Flashing back, you don't know if Adventure! can be found on this planet or that - Use your dowsing rod!

Dowsers are people looking for something; usually, they find it. The determining factor in [i]what[/i] they find is dependent on what they're satisfied with, and what they're looking for. The trick is being absolutely committed to finding it, no matter what - otherwise "not found" is the steady-state and you don't find it.

Core Value:
Dowsing for "____" - This is your determination to find ____, and what it is you're looking for. Use it defend from giving up your dowsing rod, to give up looking, etc. It's also somewhat representative of the severity of what you're looking for; a low CV in "Dowsing For Adventure" means you're happy with simple adventures; a high one represents the adventures that change worlds.

Benefit: (Suggested)
Divide your CV by 2, rounding down. You can ask your GM which way to go, to find that thing, that many time a session. If you take a Theme Descriptor meaning something like "Found it!", you get a free twist to spend on it.

No future event has happened, so a "message from the future" functionally appears ex nihilo. Conservation of mass and energy notwithstanding, if you assume time travel than you are assuming that there are points in time where there is more mass energy than there was a moment before (obviously, looking at it more hollistically clears the apparent discrepancy with conservation).

When the Transcendentals receive messages from the future, they are either receiving a message that states that such-and-such happened and are powerless to alter that course of events (single timeline - if you travel to the past you will not be changing the present because the past already has you in it), or they are receiving multiple messages from multiple futures and each one (hypothetically) gives insight as to how those futures came into being (multiple timelines - if you travel to the past you will create an alternate history, and so when you return to the present you will not be in the same timeline as you were when you departed).

Even with the alternate history theory, there is some debate as to "how much" change is needed to produce an alternate timeline, but that discussion is largely immatereal. SA virtually demands an alternate history outlook, and I think it is in the confusion between alternate history and single timeline that a lot of the question of how Transcendentals and retrocommunication work comes from.

Thus, functionally, what's happening is that the Transcendental receives information from multiple timelines. If the messages are from far enough in the future, then few of them will have any real "authority" (higher possibility of occuring) than the others. Thus, you get several equal-weight possibilities. This may well give you a fairly clear start point for your actions (in 70% of the outcomes that you want, a certain action was performed tomorrow, and that action has only a 7% chance of leading to generally bad outcome). This could also contribute to the disjointed nature of Transcendental behavior, as every day they judge immediate actions from long-term results, all without any heed as to what they're going to do the day after tomorrow. ("Today" and "tomorrow", of course, are being used in a metaphorical sense in this case.) Whatever immediate action results in the highest net utility is the action taken - in case of ambivalent, the highest gross utility is used. Essentially, the "operating system" for the Transcendentals is almost entirely dedicated to advanced felicific calculus.

This does not now necessarily revert to Timeline A. The universe is probabilistic; Quantum Mechanics tells me so. Things will not necessarily happen the exact same way again, including what happens inside someone's brain and the decisions they make.

I generally consider any system that requires meta-time to be inherently flawed. Now, it’s a common flaw (pretty much every well thought out time travel system in fiction requires meta-time, including C°ntinuum’s). But, having meta-time implies that the system is incomplete. There’s likely to be meta-timetravel, which brings us back to the original problem, ad infinitum.

When I read The End of Eternity it had this problem. It left me wondering if Asimov had made that common mistake. What surprised me is that Asimov had thought of it, and it was the second half of the plot. Only Asimov story I’ve ever read where I figured out the trick ending a third of the way through.

And yes, the "meta-time" thing is why some of my replies in this and other threads have sounded weird. It's almost like a second time variable.

Let's say that I'm a human with an oracle, instead of a T (and thus less interested in metaversal goodness and more in what happens to “me” personally).

Virtual Timeline A:
I have a bad day and get fired at work. So I send back a message to tell me not to lose my temper. Timeline invalidated.

Virtual? Timeline B:
I read the message and don't make that mistake again. I don't bother sending back the original message. Now, in theory this timeline is invalid and thus virtual. But there's no moment when I can say that the timeline is “now” invalid (because I could always decide to send the message later, or someone else might). So I continue on with my life in a theoretically virtual timeline, having benefited from another one without paying any sort of bandwidth cost. Eventually the universe ends, at which point the timeline is proovably virtual, so one can argue that at that point it goes away.

This is mostly to point out that the “and then I send a message, invalidating my timeline” is a form of meta-time. Realistically, there's no single point in time where you should be able to say “this timeline is now invalid”. It either was invalid all along, or it is never invalid. But we're (arbitrarily) saying that it goes invalid when a message does or doesn't get sent back.

You can also tack on more information when you send stuff back, such as the outcome of certain plans of action so that they can be avoided. Eventually you want to send the most efficient message possible, but it might take a few billion iterations to get it right.

Also, don't unsend seemingly useless information.

Thought experiment to demonstrate how your system works:
Virtual timeline A:
Server Violet is destroyed. I send back a message 3 days to tell myself to send inspector team 1 to prevent the destruction and tell me what happened. Sending this message invalidates the current timeline (curiously, the moment of invalidation is when the first bit is sent, do I'd better get my timing perfect).

Virtual timeline B:
I get a message from myself to send inspector team 1 to prevent the destruction and tell me what happened, and I do so. Three days later Server Violet still exists, but half the servers were destroyed and there is irreparable damage to Stored society (by quick psychohistorical analysis). I send back a message to myself to send inspector team 2, as they would be better qualified to deal with the problem. Since I never send the original message (from timeline A) and never received the message I am sending now, this timeline is also invalidated.

Final Timeline:
I get a message from myself to send inspection team 2 to deal with Server Violet, and things work out well (better than I can expect using any other team). I consider this to be good enough and choose this timeline to be final. I send back the exact same message as I received, reinforcing this timeline.

How to create an irreparable paradox using this model:
Virtual Timeline A:
Over the next two hundred years, the Mechanicans pick up the CV Sameness. It keeps the civilization from falling apart, but it also stagnates the society, destroying any potential for growth. I send back a message (three hundred years) to tell myself to prevent this. Timeline A is now invalid.

Virtual Timeline B:
I received a message a hundred years ago to prevent the Mechanicans from picking up the CV Sameness and things are going well on that front. I'm using Logician conflict to keep the Mechanicans on their toes. The Mechanican world of Wrench comes under a massive Logician metatech attack which threatens to destabilize the entire region. I send a message back a week to tell myself to tell myself to send inspection team 1 to go to Wrench and prevent this attack. Timeline B is now invalid.

Virtual Timeline C:
I still received that message a hundred years ago about the Mechanicans, and I was still using Logicians to fix it. Now I also receive a message to send inspectors to Wrench to prevent the Logician metatech attack. So, I send the inspectors to Wrench. This works well, so I send back the same message. One hundred years later I find that the Mechanicans were at no risk of picking up the CV Sameness (meaning that the resources spent on fixing the problem were waisted). This is largely because of the failed Logician attack. So I don't send back the message from timeline A, invalidating this timeline.

Virtual Timeline D:
I never received a message to remove the Mechanican CV Sameness, so I don't involve the Logicians, so there is no attack. This timeline is identical to Virtual Timeline A.

This creates a loop where we have three equally valid timelines each supporting the others. None becomes the Final Timeline. This sort of situation becomes increasingly likely as more and more temporal communication is used. Worse, it is virtually guaranteed if two or more transcendental entities are fighting.

The first is the final timeline is inconsistent. You make a decision based on the advice of lots of virtual timelines, which requires that their messages are sent to you, but the act of making that decision erases the information, which means that you couldn't make the decision in the first place.

Second: if the only message that exists at the end is the one you send from the path you eventually like, you could always choose not to send back any messages when you like the outcome. As such, buffer space is only used in virtual timelines (the ones that don't come to be). In any real timeline, the entire buffer is always clear. (in other words, the inconsistency of the timeline can be abused to create information from nothing; without paying any cost for it whatsoever).

Which also begs the question, how do those Ts deal with the infinitude of signals coming back?

They choose a particular course of events and do their best. They then send back a note to their prior selves saying, "Here are some suggestions for your life." Their prior selves then do not follow that course of events, but rather choose a different path and do something else. They then send a note back, and their prior selves do something different. In this way, from an external viewpoint they examine possibilities one at a time.

Imagine driving down fifty roads and exploring them all, and then being able to actually have gone down the road you liked best and not needing to go down the others. The Transcendentals tell themselves exactly what to do because they know it's going to work. They guess and the guess is right because it's the one they chose.

I'm not sure if that makes sense to you guys,

Now, you're in future B. So, you read A's message, write B's, and read C's. At the end, you take all three (with your message overwriting whatever was in the B slot before) and send them back. Vuala! Instant internally consistent timeline with quantum loop information. Of course, you could just be talking to yourself and two random string generators. Which brings you back to the "independent confirmation" problem.