Talk:Binary prefix

Template-table references removed from article, preserved here

Bit rates (data-rate units) Name Symbol Multiple bit per second bit/s 1 1 Metric prefixes (SI) kilobit per second kbit/s 103 10001 megabit per second Mbit/s 106 10002 gigabit per second Gbit/s 109 10003 terabit per second Tbit/s 1012 10004 Binary prefixes (IEC 80000-13) kibibit per second Kibit/s 210 10241 mebibit per second Mibit/s 220 10242 gibibit per second Gibit/s 230 10243 tebibit per second Tibit/s 240 10244

Multiple-bit units Decimal Value Metric 1000 kbit kilobit 10002 Mbit megabit 10003 Gbit gigabit 10004 Tbit terabit 10005 Pbit petabit 10006 Ebit exabit 10007 Zbit zettabit 10008 Ybit yottabit 10009 Rbit ronnabit 100010 Qbit quettabit Binary Value IEC Memory 1024 Kibit kibibit Kbit Kb kilobit 10242 Mibit mebibit Mbit Mb megabit 10243 Gibit gibibit Gbit Gb gigabit 10244 Tibit tebibit — 10245 Pibit pebibit — 10246 Eibit exbibit — 10247 Zibit zebibit — 10248 Yibit yobibit — — — Orders of magnitude of data

Multiple of a unit of measurement by an integer power of two??

WRT "A binary prefix is a unit prefix that indicates a multiple of a unit of measurement by an integer power of two". Is it? Or is it a power of 1024? Yes, they are all powers of 2, but calling them that seems misleading. That they are power of 2 doesn't seem like the most central defining property of this set of multiples. Stevebroshar (talk) 20:57, 21 December 2024 (UTC)

If we had names for other powers of two, they too would be called binary prefixes. The fact that we find only a certain subset of these prefixes convenient enough for general use to create a name does not mean that we should necessarily use the smallest (obvious) category that contains this subset. And no, it is not misleading: it is predicated on the practicality of implementing memory sizes as powers of 2, not of 1024. When the prefixes 'centi', 'deci', 'deca' and 'hecto' fall into disuse, will it be misleading to call the remaining prefixes (all of which are powers of 1000) "decimal prefixes"? (Actually, these are more commonly called "metric prefixes", but that is an even vaguer category.) —Quondum 21:13, 21 December 2024 (UTC)

@Stevebroshar: has a valid point. Binary prefixes historically are defined in positive integer powers of 1024 and are likely to continue to do so. They go back to the approximate equivalence of 1,024 to 1,000 and unlike metric prefixes are not defined for each power of the base number to a maximum and not to a minimum at all. Whether the rarely used metric prefixes fall into disuse or not is irrelevant, they would remain defined. I think we would have to find an RS to change the article to state "positive integer powers of 1024" but with one I would support such a such a change. Tom94022 (talk) 20:22, 22 December 2024 (UTC)

I agree with both of you to some extent. I agree that if there was a prefix that was binary and not a factor of 1024, it could be called a binary prefix ... as that is an accurate description. But, there aren't any. Is it reasonable to describe something that doesn't exist? And I think likely never will? ... Thing is, there's no RS for the current definition :o) This article seems to conflate a general definition of binary prefix with the IEC standard that defines powers of 1024. Is the article about the general definition? If so, what sources back that? If it's more strictly about the IEC standard prefixes, then I think it reasonable to highlight that they are powers of 1024, not 2. Stevebroshar (talk) 15:10, 25 December 2024 (UTC)

WRT metric: this article is not about metric prefixes except from a historical perspective -- the binary prefixes were created since the metric prefixes are not suitable for (binary) computing. Therefore, what's true about metric seems to have little bearing on what these binary prefixes are. But, if you want go down that rabbit hole: Metric does have some sizes that are not factors of 1000, but in the context of computing, we only use the 1000-based guys. Metric has subdivisions of the base unit that the binaries don't. They are similar yet different animals. ... One might call metric 1000-based with a few exceptions. Along that line, it seems reasonable to simplify the definition of binary prefixes as 1024-based. Stevebroshar (talk) 15:15, 25 December 2024 (UTC)

Clause 4 of ISO/IEC 80000:13-2008 is entitled "Prefixes for binary multiples" and lists the prefixes as positive powers of 2¹⁰; I think this enough of a reliable source to state binary prefixes are defined in positive integer powers of 1024. The fact that SI (metric?) prefixes happen to be defined in integer powers of ten is not particularly relevant. Should we go forward with the change and see what it provokes? Tom94022 (talk) 07:07, 26 December 2024 (UTC)

I think so, yes. ... So far I have been unable to find the IEC doc online. Maybe one has to buy it. I'm not doing that. 150,000 CHF!!!! Is there a way to include an excerpt from the doc in the ref? Stevebroshar (talk) 09:33, 26 December 2024 (UTC)

That is 150 CHF, which nevertheless is far too steep. The citation templates generally include a |quote= parameter for quoting from citations, but I doubt that this will be helpful here. The 2018 standard includes, as mentioned by Tom94022, the section heading that refers to prefixes of binary multiples, but this does not in itself define the term "binary prefix" directly. My faded memory would have me believe that the section is very brief, consisting of little more than a table that defines the prefixes 'kibi' to 'yobi' and their associated symbols. The 'little more' is of course what we might be interested in, but I am almost certain that it does not imply a definition that restricts the term to the defined prefixes or even the continuation of the defined prefixes, and we should avoid defining it that way. What it does define is the specific set of eight binary prefixes, and we can say that those eight prefixes are defined as binary prefixes. The article already adequately reflects this.

We could try to reword it to avoid implying a specific definition, but rather to reflect that the term "binary prefix" includes the formally defined prefixes. —Quondum 15:33, 27 December 2024 (UTC)

For example: 1,024 bytes = 1 kilobyte (or kibibyte)

A dispute has arisen at Power of two about the choice of words between ‘For example: 1,024 bytes = 1 kilobyte (or kibibyte)’ and ‘For example: 1,024 bytes = 1 kibibyte ≈ 1000 bytes = 1 kilobyte’. Constructive opinions are invited at Talk:Power_of_two#For_example:_1,024_bytes_=_1_kilobyte_(or_kibibyte). Dondervogel 2 (talk) 17:10, 26 December 2024 (UTC)