married? I realize that it is becoming more common to not do this, but based on a cursory google about 80% of women who marry still change...
Yes, name changes do happen, but they are fairly rare events
What about women who change their last name when they get married? I realize that it is becoming more common to not do this, but based on a cursory google about 80% of women who marry still change their name, and about 47% (let's round up to 50 because I'm pretty bad at math) will get married at some point. If women are about 50% of the population of the U.S., about 50% of women get married, and 80% of those women will change their name that's what, 20% of the US population who are going to change their name at some point in their lives? That doesn't seem all that rare.
It's entirely possible I'm misunderstanding that, though. x.D
I do think your quote of the top comment hits home. Most of the folks that are making decisions aren't ever going to need to deal with this.
Changing a name is a tedious process in any country. Which is kind of weird when you think about it, given that often times individuals are identified by some sort of numeric or alphanumeric ID...
Changing a name is a tedious process in any country. Which is kind of weird when you think about it, given that often times individuals are identified by some sort of numeric or alphanumeric ID like a SSN or ID#. You'd think that given this, information could flow from central repositories out to consumers of this information. Oh well.
I've solved this problem ages ago. Just refer to people by their social security numbers and the problem solves itself.
Seriously, though, the title is a really unnecessary exaggeration of the problem. Just because there are some poorly made applications doesn't mean that computer science has no answer to the problem.
One of the things I do when I wear my database design hat is to add an auto-incrementing ID column to a table as a primary key for every table I design. Sure, it's wrongTM, but it solves so many problems in such an easy way. I find it hard to believe that any database of users - especially ones designed to span millions or billions of them - would ever use a person's name as a reference to an account? We already know there's going to be more than one John Smith, so much so that the name is literally used as a stand in for "a common name".
I think more than anything that this is just something that isn't put as a self-service option simply because it's not a priority for most people. Yes, name changes do happen, but they are fairly rare events, so it's not something developers consider to be a priority.
I really liked the top comment, though:
You'd think. But a bug in a travel company's process accidentally charged my credit card to a trip booked by someone who had the same first and last name as me but different middle names. So people are out there doing it and for a time I had a $700 flight on my credit card to prove it.
Looks like that isn't unique to Finland @Akir. Who knew!
I'd argue that if it solves problems, it's not wrong. As long as you're aware of the issues that this raises, and think about ways to mitigate them, then having serial, enumerated IDs is super useful (which is why people have been doing it for a long time). Alternately, having a UUID as your key is also a decent way to achieve what you want, which is a unique identifier that is unrelated to name, email, telephone, etc.; an identifier that you assign to an account that is immutable and is not something that the user has control over.
That's the key to solving this issue, which as you've pointed out is certainly solved, or you and I are geniuses that should collect some kind of prize money. Maybe we should write a paper.
A better title for this would have been "Changing your name is an easy, solved problem, but 90% of programmers suck."
Edit: i successfully prevented myself from tangentially listing why Serial IDs are wrong, and then arguing against that, but if anyone is interested, I haven't held forth on something all decade, and I'd be happy to put on my pedagogy hat.
What about women who change their last name when they get married? I realize that it is becoming more common to not do this, but based on a cursory google about 80% of women who marry still change their name, and about 47% (let's round up to 50 because I'm pretty bad at math) will get married at some point. If women are about 50% of the population of the U.S., about 50% of women get married, and 80% of those women will change their name that's what, 20% of the US population who are going to change their name at some point in their lives? That doesn't seem all that rare.
It's entirely possible I'm misunderstanding that, though. x.D
I do think your quote of the top comment hits home. Most of the folks that are making decisions aren't ever going to need to deal with this.
Changing a name is a tedious process in any country. Which is kind of weird when you think about it, given that often times individuals are identified by some sort of numeric or alphanumeric ID like a SSN or ID#. You'd think that given this, information could flow from central repositories out to consumers of this information. Oh well.
I didn't even consider that, but you're right. I don't really see it much where I live, but I know it's a thing.
You had me wide-eyed for a few moments there! Ha!