General Exalted Thread

[MJ12 Commando]

Are we actually doing this? Let's do this, despite discord's insistence I don't.

So, the thing with a mathematically perfect model is...it doesn't actually model reality. You have to actually, physically perform the process, which you can't do perfectly and without leaving behind evidence. Even in our reality, a 'perfect' encryption can be broken by listening to the sound the cpu makes while performing the algorithm in question. Your encryption is perfect, but there are infinite side channels you cannot actually cover assuming an opponent with unlimited information about the current state of the universe. (unless you are a Solar/etc. with a conceptually perfect encryption effect of course). That's not even getting into things like...the information on what someone encrypted with a one-time pad is still there, in the head of whoever wrote the message. Then there's things like the least god of the computer the information was encrypted on knowing it, or the gods of various secrets knowing it, or divination/investigation magic looking into the past as you wrote the message down.

Click to shrink...

This demonstrates a complete misunderstanding of what a one-time pad is. A one-time pad is merely a ciphertext that can be 'decrypted' into any posible message up to the total the length of the ciphertext. In fact, for most of the time a pad exists, the possible message may not exist. Sure, if the Solar gets access to the message in some unencrypted form or the cipherkey, obviously the Solar should be able to decode it. But we're not talking about that. We're not even talking about side-channel attacks.

We're talking about whether a Solar can look at a ciphertext, where the actual message the ciphertext will decrypt into might not even exist for years or decades, and somehow magically deduce the contents of the message, either at the moment of the message's creation (at which point you have an uninterceptible method of instantaneous communication over all of Creation), or even before the message is created (at which point any Solar can trivially create an EXORDIUM channel to themselves in the past).

 

[Elero]

You should be clearer about what you mean then, because

We're talking about whether a Solar can look at a ciphertext, where the actual message the ciphertext will decrypt into might not even exist for years or decades

Click to shrink...

That's really not how a one-time pad is actually used. You first, in advance, share the key, which is just random data. You don't share the message in advance and then create a key afterwards. Yes, you cannot decrypt this, because it isn't a message, it is a key. You just went and switched the two around. The thing you actually write and send to the person you want to communicate when you decide what to tell them is the message. Until you create the message, there is nothing to decrypt.

(Yes, A+B = secret message, so we can call either A or B the key and the other the message in terms of getting the message out of it, but one is random garbage you created first, the other one is what you created with knowledge of the first to create meaning.)

 

[Sanctaphrax]

I guess this is happening whether I participate or not. Might as well get involved.

I think magical effects can absolutely pierce one-time pads. Moreover, I don't think extracting The True Meaning from perfectly encrypted text is any more philosophically problematic than extracting it from plaintext.

No two people speak exactly the same language. Every written or spoken word is an imperfect representation of its creator's intentions, and there is nothing inherent in a word that connects it to its meaning. Communication is only possible when people share codes, and understand the rules of each other's language games.

Consider the following thought experiment. There is a sentence that is grammatically correct and understandable in two different languages. However, its meaning is entirely different depending on the language it is spoken in. That sentence is then similar to a one-time pad in that, without context clues, it is conceptually impossible to divine the correct meaning from the words in the sentence. The different languages, giving different meanings, are analogous to different cipher-texts giving different meanings.

A one-time pad is just a different set of communication rules. It's like a one-use constructed meta-language that serves only to transmit a single message. Piercing a one-time pad is really no different from instantaneously learning a very strange new language.

If we're accepting that Solar Charms can directly extract the intended meaning from any piece of text, then we're accepting that they can directly extract the intended meaning from any piece of text. At least on a philosophical level.

And as far as I can tell that acceptance is uncontroversial.

This has some funny implications, like extracting a totally different message from an exactly identical piece of text depending on the creator's intentions, but I'm inclined to regard that as a feature rather than a bug.

PS: I really don't think the perfect defense analogy holds at all. Solar magic overcomes perfect effects all the time; in fact, Heavenly Guardian Defense beating Accuracy Without Distance is a classic example. Primacy of defense is there to keep the combat system working, and a cipher isn't actually defending against an attack.

 

[Queshire]

We're talking about whether a Solar can look at a ciphertext, where the actual message the ciphertext will decrypt into might not even exist for years or decades, and somehow magically deduce the contents of the message, either at the moment of the message's creation (at which point you have an uninterceptible method of instantaneous communication over all of Creation), or even before the message is created (at which point any Solar can trivially create an EXORDIUM channel to themselves in the past).

Click to shrink...

Yes? A Solar should totally be able to do that. I mean, imagine a solarfied Sherlock Holmes. That type of bullshit would be right on theme with them.

 

[JayTee]

While I do feel that MJ12 Commando's concerns are legitimate, I also feel that they are edge cases that most GMs would handwave aside on the grounds of "I'm not dealing with that bullshit."

 

[MJ12 Commando]

While I do feel that MJ12 Commando's concerns are legitimate, I also feel that they are edge cases that most GMs would handwave aside on the grounds of "I'm not dealing with that bullshit."

Click to shrink...

I mean, 'a proper one time pad' is an edge case in the first place, since they're ludicrously fucking impractical. To make a one-time pad work, you need to have a completely secure key transfer of a completely random key that takes place under complete secrecy, with absolutely no reuse of the key in whole or in part. Basically, if you have the capability of actually securing a one-time pad, you can just... send someone the message in plaintext with pretty much equal security.

And yes, I think that if you can manage to successfully meet all these steps, every single one of which is an epic feat in Creation, you absolutely should be able to stop a Solar from going "I haz a charm I win." Which isn't even the same thing as preventing the Solar from decoding the message, even! Again, if the Solar can manage to eavesdrop on your transfer, or find some example of how your keys were reused, or you're not using some potent artifact or Charm that exists only to generate completely random outputs, you can exploit that flaw, sure okay. It's an extremely limited suggestion that a very impractical and complex method of encryption cannot be trivialized by the use of Charms, nothing more.

And I don't think it's unreasonable, especially since Exalted has been moving away, not towards, the idea that Exalts in general are these Conceptually Perfect Beings who basically can trivialize any non-conceptual problems.

Yes? A Solar should totally be able to do that. I mean, imagine a solarfied Sherlock Holmes. That type of bullshit would be right on theme with them.

Click to shrink...

Holmes wasn't completely infallible and honestly a Solar version of him shouldn't be either. Capable of impossible feats, sure. But there's a very significant difference between "capable of impossible feats" and "can literally trivialize anything."

Frankly, if you're playing a Solar spy, it probably means that you want to do spycraft once in a while, not just throw Win At Counterintelligence Prana at all your problems. Certainly I wouldn't expect someone playing a Solar brawler to want to just throw Win At Fight Prana at any problem they face, instantly winning the resulting conflict.

And if someone's invested such massive amounts of resources in keeping a message secret... I think that absolutely deserves to be treated as a big thing that takes more than the use of a single charm to bypass.

 

[Shane_357]

To successfully use something like a one-time pad in Exalted I think you'd need to put in some legwork on removing divine 'leaks', which could mean doing it in a place secured by an Adamant Circle working that ensures that the place is outside Fate and that things created within lack small gods.

 

[JayTee]

I mean, 'a proper one time pad' is an edge case in the first place, since they're ludicrously fucking impractical. To make a one-time pad work, you need to have a completely secure key transfer of a completely random key that takes place under complete secrecy, with absolutely no reuse of the key in whole or in part. Basically, if you have the capability of actually securing a one-time pad, you can just... send someone the message in plaintext with pretty much equal security.

And yes, I think that if you can manage to successfully meet all these steps, every single one of which is an epic feat in Creation, you absolutely should be able to stop a Solar from going "I haz a charm I win." Which isn't even the same thing as preventing the Solar from decoding the message, even! Again, if the Solar can manage to eavesdrop on your transfer, or find some example of how your keys were reused, or you're not using some potent artifact or Charm that exists only to generate completely random outputs, you can exploit that flaw, sure okay. It's an extremely limited suggestion that a very impractical and complex method of encryption cannot be trivialized by the use of Charms, nothing more.

And I don't think it's unreasonable, especially since Exalted has been moving away, not towards, the idea that Exalts in general are these Conceptually Perfect Beings who basically can trivialize any non-conceptual problems.

Holmes wasn't completely infallible and honestly a Solar version of him shouldn't be either. Capable of impossible feats, sure. But there's a very significant difference between "capable of impossible feats" and "can literally trivialize anything."

Frankly, if you're playing a Solar spy, it probably means that you want to do spycraft once in a while, not just throw Win At Counterintelligence Prana at all your problems. Certainly I wouldn't expect someone playing a Solar brawler to want to just throw Win At Fight Prana at any problem they face, instantly winning the resulting conflict.

And if someone's invested such massive amounts of resources in keeping a message secret... I think that absolutely deserves to be treated as a big thing that takes more than the use of a single charm to bypass.

Click to shrink...

I guess I agree with some parts and disagree with others. If you're playing a Solar spy, then no code should be beyond your ability to crack. On the flip side, "I haz charms, therefore I win" is dumb, I agree. I think a nice middle ground would be "I haz charms, therefore I winning is possible." The charm doesn't do it for you automatically, but makes it so that you can do it as like, I dunno, an extended action or a project or something. Maybe require a few magical cryptography resources you can use to boost your own innate code braking skills.

 

[MJ12 Commando]

I guess I agree with some parts and disagree with others. If you're playing a Solar spy, then no code should be beyond your ability to crack. On the flip side, "I haz charms I win" is dumb, I agree. I think a nice middle ground would be "I haz charms, therefore I have the option to win." The charm doesn't do it for you automatically, but makes it so that you can do it as like, I dunno, an extended action or a project or something. Maybe require a few magical cryptography resources you can use to boost your own innate code braking skills.

Click to shrink...

I mean just from the narrative implications alone I don't think this works. If you're going to expend massive amounts of resources to protect a small number of messages from being decoded if someone inadvertently gets access to either the message or the key without the other component in a fit of absurd paranoia, that probably means it's super fucking plot-important and shouldn't boil down to just 'an extended roll.'

And Solars already have charms that can make it much easier for them to decode the message. They have supernatural social ability that can help them interrogate said spy and get the message that way. They have supernatural detective skills that can help them piece together evidence from the spy's safehouse. They have supernatural stealth and awareness, so that they can spy on the messenger, and supernatural combat skills so they can incapacitate the messenger or the spy. They have supernatural investigation so that they can find the deaddrops being used for these messages and keypads.

And even if they miss one message but can decode others, they can make an educated guess as to what the message is from that context, an educated guess that is itself assisted with a Solar's supernatural skills at warfare and academics and politics.

Frankly, it seems to me that the actual argument is whether "is this message encoded by an absurdly complex and impractical method" is fluff or a plot hook. And some people think it's just fluff, so you should be able to throw a charm at it and win, while I and several others think it's a plot hook, at which point it's all about what you do after you run into this impossible-to-decode message.

 

[Sanctaphrax]

If you think the ability to automagically break codes makes things problematically easy for Solars, then the whole one-time-pad thing is a pure distraction. The problem has nothing to do with applicability to this one type of code.

Personally, though, I think that intercepting a coded message is difficult enough and narrow enough that letting a specialist Solar overcome the next step easily is fine. If you're actually gonna spend xp on codebreaking specifically, you deserve to break some codes.

PS: I think that giving blocks of texts actual inherent meanings is one way to make a world feel magical. In the real world that's a human abstraction that fails in the face of scientific reductionism; making it truly real provides a detail that's clearly different from our world and yet not weird enough to create any problems.

 

[JayTee]

I think there are three elements to this discussion

1: Is the impossible message fluff or crunch.
2: What should or should not be considered "impossible" for a Solar.
3: How important is it that the message remain impossible to decipher.

If you're a GM running a game with a Solar Spy, then you're going to want that spy to feel like a cool guy for being able to decipher an impossible message, on the other hand you don't want to make it too easy, otherwise why even bother saying its an impossible message?

A lot of it also depends on if the impossible message is intended to be decoded or not. If you intend for it to be decoded, then the only thing Encryption Breaking Prana is doing is speeding that process along and you'll get to where you always were intending to go in the first place. If its not intended to be decoded, then there needs to be a reason for it to exist beyond having the players decrypt it.

Ultimately this is coming down to a per table judgement call, which is kind of a boring copout, but considering how uncommon an unbreakable code is I suppose its not that surprising.

 

[Thelxiope]

No one should start with a god damn one-time pad.

Someone is trying to send messages to their spy in your GLORIOUS SOLAR KINGDOM. They start with codes or rudimentary ciphers - and then you break them. They seethe. They write their messages in languages from the far side of Creation, and you break them. Through cunning artifice, they disguise their messages as ordinary letters, and you break them. They send messages encoded as delicate ink paintings of birds and flowers, and you break them. They send messages written in an ink that can only be read when the least gods of the paper are bribed with specific offerings, and you break them. They write their secrets on green prayer strips wrapped around blessed scytales while offering prayers to Jupiter, and you break them anyway.

Fed up with this, they summon a demon which babbles numbers in Old Realm constantly and, I dunno, ruthlessly murders any mathematician it sees. They record some of its babbling on two scrolls. At Calibration, in secrecy enforced by prayers and offerings to Jupiter and Erembour and bribes to the local spirit courts, they pass one scroll on to their most trusted agent.

Then, and only then, after seasons or possibly years of foiling this person's plans, have they resorted to a cipher that you cannot break.

Which is when you subvert their agent; or summon their rng-demon and take advantage of some quirk it has to make it repeat the numbers it gave your enemy; or offer a bigger bribe to the curious little god from your enemy's local spirit court who stayed and watched and saw the key exchange. Intercept and burn their messages, so that they have to face you more directly once they realize none of their coded orders can reach their destination.

I cannot imagine how a one-time pad could reasonably appear except as the end of such an escalation - whereupon it's perfect to force this Solar to engage in other cool espionage behaviors.

 

[SunnySprings]

how the hell are modern era spycraft techniques showing up in a setting where most of the setting is barely out of the Iron Age, so to speak

 

[Shane_357]

how the hell are modern era spycraft techniques showing up in a setting where most of the setting is barely out of the Iron Age, so to speak

Click to shrink...

One word - Dynasts.

EDIT: Also, the All-Seeing Eye!

EDIT2: Iselsi too, I suppose.

EDIT3: And that's not taking into account the fucked up methods of the Lunars and Raksha!

 

[Kuciwalker]

So you also believe that Solar Melee should be able to break Heavenly Guardian Defense?

Since we're now declaring that Solar charms can defeat provably perfect effects (and non-flawed one-time pads are mathematically perfect effects), why stop at merely destroying basic setting assumptions like "communications and travel is hard and time-consuming" and "causality exists and you can't just change the past willy-nilly?"

Click to shrink...

This demonstrates a complete misunderstanding of what a one-time pad is. A one-time pad is merely a ciphertext that can be 'decrypted' into any posible message up to the total the length of the ciphertext. In fact, for most of the time a pad exists, the possible message may not exist. Sure, if the Solar gets access to the message in some unencrypted form or the cipherkey, obviously the Solar should be able to decode it. But we're not talking about that. We're not even talking about side-channel attacks.

We're talking about whether a Solar can look at a ciphertext, where the actual message the ciphertext will decrypt into might not even exist for years or decades, and somehow magically deduce the contents of the message, either at the moment of the message's creation (at which point you have an uninterceptible method of instantaneous communication over all of Creation), or even before the message is created (at which point any Solar can trivially create an EXORDIUM channel to themselves in the past).

Click to shrink...

Exactly as described, your assumptions are bad, and false, and they're especially bad and false in Creation of all places.

A "mathematically perfect" use of a one-time pad is a non-entity. It doesn't exist. It's not real. Characters are not abstract mathematical objects carrying out operations on other abstract mathematical objects; they are (fictional, yes) people in a physical world.

So you want to use a one-time pad? OK, first you need to generate the pad. This is the first thing you can screw up: what if you thought you were generating it randomly, but were wrong? Or what if you accidentally re-used an old pad? Or what if your counterparty had mistakenly already used the pad to encrypt a message?

Now you need to transcribe your message under the pad's substitution. Easy, right? Well, are you doing this by hand? And do you care about the message at all? Then maybe your emotions subtly bled through into your handwriting of the encrypted text. Picking up that sort of minute detail is classic Solar linguistics. Especially in Creation, where calligraphy is a Big Deal!

But maybe you were really clever and built some kind of special typewriter that could be fed the pad. Cool! But maybe the strength of your keystrokes betrays your feelings and thoughts in the same way. Or maybe you just screwed up and made a mistake in the construction of the machine. Or maybe you just didn't notice that you're introducing line breaks at word boundaries, leaking information.

Maybe you somehow isolated all of these specific issues and made sure they didn't happen. But it turns out that the timing of your message leaked critical information that allowed the enemy to glean its intent.

And then, finally, since we're in Creation and everything is a little bit animist, maybe a little bit of magic-spirit-stuff imprinted itself on the message and provided enough residue for the Solar codebreaker to work off of. Maybe there's a least god or something. The assumption of no side channels is absolutely terrible in Creation, where there is magic that can do stuff like "provide a video of this location several centuries in the past". Guess what, that's not a matter of Shannon entropy or anything, it's magic. Mathematical abstractions applied to Creation leak.

But even if we don't resort to animist mumble mumble, and even if you do say "I am going to encrypt this message to my friend Joe with a one-time pad", and even if you do provide an extensive, thirty-minute stunt explaining how you didn't fuck up in any conceivable way - OK, great? First off that's like a one point stunt because it sounds extremely boring. Second, the degree to which you actually pull off that stunt is given by your dice roll. If you roll poorly then, well, I guess something in your stunt was mistaken. (Heaven forfend you botch.)

If you want to say that the use of a one-time pad makes mundane codebreaking actions generally inapplicable, sure, that's fine. And if you want to argue for them giving a circumstantial penalty to the magic-enhanced Solar codebreaker, you know what, that's cool too.

And if you use an encryption Charm on your behalf, well, that's a whole 'nother ball game. Because that Charm probably stipulates its own rules for if and how it can be contested, and those just have nothing to do with whether or not you stunt this as a one-time pad or as steganographically encoding your message in the shape of the hearts you use to dot your i's.

Prediction: you're going to try to say "no no but we're talking about whether the Solar has some kind of completely abstract algorithm that operates on platonic, unphysical sequences of bits to transform them into messages" but, guess what, I was not talking about that, and you were responding to me, so suck it. Those things don't exist and cannot exist in play unless you subscribe to particularly fanservicey notions of the Wyld.

 

[MJ12 Commando]

Exactly as described, your assumptions are bad, and false, and they're especially bad and false in Creation of all places.

Prediction: you're going to try to say "no no but we're talking about whether the Solar has some kind of completely abstract algorithm that operates on platonic, unphysical sequences of bits to transform them into messages" but, guess what, I was not talking about that, and you were responding to me, so suck it. Those things don't exist and cannot exist in play unless you subscribe to particularly fanservicey notions of the Wyld.

Click to shrink...

But we were talking about whether the Solar has some kind of completely abstract algorithm that operates on platonic, unphysical sequences of bits to transform them into messages. Because guess what? You were talking about that the moment you jumped into the one-time pad discussion. Because that was the fundamental one-time-pad discussion in the first place. Whether a Solar could defeat such a theoretical construct.

Not whether such a theoretical construct even existed in Creation. Let me point you to the actual, original post which started the argument:

Incidentally, no, I disagree there. Discerning Savant's Eye cannot break a properly implemented one-time-pad or equivalent system, because an OTP is mathematically impossible to break and thus decryption attempts are inapplicable actions. The Solar doesn't even get to make an attempt, because he can't roll the dice to try - he Just Fails Outright. Properly implemented one time pads are not vulnerable to any form of decryption or breaking - they're a real life example of inapplicability.

Click to shrink...

The argument was always about whether the Solar "has some kind of completely abstract algorithm that operates on platonic, unphysical sequences of bits to transform them into messages." It was never about the practicality of such things, and in fact everyone acknowledged that one-time pads are pretty goddamn hard to make perfect. Furthermore, it was entirely about a single 2e charm in specific that no longer has the functionality in 3e, for reasons I'm not entirely sure about but I suspect are because 3e is significantly less willing to allow players to just skip processes and get to results.

So if you want to go around screaming that other people's assumptions are "bad, and false," perhaps you should check to see what their assumptions actually are in the first place?

 

[Kuciwalker]

"Thousands of pages ago I participated in one of the stupidest imaginable conversations, that could never possibly apply to any table or even to any plausible in-universe fiction; you in your remark to someone not me are therefore bound to honor all of the rules and conventions of this fanwank" hmmmmm.

 

[Sanctaphrax]

A lot of it also depends on if the impossible message is intended to be decoded or not. If you intend for it to be decoded, then the only thing Encryption Breaking Prana is doing is speeding that process along and you'll get to where you always were intending to go in the first place. If its not intended to be decoded, then there needs to be a reason for it to exist beyond having the players decrypt it.

Click to shrink...

I would really prefer not to decide ahead of time whether players are gonna succeed at what they're doing.

how the hell are modern era spycraft techniques showing up in a setting where most of the setting is barely out of the Iron Age, so to speak

Click to shrink...

Well, that brings us back to my earlier question. How advanced is Creation exactly?

I've got a modern math degree, and I'd be inclined to try a one-time pad if I was up against supernaturally skilled codebreakers. And not as the end of an escalating series of codes, more or less right away. It's actually implementable by hand, which gives it a real edge on many modern codes.

I tend to assume that Exalted mathematicians are more capable than me, despite my modern education. Presumably the First Age had some pretty serious universities, and mathematical knowledge survives more easily than many other kinds of lore.

But on the other hand we've got people saying...

...I cannot imagine how a one-time pad could reasonably appear except as the end of such an escalation - whereupon it's perfect to force this Solar to engage in other cool espionage behaviors.

Click to shrink...

 

[Kuciwalker]

Guess what, I found the original post in the ancient kerfuffle:

Incidentally, no, I disagree there. Discerning Savant's Eye cannot break a properly implemented one-time-pad or equivalent system, because an OTP is mathematically impossible to break and thus decryption attempts are inapplicable actions. The Solar doesn't even get to make an attempt, because he can't roll the dice to try - he Just Fails Outright. Properly implemented one time pads are not vulnerable to any form of decryption or breaking - they're a real life example of inapplicability.

Click to shrink...

So, unless "properly implemented" means "literally physically unrealizable" nah, we're happily in the space of "but actually side channels always exist, even if they're not practically exploitable by mundane actions".

(Not to mention that in 2E charm paradigms making an inapplicable action applicable was a standard trope!)

 

[JayTee]

I would really prefer not to decide ahead of time whether players are gonna succeed at what they're doing.

Click to shrink...

Thats fair. In general I'm okay with a GM telling me that an ability I put XP in to wont work this time because of plot reasons. You juggle a lot of things at once when you GM, so I'm willing to give it a pass if a GM just straight forgot I had that ability and didn't account for it, or maybe had something really cool planned and my ability would ruin it and I trust them that the thing actually will be pretty cool so I'll agree not to use it this time.

I generally require that the GM also admit they messed up and I need to pretend that my ability doesn't work this time so that we're all on the same page, but so long as it doesn't happen too frequently I'm willing to work with them.

 

[Thelxiope]

But on the other hand we've got people saying...

Click to shrink...

I mean, I don't have a strong math background at all, for starters; and was being somewhat hyperbolic.

But, um, and I apologize if this wasn't clear -

I mostly don't care about the question of cryptographic techniques in Creation except as relates to a hypothetical Exalted game hypothetical-me could be running? And so in that context, if I had a Solar player who was big into spying and codes and other Night stuff, the idea of a cipher they can't bypass with a single charm activation is an appealing tool, and possibly something said player might also find interesting and fun to game?

And so it makes sense to me that, if I were using cryptography and secret messages and such in my game, I wouldn't start with a one-time pad; because it's the final boss of ciphers and therefore should be deployed only after a suitable period of cat and mouse. Dropping it too early seems to me like it would ...kind of be unnecessarily mean? Like, if I had a player that liked the whole spy-games thing, I think it would be better to accommodate that and let them have that moment where they can just go "haha nope" to a cipher, or ciphers; let them be cool and do the whole problem-bypassing thing Solars do; and then escalate to "no, you have to work around this; attack the endpoints or the flaws in its implementation. You can't just crack it."

 

[Romv]

Isn't there a forum for this stuff?
One Solar Exalted vs One One-Time Pad, Bloodlusted

 

[grommile]

Split the difference.

A correctly used one-time pad cannot be defeated by cryptanalysis, because cryptanalysis fails when the key is as long as the ciphertext; if all you have is the ciphertext, I would rule that codebreaking Charms are inapplicable against an OTP-encrypted message.

If you get hands on a sample of the actual keystream material being produced by a one-time-pad generator, on the other hand, then I'm happy for Twilight Jerzy Różycki to flex their glorious golden god-brain to identify the biases in the generating mechanism and predict/reconstruct the rest of that keystream, even if the sample obtained would be too small to mount a normal analytical attack.

So, y'know. Your Night now has an opportunity to pickpocket or burgle the suspected enemy agent to get you a scrap of their pad.

 

[MJ12 Commando]

"Thousands of pages ago I participated in one of the stupidest imaginable conversations, that could never possibly apply to any table or even to any plausible in-universe fiction; you in your remark to someone not me are therefore bound to honor all of the rules and conventions of this fanwank" hmmmmm.

Click to shrink...

I mean, @Sanctaphrax was referencing a few pages of discussion about the theoretical limits of a single Solar charm. Nobody asked you to jump in and declare that everyone on one side or the other of the argument was wrong in the most condescending possible way, and then double down on the hostility when people pointed out that it was a reference to a prior discussion.

 

[Jon Chung]

Solar Linguistics can definitely break one-time pads and it's not even a matter of "lulz perfect effects"; the argument that one-time pads are completely unbreakable / even Solar charms can't cause logical contradictions rests on very bad and false assumptions that are particularly bad and false in Creation.

Click to shrink...

You cannot break a message encrypted with a one time pad with a "decode this" charm - there is nothing to decode, it's an instruction to look something up in some other document, and you don't have the other document. Even if you have a perfect decoding charm, this does nothing to solve your problem - you understand the message just fine, but the thing's you're supposed to look stuff up in is not there, what are you gonna do?

A charm which explicitly let you pull information you don't have out of nowhere would work as we know that this is possible by the fact that Perfect Mirror exists (your perfect disguise can act correctly in ways you did not observe even if there is no way to extrapolate), but the charm in question would have to explicitly mention that before I would allow it to work, like Perfect Mirror.

Kek.

So, a charm that said "Perfectly comprehend what this message was supposed to communicate to its intended recipient" would work. A "Perfectly decode this message" charm would not work and makes no sense here, the same way looking at someone's pre-arranged signal flag without the corresponding reference of what colour flags mean what orders to execute means nothing to you since there's nothing to decode (the message is clear: the flag is red with a hexagonal icon, you just don't understand what it means without the signal book).

Note: I am assuming the OTP is executed correctly and the key is longer than the message. A "decode this" charm would work if this was not the case.