Ontology: Add term "Population"

[0UmfHxcvx5J7JoaOhFSs5mncnisTJJ6q]

Description of the issue

Population, and subclasses thereof, are used by various energy models to e.g. drive demand. Also it's listed as missing in #192

Ideas of solution

Named missing classes (#192) so far:

• Population
• Working Population (synonym Labour Force)

There are many further subdivisions possible. The UN population division for example subdivides by sex and age group (in five-year cohorts). The OECD by different age groups (I think ten-year cohorts) and education level. Other @OpenEnergyPlatform/oeo-domain-expert-energy-modelling may have additional needs.

So do we need a scheme for subdividing populations analogous to sectors, with sectors conforming_to sector divisions (#460)?

I think Populations would go under independent continuant -> immaterial entity, just like organisations. But strictly speaking, populations are composed of people, so it would be independent continuant -> material entity -> object aggregate, which … feels weird. Will need input from @OpenEnergyPlatform/oeo-general-expert-formal-ontology here.

Workflow checklist

• [ ] I discussed the issue with someone else than me before working on a solution
• [x] I already read the latest version of the workflow for this repository
• [x] The goal of this ontology is clear to me

I am aware that

• [x] every entry in the ontology should have a definition
• [x] classes should arise from concepts rather than from words

[l-emele]

Let's first find a definition, the classification arises from the definition.

[0UmfHxcvx5J7JoaOhFSs5mncnisTJJ6q]

"A population is the number of people in a city or town, region, country or the world."
–https://en.wikipedia.org/wiki/Population

[l-emele]

To reuse terms we already have, what about: _A population is the number of people in a spatial region._

[jannahastings]

I don't think we should define a population as the _number_ of people because that would make, e.g. "25" a population. Could we define it as an _aggregate_ of people?

[l-emele]

Good idea, Janna! So: _A population is an aggregate of people in a spatial region._

The number of this population (e.g. 80 million in Germany) would then be a data property a specific instance of the population linked with a specific instance of a spatial region (Germany) right?

[0UmfHxcvx5J7JoaOhFSs5mncnisTJJ6q]

A population is an aggregate of people in a spatial region.

So independent continuant -> material entity -> object aggregate …

What about subdivisions by sex and age? Elaborate scheme or just a bunch of subclasses defined as needed, like

"A population (female) is a population of female people."

One subdivision of populations we are actually using is rural/urban.

[l-emele]

What about:

• _A female population is a population of female people._
• _A rural population is a population in rural areas._
• _A urban population is a population in urban areas._

[0UmfHxcvx5J7JoaOhFSs5mncnisTJJ6q]

Probably there's no need to define "female" and "male". How do you feel about "rural" and "urban"?
And what about intersection sets (male urban population age 31-40)? Do we add all combinations?

[l-emele]

If you need that much detail, I would see these more of instances of population than classes. You have population age classes of 31-40 but another model might have age classes of e.g. 18-60 instead.

@jannahastings : What do you think?

[0UmfHxcvx5J7JoaOhFSs5mncnisTJJ6q]

If you need that much detail

I don't. But other's might ;)

[l-emele]

Okay, at the moment, we should focus on what we really need and not on what someone in future might need.

[jannahastings]

Okay, at the moment, we should focus on what we really need and not on what someone in future might need.

I agree: let's focus on including the broader classes for now, and if we need specific sub-classes for annotation in the future, we can add them with a "just-in-time" approach as they are needed. These will be compositional in any case, so we can always add them dynamically as long as we can define or re-use from external sources the required properties such as age and urban vs. rural. (Incidentally, we should check ENVO for urban vs. rural, they may have definitions we could re-use)

[0UmfHxcvx5J7JoaOhFSs5mncnisTJJ6q]

Okay, at the moment, we should focus on what we really need and not on what someone in future might need.

My concern is whether we set up a system akin to what is happening with sectors (#30, #460, #461), where there could be different definitions of "divisions of population" (for lack of a better term), or if we expect to add a bunch of subclasses as needed. Planning ahead could save us a lot of trouble. But I don't have strong feelings either way.

[akleinau]

I also prefer to stay simple now and concentrate on whats needed.
When looking at the definition ("A population is an aggregate of people in a spatial region.") I would prefer to solve this together with the long planned society class ("an object aggregate of people", #301, #302).
Right now following the definitions a population would be a subclass of a society, which sounds a bit weird?
Society is used in definitions like _organisation role_: "An organization role is a role of a society that has a collective goal and a set of organization rules. "

[jannahastings]

If society is defined as an object aggregate of people, then it should be a synonym of population.

[han-f]

Just jumping in while this popped up in my inbox: Wouldn't the difference between population and society be that society includes specific rules (laws, codes of conduct, governance) on how the people that make up the population interact with one another and as a whole?

[jannahastings]

I tend to agree that there should be more to the definition of society than just the people in it. So we need to reconsider the definition of organisation role: is it society that is needed there, or population (aggregate of persons)?

[akleinau]

I don't think people have to have anything more in common than all being people to create an organisation.
So I propose to leave society out for now, make a comment in the concerned issues, and change organisation def from society to population?

[jannahastings]

That sounds good to me. And if "population" sounds (by implication rather than definition) too large given that organisations might be small, we could also consider explicitly saying "aggregate of people" instead (which would indeed be a true synonym for population).

[stap-m]

I like the population definition.

society is not included yet. Unlike population, society and organisation role are not restricted to or defined by a spatial region, from my point of view. So, I'd stick with the current def of oraganisation role.

[akleinau]

@stap-m in the pull request I now pushed to just change the organisation role def to "aggregate of people", would you agree with that?

[stap-m]

We use lower case for class names. Also the issue numer #479 should be referenced in the term tracker item. I'll correct that.