Founded: 1996
Defunct: 1998 (2 years)
Seen at
https://musicbrainz.org/label/3e1fd80f-03c7-4ea1-9756-9086e8d9cb1e
I’m expecting to be reading Defunct: 1998 (3 years)
But https://www.ox.ac.uk/sites/files/oxford/media_wysiwyg/University%20of%20Oxford%20Style%20Guide.pdf
doesn’t help and I can’t find the relevant phrase needed to get google to assist.
Anyone know the relevant phrase?
Why? I was born in 1980. In 1981 I got one year old, not two 
And after your first birthday you had been alive more than 1 year.
And before your first birthday you weren’t 0 years old.
Or is that incorrect?
Here is an example of the usage that I expected:
https://www.myfrs.com/Faq_PensionPlan.htm
“Years of creditable service” means the total of all years and parts of years you worked in a retirement-covered position with an FRS employer.
However I still can’t find the term of art that covers the language used in this situation.
And I’m open to there being multiple contradictory protocols.
I think it’s because it’s based on an average/probable range? 1996‐12‐31–1998‐01‐01 would reasonably be considered 1 year (whole of 1997), 1996‐07‐01–1998‐07‐01 would be 2 years, 1996‐01‐01–1998‐12‐31 would be 3 years. Without more accuracy in the data, 2 years is probably the most accurate time span that can be established.
You also weren’t 1 year old. You were 0.something years old, or your age was measured in days, weeks, and/or months. Depending on how you round, you would indeed say that someone was 0 years old, because that’s how age is usually rounded (that is, down, and not to nearest natural number). “You have lived a total 0 complete earth‐around‐the‐sun revolutions.”
Some time in the past I found the way that, at least in some circumstances, measurements are counted when using whole numbers.
(There is an example of that “whole and any part” usage in the post above yours.)
I’ve heard the term “calendar years” used for something like this, I think. Is that what you’re thinking of?
That helped a bit.
Searching: count of years
We can count intervals or markers.
In some situations, sometimes “years of service”, markers get counted.
But I still have not found anything authorative.
But even “years of service” we won’t be able to tell. We don’t know from “1996–1998” that they had three years of service. With “mathemetical” rounding to nearest whole number, we get anywhere from 1 to 3 years, depending on where in ’96 and ’98 the label was founded respectively disbanded. With “age”(?) rounding, we would be at 2 years at the most (if you’re born on 1996‐1‐1 you’d turn 3 on 1999‐1‐1 and still be considered 2 y.o. on 1998‐12‐31). I’m pretty sure labels use the same rounding algorithm as artists which I’m pretty sure was designed for people artists for which it’d be most natural to use the latter, but even if we use the “mathematical” rounding to nearest whole number, the probability of “2” being correct is about 50% with “1” and “3” being at 25% each. In both of these scenarios, “2” is the best option to show.
If you want a more accurate date, you need to give the system more accurate information (e.g., the months of founding and disbanding the label)—at “year” accuracy, saying that it lasted for 2 years is the most accurate we can say.
The “years of service” doesn’t make sense to me. There’s no way this label existed for three (whole) years.
I still can not find the terms that are used around the various ways of counting years of being in business.
Clearly there are multiple ways this can be done.
And maybe the current method is producing a result that agrees with the pre-existing norm (if any) for counting “years in business”.
But maybe it dos not.