The Orion's Arm Universe Project Forums





Hi, a timeline question, and some math about encryption
#21
(05-18-2018, 05:45 AM)Crossroads Wrote:
(05-17-2018, 02:18 PM)Drashner1 Wrote: 5) You could have these events taking place in another galaxy, with one or more xenosophont civs involved and Terragens civ observing the whole thing via the Argus Array and recording it as some number of EG articles.

I concur with your post; I just wanted to quick set forth the idea of doing essentially what you mention here in point 5 when I write my proposed article on the Local Group. What would be observed in part or all of some xeno civ undergoing TS? I imagine this - the AA sees, in just a handful of years or even less time, a large chunk of a HEEC, or even an entire HEEC, stop emitting energy, with no energy bursts consistent with a conflict. Modo observers speculate that this is TS.

Does this sound right?

That certainly sounds workable. The only caveat I can think of might be that in considering TS more closely we might come up with some 'representative effects' that we would want to also be observed. But obviously its too early days to know what such would be or if we will come up with them. And if/when we do we can always retcon them to the Local Group article if it is created first.

No worries all aroundSmile

Todd
Reply
#22
So, rather than doing lots of replies to individual points in your various posts, I'm going to summarize points and my replies here to save space (And hopefully time). In no particular order:

a) Regarding an article on encryption in the setting - as we discussed earlier this would be great as long as we could make it at least reasonably layperson friendly. An added bit that I forgot to mention the first time is that it would also be quite workable to add references and links to the Design Notes and Additional Information sections of the article. These can be as technical as you want and can include PDFs hosted on our content management system, links to websites or other online articles, and links to videos on YouTube or similar resources. Graphics and diagrams within the article are also welcome. Also, I'm a bit rusty on the methodology, but the EG can also host formal equations and math formulae, although we would prefer to have enough layperson friendly content to make those sorts of things icing on the cake rather than the main event.

b) In various parts of our discussions you've mentioned 'technotelepathy' in terms of the transapients or archai being able to read sophonts minds. We do have such tech in the setting, both in terms of 'infecting' a sophont brain with replicators that build a non-destructive uploader and then make a copy of their mind which is then read out and as a common feature of most DNI systems. But I'm wondering if you're actually referring to another ability in the setting - deductive telepathy - in which a transapient essentially psychologically profiles a lower S-level mind to such a degree that they can predict/model what they are thinking to an extremely high degree of accuracy - so high that it often comes across as if they are reading the lower S beings mind.

Mainly just trying to make sure we are on the same sheet of music in our discussions. Also, regardless of which type of 'mind reading' you are referring to, could you clairify what you mean by this? I'm not entirely sure I'm seeing where you're coming from here. Thanks!

c) Re baseline friendly terminology - I think that once we have an article created (or an existing article updated) with a consistent terminology and clear definitions of what each it is, it won't take long for those with an interest to pick it up and start using it comfortably. Based on past experience, whether that terminology is 'quantum level 0) through 3)' or 'completely classical, quantum co-processor, etc.' is less important than people feeling that they have a reasonably clear understanding of what each level is and is characterized by and is capable of. At that point, they will probably start incorporating it into both forum discussions and EG articles, as well as potentially developing alternative terminologies with the idea that different systems of notation and terminology and 'schools of thought' exist in different places in the setting.

d) Re creating (or not creating) large scale quantum entanglements in hot and messy environment and/or gaming such systems - this all sounds cool. For an added thought - I wonder if the archai could use chaos wand technology to essentially create stable zones in some environments for QC to take place in?

e) Re magmatter computronium - I imagine that this type of hardware (quantum or classical) is used in conditions where its extreme durability is considered to outweigh its mass. IIRC magmatter is superconducting up to extreme temperatures (the surface of the sun is a practically absolute zero vacuum as far as it is concerned). Also, mag-computronium would be able to engage in processing at speeds/energy levels that would exceed the strength of chemical bonds and tear conventional matter processors apart. It's sort of like having mini-neuron star matter at your disposal - although not held together by gravity.

f) Re S4 magmatter minds - Such might be possible, but would need to be distributed into some kind of swarm of units using comm-links. A mass of magmatter smaller than a lunar mass will not collapse into a black hole. But anything larger pretty much is guaranteed to unless properly spread out. Of course, such a swarm could consist of lunar masses of magmatter inside void bubbles and then great merriment could ensue...

g) Re plasma processors - the current numbers on these are not as concrete as (for example) those for Ultimate Chips. They are a mix of BOTE, what seems plausible, and what came from an article I read some years back about the ultimate limits of computation by Seth Lloyd. The original article I saw was in Ars Technica and I'm not able to find it now. But I was able to find a paper by Lloyd on the same subject (see attached). At the time, I thought the article was interesting (insofar as I could understand it), but not really relevant to OA or much else. However, I then came across an SF story (Cusp) which is an interesting semi-hard SF story with lots of big ideas. It makes a passing mention of a plasma processor computer made from a very small fusion reactor. And then somewhere in all that magmatter was introduced to the setting.

Putting this all together (and bearing in mind I don't entirely remember all my thinking at the time), I believe I assumed a plasma temp of around 100,000 degrees (IIRC the 'ultimate laptop' is described at running in the billion degree temp range, so I assumed a 4-5 orders of magnitude less capable system to be conservative). If the UL is capable of approx 2e50 operations per second, then a system 10,000x less capable could do 2e46 ops and one 100,000x less capable can do 2e45. An S3 is described as processing 2.98e38bps and storing 1.64e45bits. So allowing for even more efficiency and capability losses to be even more conservative, fitting an S3 into a kg of plasma processor seemed reasonable. IIRC I also assumed 1-bit per atom data storage to be conservative.

For an S4 we are tentatively assuming 3 Earth masses of plasma processor in the core of a Jupiter Brain (meaning a gas giant converted into computronium) using a 3d mesh of magcarbon nanotube fitted with tiny conversion reactors that suck in part of the gas giants mass and convert it to energy to energize a plasma that is held along the lines of the mesh by the massive magnetic fields and controlled by them to act as a processor. And the metallic H2 surrounding the core is (somehow - maybe by introducing impurities?) converted into extreme high temp superconducting processor for memory storage and secondary processing. Hm. Could metallic H2 support quantum computing or data storage? Extreme high temp but also under extreme conditions (pressure and gravity) that might hold it very stable.

For S5 we're currently toying with using a 10 solar mass star that has its entire volume engineered into plasma processors, much of them operating at much higher temps than S4 can manage.

All this is obviously very speculative and not as worked out as we would like yet. Any thoughts or computational assistance you could provide in these areas would be much appreciated.

h) Re a taxonomy of quantum computing - Sounds good to meSmile Either as its own article or by expanding the existing taxonomy article sounds good.

i) Re reversible computation - this has come up in discussion on the forum before (waste heat as a limiting factor gets talked about a lot here), but this is again an area where none of the currently active members have been able to claim much solid expertise and so we've not known quite what to do with it or what limitations it has. It somewhat comes across as a bit of a free lunch to some folks and so there is a tendency to be leery of it out of the fear that we might invest a lot of creative time on it only to have the other shoe drop and then have to undo a lot of cool stuff. Again, if you can provide some layperson friendly descriptions/EG entries/references we'd be happy to add this to the EG and the OA setting.

j) Re entropy piping - sounds cool - another potential EG article I would thinkSmile That said, in past discussions there has been mention made of something called 'bit rot' - apparently having to do with the entropy of computation and getting you waste heat and causing other limits on capability. Can you explain what that is and how would it factor into all of this?

k) Re running cold vs running hot computronium - sure. While we so far haven't delved into this issue in great detail in the setting, it seems logical that Terragen civ has a vast number of different designs for computronium, each optimized for different environments, computational or storage needs, and probably some large number of other factors that are all dealt with as a matter of routine whenever some new computing installation or infrastructure is being created. So you have one set of designs for when you want to operate in high energy environments and others that are designed to work in deep space at near absolute zero, and other designs for every environement in between. And it all meshes together seamlessly from the end users point of viewSmile

l) Re quantum channels - it sounds like before we're done we should have an article on this, including the applications. Actually, I've occasionally toyed with the idea that OA should have an entire section on 'quantum tech', a whole family of technology based on various quantum weirdness - but don't have a sufficient understanding of what that might look like, what the capabilities and limitations would be, etc. Also so much going on haven't had time to learn much, and a lot of the into that is out there is more technical than I can manage.

Ok, I think that about covers everything I wanted to respond toSmile

Looking forward to hearing your thoughts on the matter(s),

Todd


Attached Files
.pdf   Ultimate_Limits_Computation.pdf (Size: 187.76 KB / Downloads: 4)
Reply
#23
A side note: as I've seen it used, "bit rot" is a facetious term for the phenomenon that programs that worked perfectly before seem to "rot" and suddenly break eventually, even though what's actually happening is the system they're run on becoming newer.

My lifelong goal: To add "near" to my "baseline" classification.

Lucid dreaming: Because who says baseline computronium can't run virches?
Reply
#24
On the quantum infotec stuff, I've been writing up a first draft of an interlinkled set of articles -- it's given me an excuse to do some more reading on quantum computing and quantum algorithms, and clarify some of the stuff I half-knew, which has been fun. (Trying to explain quantum mechanics to non-physicists -- and occasionally to graduate physics students -- is hobby of mine :-) ) Once I have them in draft form, I'll start another thread for them in the appropriate section. They have a few minor retcons for various other articles, but they're generally of the form "the silk god/the great whale/lum/vecs/muuh archives/whatever is cooler than you previously thought because (nifty quantum infotech)". Similarly for the article on computational complexity and encryption (which will be heavy on dubifiers and "high level transapiences assure us that it's safe to...") -- it actually makes sense to do that second, since it refers pretty heavily to the first, or maybe I'll bundle them.

On technotelepathy vs deductive telepathy -- you're right, I had the terminology mixed up. Technotelepathy on quantum sophonts is a whole different question (basically, if you try it while they're being fully-quantum (which requires quantum infotech), they pass out, or at last go mentally blank, due to the No Cloning Theorum for quantum information -- taking piece of quantum information from their head literally removes it, they no longer have it; or if you use classical infotech you're basically a noise source, so either you learn nothing or you leran almost nothing and their algorithm breaks spectaculalry. So assuming decent autosentience their brain is as as protected from eavsdropping as a quantum encrypted channel, though there are subtleties around what I'm calling "classical checkpointing" or when quantum computronium is running a classical algorithm or interacting with the classical world -- the short version is, hack their backup system instead, it's far easier -- I'll address both in my draft writeup)

On the future stuff, I thought of a possible loophole assuming we want to keep the main setting in 10,600. Could we have the article be derived from multiple copies of the EG from inside the "Orion's Arms" virch as 'leakage from the distant future' in via some obscure Roman loop phenomenon or whatever (or some similar solution) that seems consistent enough across multiple parallel histories and significant enough in content that the real EG editors figured it was worth publishing it anyway in 10,600 (with a detailed warning paragraph at the top)? So for the article on the Chaos, for example, we'd both keep the current article, and add a "possible future version of questionable provenace" as well? That Orion's Arms virch is presumably a prediction of the future presumably run by some archai, possibly up to an S6 (in which case it's probably Yesod), possibly even derived from one-or-more Tipler oracle runs -- so it's more of "a plausible prediction" than "a guaranteed outcome", and has trustworthiness questions (particulary around S6 -> S7 ascentions, which is a lot of what I want to talk about). And it gives anyone else who feels like writing future material an excuse -- plus they don't even need to be fully mutually consistent with what I wrote (though they should probably have similar overall directions, like "S7 is hard"), because the "Orion's Arms" virch has parallel histories.
Reply
#25
(05-19-2018, 09:26 AM)Roger Wrote: On the quantum infotec stuff, I've been writing up a first draft of an interlinkled set of articles -- it's given me an excuse to do some more reading on quantum computing and quantum algorithms, and clarify some of the stuff I half-knew, which has been fun. (Trying to explain quantum mechanics to non-physicists -- and occasionally to graduate physics students -- is hobby of mine :-) ) Once I have them in draft form, I'll start another thread for them in the appropriate section. They have a few minor retcons for various other articles, but they're generally of the form "the silk god/the great whale/lum/vecs/muuh archives/whatever is cooler than you previously thought because (nifty quantum infotech)". Similarly for the article on computational complexity and encryption (which will be heavy on dubifiers and "high level transapiences assure us that it's safe to...") -- it actually makes sense to do that second, since it refers pretty heavily to the first, or maybe I'll bundle them.

Sounds good. As an aside, this sounds like it might impact the Polarizers article as well.

(05-19-2018, 09:26 AM)Roger Wrote: On technotelepathy vs deductive telepathy -- you're right, I had the terminology mixed up. Technotelepathy on quantum sophonts is a whole different question (basically, if you try it while they're being fully-quantum (which requires quantum infotech), they pass out, or at last go mentally blank, due to the No Cloning Theorum for quantum information -- taking piece of quantum information from their head literally removes it, they no longer have it; or if you use classical infotech you're basically a noise source, so either you learn nothing or you leran almost nothing and their algorithm breaks spectaculalry. So assuming decent autosentience their brain is as as protected from eavsdropping as a quantum encrypted channel, though there are subtleties around what I'm calling "classical checkpointing" or when quantum computronium is running a classical algorithm or interacting with the classical world -- the short version is, hack their backup system instead, it's far easier -- I'll address both in my draft writeup)

Sounds good. Just so I'm clear on this - technotelepathy (directly reading or making a copy of a mind-state) on a quantum sophont is basically unworkable (with the exception of their backup), but deductive telepathy (essentially psychological profiling cranked to 11) is perfectly doable - correct?

(05-19-2018, 09:26 AM)Roger Wrote: On the future stuff, I thought of a possible loophole assuming we want to keep the main setting in 10,600. Could we have the article be derived from multiple copies of the EG from inside the "Orion's Arms" virch as 'leakage from the distant future' in via some obscure Roman loop phenomenon or whatever (or some similar solution) that seems consistent enough across multiple parallel histories and significant enough in content that the real EG editors figured it was worth publishing it anyway in 10,600 (with a detailed warning paragraph at the top)? So for the article on the Chaos, for example, we'd both keep the current article, and add a "possible future version of questionable provenace" as well? That Orion's Arms virch is presumably a prediction of the future presumably run by some archai, possibly up to an S6 (in which case it's probably Yesod), possibly even derived from one-or-more Tipler oracle runs -- so it's more of "a plausible prediction" than "a guaranteed outcome", and has trustworthiness questions (particulary around S6 -> S7 ascentions, which is a lot of what I want to talk about). And it gives anyone else who feels like writing future material an excuse -- plus they don't even need to be fully mutually consistent with what I wrote (though they should probably have similar overall directions, like "S7 is hard"), because the "Orion's Arms" virch has parallel histories.

Hm. Interesting ideaSmile I just read back through the Orion's Arms virch article and what you suggest seems workable on general principles. We might need to tweak the current article a bit here and there, but that seems workable enough. We might also want to do some tweaking to put greater emphasis on the controversial nature of the virch as a possible bottleworld (we've cranked up the negative view of these a bit since this article was first created) and possibly either power up Query or eir mysterious master to support the kind of computronium it would take to the level of simulations described and then the kind of stuff it would take to simulate an S7 even remotely. Not uber detailed stuff really, we can crank up the mystery and conspiracy theory element a good bit on this one, but enough to make it plausible that such high level toposophic events could be happening.

On a related note - would you like to present this as something that is has been an ongoing thing (just part of the Orion's Arm virtuality since it started doing its thing? Something that has a definite starting point, but that point is in the past? Or something that has just started to happen/be reported in the 'now' of 10,600AT (so it's making the equivalent of the nightly news)?

Todd
Reply
#26
(05-19-2018, 09:26 AM)Roger Wrote: On the quantum infotec stuff, I've been writing up a first draft of an interlinkled set of articles -- it's given me an excuse to do some more reading on quantum computing and quantum algorithms, and clarify some of the stuff I half-knew, which has been fun. (Trying to explain quantum mechanics to non-physicists -- and occasionally to graduate physics students -- is hobby of mine :-) ) Once I have them in draft form, I'll start another thread for them in the appropriate section. They have a few minor retcons for various other articles, but they're generally of the form "the silk god/the great whale/lum/vecs/muuh archives/whatever is cooler than you previously thought because (nifty quantum infotech)". Similarly for the article on computational complexity and encryption (which will be heavy on dubifiers and "high level transapiences assure us that it's safe to...") -- it actually makes sense to do that second, since it refers pretty heavily to the first, or maybe I'll bundle them.

Sounds good Smile I'm interested to read about how QC may work in OA and what the possible applications would be (particularly every-day ones).

(05-19-2018, 09:26 AM)Roger Wrote: On the future stuff, I thought of a possible loophole assuming we want to keep the main setting in 10,600. Could we have the article be derived from multiple copies of the EG from inside the "Orion's Arms" virch as 'leakage from the distant future' in via some obscure Roman loop phenomenon or whatever (or some similar solution) that seems consistent enough across multiple parallel histories and significant enough in content that the real EG editors figured it was worth publishing it anyway in 10,600 (with a detailed warning paragraph at the top)? So for the article on the Chaos, for example, we'd both keep the current article, and add a "possible future version of questionable provenace" as well? That Orion's Arms virch is presumably a prediction of the future presumably run by some archai, possibly up to an S6 (in which case it's probably Yesod), possibly even derived from one-or-more Tipler oracle runs -- so it's more of "a plausible prediction" than "a guaranteed outcome", and has trustworthiness questions (particulary around S6 -> S7 ascentions, which is a lot of what I want to talk about). And it gives anyone else who feels like writing future material an excuse -- plus they don't even need to be fully mutually consistent with what I wrote (though they should probably have similar overall directions, like "S7 is hard"), because the "Orion's Arms" virch has parallel histories.

The archai's forecasts generally aren't totally available to modos (or anyone else really) but it's certainly possible that there could be a future modelling effort that used nuggets of information passed down and that the scenario you propose is the most terrifying, notable or perhaps even the most likely (so long as there's some wiggle room in it).
OA Wish list:
  1. DNI
  2. Internal medical system
  3. A dormbot, because domestic chores suck!
Reply
#27
Roman loops can't be used for time travel backwards in time, at least as far as any modosophont is aware. There are probably a few belief systems/religions/conspiracy theories that propose that the Highest Archailects can affect the past, but these belief systems do not have any proof- in some models of time travel, there could not be any credible proof of reverse causality, since each altered timeline would be separate and out of communication with every other one.
Reply
#28
(05-20-2018, 02:18 AM)stevebowers Wrote: Roman loops can't be used for time travel backwards in time, at least as far as any modosophont is aware. There are probably a few belief systems/religions/conspiracy theories that propose that the Highest Archailects can affect the past, but these belief systems do not have any proof- in some models of time travel, there could not be any credible proof of reverse causality, since each altered timeline would be separate and out of communication with every other one.

Of course - no FTL or time travel in OA. But Orion's Arms is a virch (or more exactly, a media feed from the inside of a closed virch) which (appears to be) exploring what could happen if a single-wormhole Roman loop could actually be stabilized with S:5 technology, and then allowed travel into the past or future parallel alternative histories (which branch off when the Roman loop forms, and diverge thereafter -- so you can't go earlier than that). Initially, that's really interesting -- but as time passes, these histories diverge, and they're now alternate histories that diverged at some point in the past. Since history is contingent and subject to butterfly effects (as described in the Psychohistory article and the section on limits on transapients), they're not going to come true. But for broad patterns that have strong antecedents from before the split, and aren't very contingent, they would still be informative (to the extent that the archai creating the simulation was good at predicting).

Assuming the media feed somewhere includes a large set of EGs from a set alternate histories, I would expect the Encyclopedia Galactica Institute to be interested -- if nothing else, it should help them find typos in and alternative phrasing for recently added articles. And if they see a consistent pattern across a large set of "future articles" from (alternate histories who branching point wasn't too far back, that were interesting or alarming enough (say >90% of them agree that minor variations on some specific major event occurs), then I can imagine them actually publishing an averaged or 'typical' version of these, with a prominent warning mark on it, an analysis of the range of variation, and a link to a detailed explanation of Orion's Arms, what's know or speculated about the source, and comments on their likely level of reliability or otherwise.
Reply


Forum Jump:


Users browsing this thread: 1 Guest(s)