I write about radios a lot, and use the abbreviation MHz and kHz with frequencies, but I’ve always wondered why one is capitalized and the other is not. I asked ChatGPT:
Why is the first letter of the abbreviation MHz usually capitalized, but the first letter of kHz is not?
ChatGPT replied:
The reason for this difference in capitalization between “MHz” and “kHz” has to do with the SI (International System of Units) rules for unit symbols.
In the SI system, unit symbols are always written in lowercase letters, except when the symbol is derived from a proper noun. In the case of “MHz”, the unit is derived from “megahertz”, which is a combination of the metric prefix “mega-” and the unit “hertz”. Since “mega-” is a proper prefix derived from a Greek word meaning “great”, the first letter of the symbol “MHz” is capitalized to indicate this derivation from a proper noun.
On the other hand, “kHz” is derived from “kilohertz”, which is also a combination of the metric prefix “kilo-” and the unit “hertz”. However, “kilo-” is not a proper prefix in the same way that “mega-” is, so the symbol “kHz” does not capitalize the first letter.
It’s worth noting that while these rules are widely followed, there may be variations or exceptions depending on the specific context or style guide being used.
I didn’t quite follow the rationale, so I asked a follow-up question:
So why is “kilo-” not a proper prefix, but “mega-‘” is?
ChatGPT was happy to get into the weeds on this:
The reason “mega-” is considered a proper prefix, while “kilo-” is not, has to do with their origins and usage.
The prefix “kilo-” comes from the Greek word “khilioi,” which means “thousand”. It has been used in the metric system since its inception in the 18th century and is used to represent a factor of 1000. While “kilo-” is a commonly used prefix, it is not considered a proper prefix because it is not used exclusively within the SI system. For example, the prefix “kilo-” is also used in computer science to represent a factor of 1024, which is not part of the SI system.
On the other hand, “mega-” is derived from the Greek word “megas,” which means “great” or “large”. It is used in the metric system to represent a factor of one million and is exclusively used within the SI system. Therefore, “mega-” is considered a proper prefix because it is only used in the context of the SI system.
In summary, while “kilo-” is a commonly used prefix, it is not considered a proper prefix because it is used in contexts outside of the SI system. “Mega-“, on the other hand, is exclusively used within the SI system and is considered a proper prefix.
I wasn’t having any of that. I wrote:
That doesn’t make sense. “Mega-” is used in computer science the same way “kilo-” is.
ChatGPT responded:
You are correct that the prefix “mega-” is commonly used in computer science to represent a factor of 1,048,576, which is equal to 2^20 (i.e., 2 to the power of 20). This is sometimes called a “binary megabyte” or “mebibyte,” and it is used to represent the amount of computer memory or storage capacity.
However, it’s important to note that the use of “mega-” in computer science in this way is not part of the SI system of units. The SI system officially defines “mega-” as representing a factor of 1,000,000, and this definition is used consistently across all fields that use the SI system, such as physics and engineering.
To avoid confusion between the SI definition of “mega-” and its use in computer science, the International Electrotechnical Commission (IEC) established a separate set of binary prefixes for use in computer science, including “mebi-” to represent a factor of 2^20. These binary prefixes are not part of the SI system, but they are used in many technical contexts in computer science and related fields.
So to summarize, while “mega-” is used in computer science to represent a factor of 2^20, this usage is not part of the SI system and is not considered a proper use of the prefix within the SI system. The SI definition of “mega-” represents a factor of 1,000,000, and this definition is used consistently across all fields that use the SI system.
ChatGPT can be asked to regenerate a response. I wasn’t satisfied, so here is the regenerated response:
You are correct that “mega-” is used in computer science to represent a factor of 1,048,576 (2^20), which is not a multiple of 1,000, and is different from its use in the SI system, where it represents a factor of 1,000,000 (10^6).
In the context of computer science, the prefixes “kilo-“, “mega-“, “giga-“, and so on, are used to represent powers of 2 rather than powers of 10. This is because computers work in binary, which is a base-2 system, rather than decimal, which is a base-10 system. Therefore, in computer science, 1 kilobyte is equal to 2^10 bytes, or 1,024 bytes, rather than 1,000 bytes.
In general, while the prefixes “kilo-” and “mega-” are used in both the SI system and in computer science, their meaning and usage can differ depending on the context. However, regardless of context, the rule for capitalization of prefixes in SI units remains the same: only proper prefixes derived from proper nouns are capitalized in unit symbols, and other prefixes are not capitalized.
I think ChatGPT veering into computer science was a red herring, and adds nothing to the explanation. It’s irrelevant, misleading and confusing. The key point is how the prefix can be part of a proper noun. One would probably capitalize “Great Caesar” as a title, but they wouldn’t capitalize “one thousand Caesars.” The number 1024 for computer memory and 1000 for SI units has nothing to do with the distinction.
Nevertheless, ChatGPT plus human reflection did result in an answer.
And don’t forget, mHz is the abbreviation for millihertz.