3 ms·

Counterpoint: plenty of software doesn't provide any third options, because representing the data a just a single bit is convenient and the idea that there poss

Counterpoint: plenty of software doesn't provide any third options, because representing the data a just a single bit is convenient and the idea that there possibly being more than two options never even crossed the developer's mind. When you do get more than two options it's because the developer heard that it might be a possibility and they wanted to account for it, even it's never used. I don't think you can gleam much information about the cultural context from what options are available in software forms.

tablespoon 4y ago
> When you do get more than two options it's because the developer heard that it might be a possibility and they wanted to account for it, even it's never used. I don't think you can gleam much information about the cultural context from what options are available in software forms. Not necessarily. There are programmer-ergonomics benefits to using a enums instead of booleans. Personally, I'd never represent anything except a literal true/false value as a boolean. That has nothing to do with how I conceive of the possible values; it's mainly because I like to give things names that I can read.
- fluoridation 4y ago
  You would still need to conceive ahead of time of the possible values that the enum may take.
  - defrost 4y ago
    Not at all - there's sufficient range in a 32 bit unsigned integer and any values used after the initial assigned values can have details in a lookup table modifiable as time passes. In actual practice, the Australian Government found it sufficient to expand gender on passports from [M]ale [F]emale to simply include [O]ther (2011) [1]. [1] https://www.theguardian.com/world/2011/sep/15/australian-passports-third-gender-option https://www.theguardian.com/world/2011/sep/15/australian-pas... [2] https://www.passports.gov.au/getting-passport-how-it-works/documents-you-need/sex-and-gender-diverse-passport-applicants https://www.passports.gov.au/getting-passport-how-it-works/d...
    - fluoridation 4y ago
      Yes, it's expandable later. But in your initial implementation you'll still need to decide on a set of possible values, yes?

tablespoon 4y ago

> When you do get more than two options it's because the developer heard that it might be a possibility and they wanted to account for it, even it's never used. I don't think you can gleam much information about the cultural context from what options are available in software forms. Not necessarily. There are programmer-ergonomics benefits to using a enums instead of booleans. Personally, I'd never represent anything except a literal true/false value as a boolean. That has nothing to do with how I conceive of the possible values; it's mainly because I like to give things names that I can read.

fluoridation 4y ago
You would still need to conceive ahead of time of the possible values that the enum may take.
- defrost 4y ago
  Not at all - there's sufficient range in a 32 bit unsigned integer and any values used after the initial assigned values can have details in a lookup table modifiable as time passes. In actual practice, the Australian Government found it sufficient to expand gender on passports from [M]ale [F]emale to simply include [O]ther (2011) [1]. [1] https://www.theguardian.com/world/2011/sep/15/australian-passports-third-gender-option https://www.theguardian.com/world/2011/sep/15/australian-pas... [2] https://www.passports.gov.au/getting-passport-how-it-works/documents-you-need/sex-and-gender-diverse-passport-applicants https://www.passports.gov.au/getting-passport-how-it-works/d...
  - fluoridation 4y ago
    Yes, it's expandable later. But in your initial implementation you'll still need to decide on a set of possible values, yes?