Talk:Decade (log scale)

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia
WikiProject iconMeasurement Start‑class (defunct)
WikiProject iconThis article is within the scope of WikiProject Measurement, a project which is currently considered to be defunct.
StartThis article has been rated as Start-class on Wikipedia's content assessment scale.

Suggested change[edit]

Begin "One decade is a power of 10 difference ..." since a factor of 10 difference might suggest x-y=some factor of 10 (eg. 1, 2, 5, 10) — Preceding unsigned comment added by 82.132.216.182 (talk) 16:09, 13 October 2015 (UTC)[reply]

I reworded the lead to eliminate this ambiguity. —Quondum 22:46, 31 December 2017 (UTC)[reply]

Decade indicates absolute or relative position on log10 scale?[edit]

The article's sources use "decade" as a log10 measure of the size of the (ratio) interval between a pair of values, with the twist that it is also refers to that interval as a set. Other than the logarithm, this is the same as the use of "decades" as a calendrical term to denote both specific intervals of years and a measure of the length of those intervals in multiples of 10.

However, some currently unsourced material in the section on scientific notation uses "decade" as a synonym for the older term characteristic (of the base 10 logarithm), to mean the absolute position of an individual value, or equivalently its position relative to 1. That's an interesting proposal, but is the term actually used that way and are there sources for such use?

These two ways of speaking are not quite compatible. As an attribute of pairs, the interval from 67.3 to 673 is a single decade. "Decade" as claimed in the scientific notation section would say that values from 67.3 to 100 are assigned to the decade (10,100) and those between 100 and 673 belong to (100,1000). The calendar equivalent would be being able to talk about the 1920's or the 1930's as decades, but not the decade of the Great Depression, because decades are declared to only begin in years divisible by 10. Also, we don't talk about the years of the 1960's as having a decade of 196, but what's currently in the article would prescribe statements like "673 has decade equal to 3" and "the decade of 67.3 is 2".

The same issue is currently arising in order of magnitude and scientific notation and their talk pages, and occurs in a few related articles. My understanding is that the use of "decade" is, other than the mental image of an interval of values, the same as "order of magnitude" and that virtually all scientific use of such terminology is as a relation between pairs of quantities, not an attribute of individual ones. In any case, this being Wikipedia there is also the question of what sources say. 73.89.25.252 (talk) 07:25, 11 December 2020 (UTC)[reply]

To put the question more succinctly: is "decade" used (outside this Wikipedia article) to mean "an interval of values from 10^K to 10^(K+1), for integer K, and only integer values of K are allowed"? 73.89.25.252 (talk) 14:03, 11 December 2020 (UTC)[reply]
In acoustics, the decade is used to mean a factor of 10 in frequency. That can be from 10 Hz to 100 Hz. It can also be from 8.9 Hz to 89 Hz. I've seen both. Dondervogel 2 (talk) 19:44, 11 December 2020 (UTC)[reply]
Unless there's a source supporting the usage in the recently added section on scientific notation, I suggest restoring the article to its July 3 2020 version (https://en.wikipedia.org/w/index.php?title=Decade_(log_scale)&oldid=965766495 ) and adding the statement that decade also refers to intervals (such as frequency bands) rather than only the ratio or log-ratio of the endpoints. 73.89.25.252 (talk) 22:10, 11 December 2020 (UTC)[reply]
The meaning of the new additions you refer to is unclear to me. Dondervogel 2 (talk) 09:23, 12 December 2020 (UTC)[reply]
It's a WP:OR attempt to fix some wrong or fringe usage of "order of magnitude" being debated at the scientific notation and order of magnitude talk pages, by replacing it with a newly invented use of "decade". The new thing shares the same problems as the old thing which might be why it is not in use.
It's true that any normalization scheme for scientific notation amounts to a choice of one decade (as an interval) of positive numbers allowed as mantissae. If people want that stated in this article I have no reason to object, it's a simple math fact that can just be asserted if there is no reference, or a proof written out if needed. 73.89.25.252 (talk) 01:23, 18 December 2020 (UTC)[reply]
I edited the Scientific Notation section to remove errors and connect it to the topic of the article, by converting it to an example of "decades" in practice. The source cited there is not WP:RS, but rather than remove it I made the text conform to what the source actually says about decade (which is almost nothing) and exponent (which is that the b of a x 10^b is called by that name). This removes current problems until some decision is reached on whether to have that section in the article at all. 73.89.25.252 (talk) 05:24, 18 December 2020 (UTC)[reply]

Decade used for things other than frequencies[edit]

The current top paragraph of article describes decade as a general mathematical construct, log of ratio. The earlier (3 July 2020) version stated that it is applied to frequencies:

- - "One decade (symbol dec) is a unit for measuring frequency ratios on a logarithmic scale, with one decade corresponding to a ratio of 10 between two frequencies (an order of magnitude difference). It is especially useful when describing frequency response of electronic systems, such as audio amplifiers and filters."

( https://en.wikipedia.org/w/index.php?title=Decade_(log_scale)&oldid=965766495 )

Is "decade" used outside the context of frequencies? 73.89.25.252 (talk) 08:44, 21 December 2020 (UTC)[reply]

Good question. I can't think of any examples other than a frequency range. Dondervogel 2 (talk) 09:35, 21 December 2020 (UTC)[reply]