English letter frequencies in json format

letter_freq.json

{
    "a": 8.167,
    "b": 1.492,
    "c": 2.782,
    "d": 4.253,
    "e": 12.702,
    "f": 2.228,
    "g": 2.015,
    "h": 6.094,
    "i": 6.966,
    "j": 0.153,
    "k": 0.772,
    "l": 4.025,
    "m": 2.406,
    "n": 6.749,
    "o": 7.507,
    "p": 1.929,
    "q": 0.095,
    "r": 5.987,
    "s": 6.327,
    "t": 9.056,
    "u": 2.758,
    "v": 0.978,
    "w": 2.360,
    "x": 0.150,
    "y": 1.974,
    "z": 0.074
}

Can I ask the source from where you got this data?

(From what I can tell, these numbers all look highly plausible, but I’d just feel better if I knew where the data was derived from?)

@handcoding no clue after 11 years. I suspect I would have stolen it from a wikipedia source but I can't be certain at this point.

This frequency table doesn't appear to be accurate for 'json keys' in my opinion, because of the high usage of "size" in tech. That alone seems like it would put 'z' at a higher frequency than 'q'.

Here is my attempt at this:
'A', 'B', 'C', 'D', 'E',
'F', 'G', 'H', 'I', 'J',
'K', 'L', 'M', 'N', 'O',
'P', 'Q', 'R', 'S', 'T',
'U', 'V', 'W', 'X', 'Y',
'Z', '_', '-'

68, 13, 29, 39, 110,
11, 15, 7, 64, 10,
10, 32, 30, 48, 43,
26, 5, 57, 61, 71,
23, 10, 10, 10, 18,
20, 41, 10

This is derived by choosing a random sampling of 100 reasonable keys and taking their frequency, then modifying a bit.

Modifications:
Add a high likelihood of underscore.
Add a lesser arbitrarily chosen amount for dash.

JSON keys only allow uppercase, lowercase, numbers, underscore, and dash.

Here is my frequency normalized:
A 7.63
B 1.46
C 3.25
D 4.38
E 12.35
F 1.23
G 1.68
H 0.79
I 7.18
J 1.12
K 1.12
L 3.59
M 3.37
N 5.39
O 4.83
P 2.92
Q 0.56
R 6.40
S 6.85
T 7.97
U 2.58
V 1.12
W 1.12
X 1.12
Y 2.02
Z 2.24
_ 4.60

• 1.12

The biggest difference I see actually is that Q frequency in the above list is abnormally low.