Ha, good one! Got me interested, how the standards bodies handle something like that. Found this in the Wikipedia article for byte:

The unit symbol for the byte is specified in IEC 80000-13, IEEE 1541 and the Metric Interchange Format[10] as the upper-case character B.

In the International System of Quantities (ISQ), B is the symbol of the bel, a unit of logarithmic power ratio named after Alexander Graham Bell, creating a conflict with the IEC specification. However, little danger of confusion exists, because the bel is a rarely used unit. It is used primarily in its decadic fraction, the decibel (dB), for signal strength and sound pressure level measurements, while a unit for one-tenth of a byte, the decibyte, and other fractions, are only used in derived units, such as transmission rates.

Somewhat disappointing. “There’s a conflict but it’s fine”.

But there is also nothing contradictory or wrong with the unit MB.

And I didn’t say that. Admittedly, the constraint “If the image size is defined as binary” probably could’ve been expressed better (I’m not a native speaker). File sizes are usually calculated in binary units (at least by Windows and the Linux distros I know - even though Windows continues to claim those are megabytes and Linux adopted the standardized units) and I’d bet that’s the case for file.coffee, too.

Oh, well, I’m pretty sure we’re not really disagreeing anyway. So let’s conclude with the obligatory relevant xkcd.

No.

Where’s the contradiction? Yes, it’s just a different prefix but it results in a different number. What I meant to adress is that very often people write MB/megabytes (10⁶ = 1,000,000 bytes) but actually mean MiB/mebibytes (2²⁰ = 1,048,576 bytes). RAM vendors possibly most prominently.

Well, acshually, bit doesn’t have a metric symbol and ‘b’ is defined as barn. So Mb would be a megabarn.

Edit: And to be even more nitpicking: If the image size is defined as binary, it should be MiB (mebibyte) since “mega” is defined as base 10.