What's the mathematical term for numbers not meant to be read as whole numbers?

[Sampiro]

The title may not make sense, but some examples:


-If your street address is 3465 Mongoose Lane, you may give it to the pizza place as "three four six five" or "thirty four sixty five" Mongoose Lane, but you're almost certainly not going to say "three thousand four hundred sixty five Mongoose Lane". The number is sequential and it's not meaningless- your next door neighbors are probably going to live at 3463 and 3467 respectively (though they might live at 3455 and 3475 or something else, and the houses in front of you are probably 3464 or if there's a park or church that takes up the entire block it may be 3400)- but chances are there aren't really 3,400+ bulidings or lots on Mongoose Lane, and in fact the next street over could well be the 3500 block or might be the 3800 block due to some odd numbering (I've certainly seen odder).

Other examples: while your SSN isn't randomly assigned, neither is it numerical (i.e. if it's 987-65-4321 you're not going to say "nine hundred eighty seven dash sixty five dash four thousand three hundred twenty one". Same with phone numbers, credit-card numbers, bank routing numbers, confirmation number, etc.: the sequence may be orderly but it's not meant to be read as a continual number, and while street addresses are usually sequential and routing numbers/credit card numbers are assigned due to some order (every part of a SSN/routing number/credit card number/etc. has some significance based on geography and or issuing agency and or customer number, etc.) there's not really a base because it's not a whole number.

So, hope I made it clear what I'm asking. Is there an actual term for these numbers?

[Sampiro]

Ah, and a related question: in some classification systems and inventory systems you'll see something like

3/5

which isn't meant to be read as the fraction three-fifths but as "part 3 of 5". Is there a name for this (or is it the same as above)?

[cmyk]

I can see what you're asking in the OP, and I'll be curious if there is an actual general term for these kinds of numbers (besides getting particular, i.e. Address, SS Number, Zip code, etc.).

As for 3/5, it still remains a fraction, doesn't it? 3 of 5, or 3 parts of 5 units that make the whole. If anything, I think it was probably adapted from the mathematical as a shorthand.

Complete WAG, though.

[Sage Rat]

If I had to call it something, I'd probably go for either a "code" or an "ID" (or "UID")

You can also call it a "number", like a phone number or a social security number.

[naita]

I doubt there's a mathematical term for this, as it's a non-mathematical use of numbers. The everyday term is... number. From m-w:

4 a: a word, symbol, letter, or combination of symbols representing a number b: a numeral or combination of numerals or other symbols used to identify or designate <dialed the wrong number>

4a is the mathematical number, 4b is the non-mathematical

[milquetoast]

I believe the term you are looking for is Nominal Number.

I think same term can be applied individually to the 3 in the 3 /5 scenario, as you are labeling that package as "3". But the combined "3 / 5" is mathematical, as it involves counting and, as cmyk says, an implicit fraction.

[LSLGuy]

I disagree with the implicit fraction idea.

Imagine I have 5 packages or component parts labeled "1 of 5", "2 of 5", "3 of 5", "4 of 5", and "5 of 5". The first number of each label is an indicator of the item's place in a sequence, i.e. an ordinal number, while the second number in the label is an indicator of the overall sequence length, i.e. a quantitative number.

Now imagine the labels are re-written as "1/5", "2/5", "3/5", "4/5", and "5/5". In no sense is the last item somehow the whole (5/5=1 in conventional fractional notation), and in no sense is item 4/5 a larger part of the whole than item 1/5.

Arguing by analogy to units analysis in physics, you can't divide an ordinal by a quantitative & expect a useful result. Division is not well-defined across that type boundary.


Going back to the OP, the critical distinction is that numbers as you describe are really labels, arbitrary symbols which serve to identify particular instances of whatever.

Labels can be formed using any symbols understandable to the audience. Which is why addresses like 1234A Main St. can be used. We can accept a label ("1234A") which mixes digits and letters, even though you couldn't perform arithmetic on it. The mere fact that a label is composed of the symbols 1, 2, 3, 4, 5, 6, 7, 8, 9, & 0 does not make it a number in the conventional mathematical definition of the term "number".

As noted above, "Nominal number" is one semi-official term. I'd suggest other terms terms like "label", or as suggested above, "ID", "identifier", etc. are also good descriptors.

If you really wanted a word for digit-only labels within the taxonomy of mathematical numbers, I'd suggest "non-arithmetic", to indicate that although they are written as sequences of digits, they cannot participate in even the simplest common-sense mathematical operations.

Note that non-arithmetic numbers certainly can participate in other branches of math just fine, set theory in particular. E.g. we can speak intelligently about the union or intersection of a collection of street addresses even though we can't add or multiply them.

And a tremendous amount of software depends very heavily, through the SQL data storage model, on the idea of performing operations over sets of labels. And these labels are often constrained to be strings of digits.

[yabob]

Originally Posted by milquetoast

I believe the term you are looking for is Nominal Number.

I think same term can be applied individually to the 3 in the 3 /5 scenario, as you are labeling that package as "3". But the combined "3 / 5" is mathematical, as it involves counting and, as cmyk says, an implicit fraction.

That's an interesting term, but I don't agree with it for postal addresses, which generally retain order over some local area, though they might not be consecutive. The number conveys something more than ID - I expect to be able to use their ordering to help find a house.

[jasonh300]

I think the term you're looking for might be Serial Number. Social security numbers and addressed are issued in a series (even though they aren't necessarily in increments of 1).

[Shoeless]

As a software developer, I would refer to what the OP is asking about as a "numeric string", i.e. not a numeric value but a string of numeric characters.

[milquetoast]

Originally Posted by yabob

That's an interesting term, but I don't agree with it for postal addresses, which generally retain order over some local area, though they might not be consecutive. The number conveys something more than ID - I expect to be able to use their ordering to help find a house.

I think the term still applies to postal addresses. The numbers are used as labels, not mathematical representations. It just so happens that the labels are "alphabetized" for ease of finding the address.

Originally Posted by LSLGuy

I disagree with the implicit fraction idea...

On second thought, I think you are correct. The text string "3/5" does not represent a fraction, even implicitly, in this context.

[Chronos]

Now imagine the labels are re-written as "1/5", "2/5", "3/5", "4/5", and "5/5". In no sense is the last item somehow the whole (5/5=1 in conventional fractional notation), and in no sense is item 4/5 a larger part of the whole than item 1/5.

The individual boxes, no. But (assuming you recieve the items in order), when you get the one labeled 5/5, you have in fact gotten the whole shipment, and when you've gotten the one labeled 3/5, you've gotten a little over half of it.

[LSLGuy]

True, but that's a property (by definition really) of ordinal numbers when processed in order.

In the typical case, order is not guaranteed. When you receive box 4/5 first, followed by box 1/5, the "4" & "1" provide no useful information about where you are in the reception process. The fact you have counted 2 receptions of 5 expected, now that is useful information. But that does not depend on which labels you received.

[Sampiro]

Originally Posted by milquetoast

I believe the term you are looking for is Nominal Number.

That works perfectly for what I'm using, thanks! [size=1]I'm designing a classification system for genealogists and I know from experience with library users that most people will, understandably, default to reading 222 and 104 as "two hundred twenty two" and "one hundred four" (222 being greater than 104). With this numbering system it's important for them to understand that 222 is neither greater or lesser than 112 but is just a logical/orderly/easily readable way to specifically identify a particular person in a family tree. [/quote]

[t-bonham@scc.net]

Originally Posted by Shoeless

As a software developer, I would refer to what the OP is asking about as a "numeric string", i.e. not a numeric value but a string of numeric characters.

In COBOL, these are also referred to as 'DISPLAY NUMERICS' (as opposed to 'COMPUTATIONAL NUMERICS', which are numbers used in calculations).