I find that asking yourself “what am I going to use this for” is critical piece of using names in software. If you have more than one use, you probably want a field per use, with some sort of “default” so if someone does have a name they use for everything, they don’t need to work hard.
Example of uses:
The name that shows you that you’re logged in (e.g., in the top-right corner of a Mac screen or equivalent in web applications) – just let the user pick whatever.
The name that social-media apps show to other users – you might need to support features like “report” for this name, so if someone chooses a name like “kill all $ETHNIC_GROUP”, people can use the same report facility as they do for content.
The name that your custom support reps will use to greet them. This might need to allow a “Mr./Mrs./Ms./Miss/Mx.” field (probably open text for custom honorifics like “Lord”).
The name that you send the payment provider. Check the relevant API for the payment provider, and allow anything the API allows, and nothing that the API doesn’t allow. If the user has a problem, let them take it up with Visa or AMEX.
The name that you put on postal notices. It’s reasonable to have a max-limit on the characters here if you need to fit it into a physically constrained space like an address sticker.
The name under which they are legally authorized to work in $COUNTRY. They need to put exactly the characters on the work-authorization document that they plan to use to show eligibility for employment. It makes sense to show them the list of possible documents, and tell them to put the name exactly as it appears on that document.
Here’s the thing: until you do have a use case for a name, don’t ask for it. Once you do, your use case might mean “I’m sorry there’s no way to write your real name, but I need to put something on the snail mail, so put whatever thing would get this to you.”
I have spent so much time trying to explain the principles you outlined, and how they relate to gender markers and to trans people, to medical professionals over the past few years. (I do this primarily in my capacity as a person who makes noise and tries to push for improvement, rather than as a patient. Sometimes these conversations are made possible because of my employment, and sometimes because of personal connections.)
I wish it were as simple as just stating those principles and having them be applied. There’s just so much resistance to that. Nonetheless, you’re absolutely correct, and stating the principles clearly is a vital step.
This article is a a lot better than the regular “Falsehood” articles. Usually they give of a certain smug aura, implying the reader is supposed to know these things, and then proceeds to shower them with vague references or unclear implications. Actually explaining what you mean, what’s possible makes the issue much more approachable and understandable for someone who doesn’t have too much experience with the issues listed here (my name for example can’t be written in ASCII, but that’s basically just replacing a ‘ć’ with a ‘c’, which hasn’t ever been too much of an issue).
Another step forward would be if they could suggest, to a programmer what to do. Should all names just be Unicode Strings? Is no abstraction possible? Do no conceivable categories of names exist? Under what circumstances can it be said that sacrificing “absolute accuracy” for pragmatism/deadlines is justified?
Considering this, I don’t find it s uprising that the author says
Did programmers sit up, take notice and change their attitudes to names? Sadly, not really.
These articles always seem to miss the point completely. These names are put into databases for some reason, usually know what to call you on the website or in an email, or to collect data. They are not meant to faithfully reproduce every aspect of your cultures naming system. UPS doesn’t care if I use my ‘stage name’ or my ‘legal name’ or just make one up, they just want to know what bell to ring when they are in front of my house.
That might be true for the stage name/legal name thing, and for unimportant cases like your UPS example, but it’s extremely frustrating when I’m asked to input my name “exactly as it is on my debit card” or whatever payment processors say, and then don’t allow me to write my name because it “contains illegal characters”. Or when I’m applying for an ESTA from the US government, where they presumably care that my details are correct and correspond to what’s on my passport, but don’t allow me to input the correct name. Or when I’m not allowed to update my publicly visible name for a bunch of services without jumping through way too many hoops for each and every one of them (possibly trying to contact some administrator because there’s no system for changing names).
Not to be morbid, but two more familiar examples for #40, “People have names”, is abandoned babies and unidentified patients/bodies in hospitals. (Yes, in the latter, there’s the distinction that the person probably has a name that the hospital doesn’t know, but that doesn’t help when you’re the nurse with an intake form that the validation in a records app won’t let you save.)
Based on a conversation with @wink on IRC, I was prompted to research[1] this
In many countries, it is customary for a person being confirmed in some dioceses of Roman Catholic Church and in some Anglican dioceses to adopt a new name, generally the name of a biblical character or saint, thus securing an additional patron saint as protector and guide. This practice is not mentioned in the official liturgical book of the rite of confirmation and is not in use in Spanish and French-speaking lands, nor in Italy or the Philippines.
Even if we eliminate such curiosities, there are (a few) alphabets which are not yet covered by Unicode (perhaps the most realistic example is Aymara, a script for a language spoken by well over a million people in South America;
Huh? Aymara is written with the ordinary Latin alphabet, along with a few diacritics that are also found in the writing systems of much more widely spoken languages such as Spanish. Nothing that wouldn’t be comfortably within the range of writing systems specified by Unicode.
I find that asking yourself “what am I going to use this for” is critical piece of using names in software. If you have more than one use, you probably want a field per use, with some sort of “default” so if someone does have a name they use for everything, they don’t need to work hard.
Example of uses:
Here’s the thing: until you do have a use case for a name, don’t ask for it. Once you do, your use case might mean “I’m sorry there’s no way to write your real name, but I need to put something on the snail mail, so put whatever thing would get this to you.”
I have spent so much time trying to explain the principles you outlined, and how they relate to gender markers and to trans people, to medical professionals over the past few years. (I do this primarily in my capacity as a person who makes noise and tries to push for improvement, rather than as a patient. Sometimes these conversations are made possible because of my employment, and sometimes because of personal connections.)
I wish it were as simple as just stating those principles and having them be applied. There’s just so much resistance to that. Nonetheless, you’re absolutely correct, and stating the principles clearly is a vital step.
My example for 40: an baby who died before getting a name.
I have heard https://en.wikipedia.org/wiki/Ishi given as an example.
This article is a a lot better than the regular “Falsehood” articles. Usually they give of a certain smug aura, implying the reader is supposed to know these things, and then proceeds to shower them with vague references or unclear implications. Actually explaining what you mean, what’s possible makes the issue much more approachable and understandable for someone who doesn’t have too much experience with the issues listed here (my name for example can’t be written in ASCII, but that’s basically just replacing a ‘ć’ with a ‘c’, which hasn’t ever been too much of an issue).
Another step forward would be if they could suggest, to a programmer what to do. Should all names just be Unicode Strings? Is no abstraction possible? Do no conceivable categories of names exist? Under what circumstances can it be said that sacrificing “absolute accuracy” for pragmatism/deadlines is justified?
Considering this, I don’t find it s uprising that the author says
These articles always seem to miss the point completely. These names are put into databases for some reason, usually know what to call you on the website or in an email, or to collect data. They are not meant to faithfully reproduce every aspect of your cultures naming system. UPS doesn’t care if I use my ‘stage name’ or my ‘legal name’ or just make one up, they just want to know what bell to ring when they are in front of my house.
That might be true for the stage name/legal name thing, and for unimportant cases like your UPS example, but it’s extremely frustrating when I’m asked to input my name “exactly as it is on my debit card” or whatever payment processors say, and then don’t allow me to write my name because it “contains illegal characters”. Or when I’m applying for an ESTA from the US government, where they presumably care that my details are correct and correspond to what’s on my passport, but don’t allow me to input the correct name. Or when I’m not allowed to update my publicly visible name for a bunch of services without jumping through way too many hoops for each and every one of them (possibly trying to contact some administrator because there’s no system for changing names).
Not to be morbid, but two more familiar examples for #40, “People have names”, is abandoned babies and unidentified patients/bodies in hospitals. (Yes, in the latter, there’s the distinction that the person probably has a name that the hospital doesn’t know, but that doesn’t help when you’re the nurse with an intake form that the validation in a records app won’t let you save.)
All of my children are named
MAX(id)+1Not here in Germany. Seems pretty catholic to me.
So much for falsehoods :)
Based on a conversation with @wink on IRC, I was prompted to research[1] this
(https://en.wikipedia.org/wiki/Confirmation#Confirmation_name)
I’m not denying it actually happens, just that it may not be that widespread.
[1] “research” == google and hit the first Wikipedia entry
Also we got confirmation from two independent sources it’s practiced in Poland.
Huh? Aymara is written with the ordinary Latin alphabet, along with a few diacritics that are also found in the writing systems of much more widely spoken languages such as Spanish. Nothing that wouldn’t be comfortably within the range of writing systems specified by Unicode.