2

This is a 1024-byte length string.

How to pronounce the sentence above?

This is a one-thousand-and-twenty-four byte length string.

or

This is a one-oh-two-four byte length string.

xmllmx
  • 2,698
  • 3
    Voting to close because answers can only be opinion based. –  Mar 26 '14 at 02:40
  • But I use the second form :-) –  Mar 26 '14 at 02:41
  • 1
    @andy256: I don't think it should be closed. I think it should be treated as a poll, and vote your preferred pronunciation up. If you have pronunciation distinct from those already posted, add it. – Cyberherbalist Mar 26 '14 at 16:01
  • It should be possible to write answers based on facts or expertise here. There likely isn’t a single correct pronunciation, but there’s probably a very short list of common ones. – Bradd Szonye Mar 27 '14 at 23:31
  • That's easy, '1K' – Mitch Dec 26 '16 at 20:17

6 Answers6

7

"This is a ten-twenty-four byte length string."

I've found this to be one of the more common ways to say it. This is also typical of other numbers over 1024, such as 2048 ("twenty-forty-eight") and 4096 ("forty-ninety-six").

meiguoren
  • 171
2

This is a question of convention. There are several correct ways to pronounce it. Complicating the question, it's poorly-enforced convention.

In general, conventions follow convenience. Due to some neuro/psychological factors (humans prefer to not count beyond three, humans have a hard time remembering unstructured lists, etc), most long numbers are pronounced in batches of shorter numbers.

Consider this list of numbers, and their common pronunciations:

  • 1024 - "ten twenty-four"
  • 124 - "one twenty-four"
  • 104 - "one oh four" (alternatively, "one hundred and four")
  • 100 - "one hundred" (alternatively, "a hundred")
  • 1000 - "one thousand" (alternatively, "a thousand")
  • 1004 - "one thousand four" (alternatively, "one thousand and four")

Some general preferences are apparent:

  • Grouping numbers into two-digit components is good
  • Grouping numbers with very different magnitudes (at least three orders) is bad
  • Fewer components, where components are either grouped or ungrouped numbers by the above preferences, is good
  • If there are a mix of grouped and ungrouped numbers, the ungrouped should be on the left if it possible under the first two preferences. (e.x. 124 is read like 1|24 and not 12|4, whereas 37,001 would be read like 37000|1, not 3|7001)

These aren't ironclad rules. For example, when we left the " 'oughts" (2000-2009), there was a confusion whether we should go from reading it as "two thousand nine" to "two thousand ten" (2000|10) or switch to "twenty ten" (20|10). Obviously, under the above preference, 20|10 is the conventional way to say it, but there's a certain amount of conversion cost, so people were divided on whether it was worth converting.

rsegal
  • 2,420
2

After the or a it’s common to omit the leading one from numbers in speech:

This is a thousand-twenty-four-byte string.

This is in contrast to other leading digits:

That is a two-thousand-forty-eight-byte string.

But as others have noted, a programmer is quite likely to abbreviate these numbers:

This is a one-K string. That is a two-K string.

Note that the K implies “kilobytes,” so you wouldn’t pronounce “bytes.”

Bradd Szonye
  • 15,402
  • K is generally pronounced simply as "kay" rather than "kilo", and M as "meg" rather than "mega". Note that in programmerese, the Kilo- prefix usually refers to 1024 , not 1000 as in normal SI/"metric" usage. There's an informal convention of using uppercase K, rather than the lowercase k, to denote this but not everyone honors that convention... and there's no consensus on how to distinguish 1K1K from 1k1k, M is used for both. – keshlam Mar 28 '14 at 05:46
  • @keshlam Agreed – I wrote it that way to imply the pronunciation “one-kay.” (To expand on your final sentence, people sometimes even mix binary and decimal thousands, especially when measuring storage.) – Bradd Szonye Mar 31 '14 at 23:37
  • I debated mentioning the fact that hard disk marketing traditionally quotes disk sizes in terms of decimal powers rather than binary powers, but didn't want to get into debating the (mostly bogus) historical justifications for doing so. – keshlam Apr 01 '14 at 12:32
1

Either one thousand and twenty-four or ten twenty-four, but most computer folks would probably read it as one K, since it's obvious from the context that bytes are what are being discussed.

If it was a memory size, it might be necessary to say the i1103 is a one K-bit memory or the AM27S37 is a one K-byte memory.

Spehro Pefhany
  • 10,419
1

This is a one-oh-two-four byte length string.

As a part-time programmer, that's the way I have always mentally said it, so that's the way I'd say it out loud.

I wouldn't shorten it to "1K", unless you're being fairly non-technical. Because an important characteristic of 1024 is that it's divisible by 2, 4, 8, 16, 32, etc. So if the term arises during a discussion that involves arithmetic, I'd stick to the exact formulation of the length.

Terpsichore
  • 3,364
  • 1
  • 14
  • 13
0

In programming, it would be most common to convert bytes to a more manageable unit so you don't have to express the full number.

eg. it is much easier to just state "one kilobyte" than "one thousand twenty four bytes" even though the latter is still correct. You certainly wouldn't say "eight thousand one hundred ninety-two bits" :)

You would never say the numbers individually as in your second example.

mattacular
  • 765