What is an address mention ?

287 viewsaddress

What is an address mention in terms of data structure? What should it contain? Should it contain only the information content that is in the source or more? How should it cite its source? How precise should the parsing go? Should it enable to display the address on a map? Should it contains already manually or automatically corrected version for each for the field?

Answered question
Add a Comment

Bold / audacious statement: you perhaps ask the wrong question.

An “address mention” quite obvious is some contextualized reference to an address. It is the actual address as an object of the mention / reference that contains more information than the literal text of some reference. This is at least a bit pedantic, because you most likely mean the ontological definition of an address as an abstract concept. The reason I emphasize the meaning of “mention” in a technical sence as the instantiated object of the class Address, is because I think that is how most people would respond to the word “mention”. An alternative word in this context to think about is “attestation”. Attestations as instances of something a bit more abstract (such as the actual address as a concept) are important.

I thus would like to rephrase the question as: “What is an address?”, leaving out the word “mention”. But that is besides the point you rise.

Addresses are complicated. They always consist of at least a number (integer) within some broader area being defined. They often have some form of suffix, like “bis”. Addresses do not always refer to a street. That is relatively modern. Earlier, the address-numbers were unique per city, and later on per neighbourhood or borough (or both at the same time). The concept of street already existed then, but was separate from what was seen as an address. The same street might have run through multiple boroughs, and thus could potentially have multiple occurrences of the same number. So the definition will need to take this into account.

You can also dispute the broadness of the term. This pertains especially to the original goal with which they were used. One could pose we limit addresses to the kind of stuff people sent letters to, and used in speech to signify where they lived. In the early nineteenth century (1811 in The Netherlands) a house number was formally established as required by the central government. Those most definitely are addresses. But there are a lot of cases where establishing that, is a lot less clear. For example, numbers for taxation-purposes from a single city that were required to be painted on a certain area above or next to the front door. You could say those don’t meet the requirement of being a common way to signify where you lived. I have come by examples of a clerk announcing some new system in a city, by immediately, in a newspaper, mentioning his own address, with “you can find me at …”. Although minor, this thus signifies those addresses to be used with the specific goal in mind of finding a house.

You may have noticed this is a pet peeve of mine. I also explicitly use the phrasing “cadastral address” to signify the formal appointing of a parcel with a parcel-number (with or without a suffix). I fail to see the distinction with most definitions of address, and thus would pose these are addresses too. In formal documents these cadastral addresses are (still) consequently used to address (pun intended) a specific plot of land, i.e. a parcel. For example, “Amsterdam, section A, parcel number 123 bis”. This is besides the point of formatting and parsing strings as addresses. You could translate this as “Amsterdam A123/bis” but also “Amsterdam-A-123-bis”. I would argue this is exactly the same address, but a different instantiation of the address as an abstract concept. The same applies for polygons or points on a map.

So, I propose a definition of an address as:

An address is an explicit reference to the (spatial) location of an object, consisting of at least, but not limited to a numeric part and some (optional for small collections) reference to 1 or more sub-divisions.

This is quite abstract, but does allow for all of the possibilities we have to deal with, and not excluding things like for example archival addresses, as indications of where to find some archival piece.

If you want to use a more specific definition for the common (day-to-day speech) case, you could for example choose to name this as a “house address”, and then declaring those to be references to the spatial location of building within the context of a city or municipality, most often pertaining to an explicitly declared street. Those addresses thus are a subset of the first, more abstract, kind of address.

Answered question
Add a Comment
Write your answer.