det

But Hebrew and Arabic speakers read numbers left to right (even though they read everything else right to left). So they treat the numbers as big-endian.

More evidence in favor of big-endian: In modern Hebrew and Arabic, numbers are written in the same direction as in English: e.g. 

.שטחה של המדינה הוא 22,072 קמ"ר 

As a native English speaker (and marginal Hebrew reader), I read each word in that Hebrew sentence right-to-left and then read the number left-to-right.

I never considered the possibility that native Hebrew speakers might read the number from right to left, in a little-endian way. But my guess is (contra lsusr) nobody does this: when my keyboard is in Hebrew-entry mode, it still writes numbers left-to-right.^[1] 

This indicates that even when you give little-endian an advantage, in practice big-endian still wins out.

1. ^{^}

   I also tested in Arabic-entry mode, and it does the same even when using the Eastern Arabic numerals, e.g ١٢٣٤٥٦٧٨٩. 
   
   It's hard to Google for this, but this indicates that modern Arabic also treats numbers as left-to-right big-endian [I just verified with an Arabic speaker that this is indeed the case]. It's possible this was different historically, but I'm having a hard time Googling to find out either way.

SMTM has a follow-up post that goes into how confusing citrus classifications are.

In particular:

• The British called all citrus "lime" or "lemon" interchangeably (like you say).
• Lemons can be green and limes can be yellow, so you can't clearly distinguish based on color.
• We still use the word "lime" for a bunch of different kinds of citrus, so we're not that much better.