Sending Encrypted Text Messages via BlockChain

[Phinnaeus Gage]

I thought of this last night, and not sure if this approached has been discussed before. The following is an example of how a rogue entity could send encrypted text messages back and forth to one another, and still keep the coins between themselves.

.011002
.011111
.01102
.0102
.0111
.01111
.0102
.011002
.01101
.01001
.011101
.011
.0102

.020101
.020002
.0211102
.021
.02101
.021
.02011
.020111
.0211
.02011
.02001
.021

The above that I devised is based on the simplest code readily available, but a more complex coding system, with each person having the key, can create a more complex one that would be a hell of a lot harder to crack. And it could all be done in the open for all to see.

In fact, a simple program could be built to create the sums on demand, ready to be hand-sent one-by-one or automatically, not all at once, but at a rate as if a human were inputting the amounts.

I started this post to bring awareness that this option exist and should be deeply considered.

Can you crack the code above?

~Bruno~

[teflone]

YOU TAKE THAT BACK ABOUT MY MOTHER!!!!   

[Xenland]

Hmmmm this could be made into a wevsite easily!!!

[legolouman]

Great idea, but what if the network accepts the payments in a funny order? Certainly won't always be the case, but it might happen. Your message of "cheese" could turn into "echsee"

[Xenland]

Great idea, but what if the network accepts the payments in a funny order? Certainly won't always be the case, but it might happen. Your message of "cheese" could turn into "echsee"

Hmm maybe there is a timestamp when they were "sent" to the nodes (instead of looking at the timestamps when they were mined into a block)

[fatigue]

Either that or the transactions would have to be sent over a longer time period to ensure that they are confirmed in order.

[Serge]

nah, you could just put 01,02,03... prefix or suffix to make sure correct order and send them all in one batch

[adamstgBit]

do not needlessly pollute the blockchain

[Xenland]

Either that or the transactions would have to be sent over a longer time period to ensure that they are confirmed in order.

Yeah, basically you'd have to cron job the block chain for confirmed transaction then send the next letter so the time would take (if your lucky) 10 minutes per letter :/

You could however have one transaction represent multiple letters (maybe up to 5 letters) for example

with single letters(to show a comparison)

A(0.00000001 BTC)
-
Z(0.00000026 BTC)

Multiple letters

AA(0.00000027 BTC)
AB(0.00000028 BTC)
....
....
AZ(0.00000054 BTC)
....
....
BZ(0.00000080 BTC)

You get the point...

[bbit]

there you go again taking shots at my mom agian! 

[austonst]

Correct me if I'm wrong, but if a transaction has multiple outputs, doesn't the client generating the transaction have complete control over the numbering of the outputs? This would make it easy to read them in a certain order.

[fatigue]

Either that or the transactions would have to be sent over a longer time period to ensure that they are confirmed in order.

Yeah, basically you'd have to cron job the block chain for confirmed transaction then send the next letter so the time would take (if your lucky) 10 minutes per letter :/

You could however have one transaction represent multiple letters (maybe up to 5 letters) for example

with single letters(to show a comparison)

A(0.00000001 BTC)
-
Z(0.00000026 BTC)

Multiple letters

AA(0.00000027 BTC)
AB(0.00000028 BTC)
....
....
AZ(0.00000054 BTC)
....
....
BZ(0.00000080 BTC)

You get the point...

Interesting concept but it would be limited by the fact that for example .00000028 BTC could be misinterpreted as BA.

What would be more susceptible to observation is using all of the decimal places but denoting 2 places to each letter

A(01)
B(02)
C(03)
D(04)
E(05)
F(06)
G(07)
H(08)
I(09)
J(10)
K(11)
L(12)
M(13)
N(14)
O(15)
P(16)
Q(17)
R(18)
S(19)
T(20)
U(21)
V(22)
W(23)
X(24)
Y(25)
Z(26)

Although the key to the code would obviously have these more random assignments

Message: This is coded

Transaction 1: This (.20080919)
Transaction 2: isco (.09190315)
Transaction 3: ded (.04050400)

[austonst]

Using base 27, where 0 is a space, A=1, ..., Z=26, it would be possible to fit 5 letters into 8 decimal digits since 27^5=14,348,907. So, conversion for "Your mom" would be as follows:

Code:

Y: 25 * 27^4 = 13,286,025
O: 15 * 27^3 = 295,245
U: 21 * 27^2 = 15,309
R: 18 * 27^1 = 486
 :  0 * 27^0 = 0
Result: 0.13597065

M: 13 * 27^4 = 6,908,733
O: 15 * 27^3 = 295,245
M: 13 * 27^2 = 9,477
 :  0 * 27^1 = 0
 :  0 * 27^0 = 0
Result: 0.07213455

Decoding would be similar:

Code:

13,597,065 / 27^4 = 25 r311,040
311,040 / 27^3    = 15 r15,795
15,795 / 27^2     = 21 r486
486 / 27^1        = 18 r0
0 / 27^0          = 0
25 15 21 18  0
 Y  O  U  R  _

And so on for the next lines. You could extend this further by spending full coins (one preceding A-E won't end up going to the 9th digit, but F might, and anything higher will), but it would probably be a lot cheaper to just have multiple outputs and stick them together.

This may have been what Xenland was going for, but his look a little different... Of course, this works with a full alphabet and a character for a space or end of line, and would work with any sort of encryption that uses a 26 character alphabet. Slightly more compact than fatigue's, but a bit more difficult to work with.

[fatigue]

Using base 27, where 0 is a space, A=1, ..., Z=26, it would be possible to fit 5 letters into 8 decimal digits since 27^5=14,348,907. So, conversion for "Your mom" would be as follows:

Code:

Y: 25 * 27^4 = 13,286,025
O: 15 * 27^3 = 295,245
U: 21 * 27^2 = 15,309
R: 18 * 27^1 = 486
 :  0 * 27^0 = 0
Result: 0.13597065

M: 13 * 27^4 = 6,908,733
O: 15 * 27^3 = 295,245
M: 13 * 27^2 = 9,477
 :  0 * 27^1 = 0
 :  0 * 27^0 = 0
Result: 0.07213455

Decoding would be similar:

Code:

13,597,065 / 27^4 = 25 r311,040
311,040 / 27^3    = 15 r15,795
15,795 / 27^2     = 21 r486
486 / 27^1        = 18 r0
0 / 27^0          = 0
25 15 21 18  0
 Y  O  U  R  _

And so on for the next lines. You could extend this further by spending full coins (one preceding A-E won't end up going to the 9th digit, but F might, and anything higher will), but it would probably be a lot cheaper to just have multiple outputs and stick them together.

This may have been what Xenland was going for, but his look a little different... Of course, this works with a full alphabet and a character for a space or end of line, and would work with any sort of encryption that uses a 26 character alphabet. Slightly more compact than fatigue's, but a bit more difficult to work with.

Ahh but the complication is key to preventing decryption. I like it! The only spot i could see a possible misinterpretation is the 0 being transmitted. how would the receiver know clearly if there was a space or if the word was continued in the next transaction?

EDIT: i see now that it would be a standard to perform the decryption method 5 times in all situations.

[SgtSpike]

Speaking of this...

Someone seemingly sent me such a message to my main public BTC address a while back in block 181957.  I never did figure out what it was supposed to say...

18TKNbSLTrd3a2W8mtoH5uNzFhWRWNcuHU: 0.0100111
18TKNbSLTrd3a2W8mtoH5uNzFhWRWNcuHU: 0.01101001
18TKNbSLTrd3a2W8mtoH5uNzFhWRWNcuHU: 0.01100011
18TKNbSLTrd3a2W8mtoH5uNzFhWRWNcuHU: 0.01100101
18TKNbSLTrd3a2W8mtoH5uNzFhWRWNcuHU: 0.001
18TKNbSLTrd3a2W8mtoH5uNzFhWRWNcuHU: 0.01001101
18TKNbSLTrd3a2W8mtoH5uNzFhWRWNcuHU: 0.01101001
18TKNbSLTrd3a2W8mtoH5uNzFhWRWNcuHU: 0.0110111
18TKNbSLTrd3a2W8mtoH5uNzFhWRWNcuHU: 0.01100101
18TKNbSLTrd3a2W8mtoH5uNzFhWRWNcuHU: 0.01100011
18TKNbSLTrd3a2W8mtoH5uNzFhWRWNcuHU: 0.0111001
18TKNbSLTrd3a2W8mtoH5uNzFhWRWNcuHU: 0.01100001
18TKNbSLTrd3a2W8mtoH5uNzFhWRWNcuHU: 0.0110011
18TKNbSLTrd3a2W8mtoH5uNzFhWRWNcuHU: 0.011101
18TKNbSLTrd3a2W8mtoH5uNzFhWRWNcuHU: 0.001
18TKNbSLTrd3a2W8mtoH5uNzFhWRWNcuHU: 0.0100001
18TKNbSLTrd3a2W8mtoH5uNzFhWRWNcuHU: 0.01110101
18TKNbSLTrd3a2W8mtoH5uNzFhWRWNcuHU: 0.01101001
18TKNbSLTrd3a2W8mtoH5uNzFhWRWNcuHU: 0.011011
18TKNbSLTrd3a2W8mtoH5uNzFhWRWNcuHU: 0.011001
18TKNbSLTrd3a2W8mtoH5uNzFhWRWNcuHU: 0.8021192

[austonst]

It's ASCII: "Nice Minecraft Build"

Not sure about the last one. Change? IP address?

[SgtSpike]

Ah, I tried that, but couldn't seem to make it work.  Maybe I failed to add the trailing zeros or something.  Anyway, thanks for translating. 

[Xenland]

Interesting concept but it would be limited by the fact that for example .00000028 BTC could be misinterpreted as BA.

What would be more susceptible to observation is using all of the decimal places but denoting 2 places to each letter

A(01)
B(02)
C(03)
D(04)
E(05)
F(06)
G(07)
H(08)
I(09)
J(10)
K(11)
L(12)
M(13)
N(14)
O(15)
P(16)
Q(17)
R(18)
S(19)
T(20)
U(21)
V(22)
W(23)
X(24)
Y(25)
Z(26)
AB(27)
AC(28)
AD(29)
AE(30)
AF(31)
AG(32)
AH(33)
AJ(34)
AK(35)
AL(36)
AM(37)
AN(38)
AO(39)
AP(40)
AQ(50)
AR(51)
AS(52)
AT(53)
AU(54)
AV(55)
AW(56)
AX(57)
AY(58)
AZ(59)
BA(60)
BB(61)


I Counted wrong but yeah my point was to include multiple characters into on transaction.

with single letters(to show a comparison)

A(0.00000001 BTC)
-
Z(0.00000026 BTC)

Multiple letters

AA(0.00000027 BTC)
AB(0.00000028 BTC)
....
....
AZ(0.00000054 BTC)
....
....
BZ(0.00000080 BTC)

You get the point...

[Xenland]

nah, you could just put 01,02,03... prefix or suffix to make sure correct order and send them all in one batch

You'd have to have less characters in the transaction or use big numbers like 1BTC then 2BTC to signifiy order I don't think its worth it

[Phinnaeus Gage]

I didn't take into consideration the time aspect, therefore I modified the same message. My intent was to create an encrypted message as simple as possible. With the new coded message below--first part is a question and second part is the answer--it now does not matter in what order the coins are sent or received.

.0711002
.0821111
.081102
.08402
.05911
.004111
.05202
.0351002
.036101
.087001
.0471101
.0131
.06602

.1140101
.1150002
.19211102
.1651
.135101
.1891
.179011
.1320111
.13811
.146011
.126001
.1431

Feel free to still try to crack it. For a clue or two, it's in American English using very common words. Basically, both are complete sentences, albeit short and to the point. As you can clearly see, I did not change the order of the letters, but I easily could have, and the resulting text would still be the same.

Most here know that I'm not a coder, and a lot of the dialog above is kinda over my head, though I get the gist of what's being relayed and am capable of figuring out all that's been posted on this thread to date.

~Bruno~