Create your own terminology

[confocaloid]

Semiomniperiodicity: An alternative of omniperiodicity.
If all periods are proven impossible or possible, then the rule is semiomniperiodic.
For example, B/S and B/S012345678 are semiomniperiodic, because all periods are impossible.
WireWorld is semiomniperiodic too, becuase It has every period other than 2 and P2 is impossible.

The concept "omnivelocital" needs further discussion.

It sounds like that would create a fundamental difference between the existing concept of omniperiodicity and the proposed notion of "semiomniperiodicity":

• Whether or not a cellular automaton is omniperiodic, cannot change over time. Being or not being omniperiodic is a mathematical property of a CA. (It is often currently unknown whether or not a given CA is omniperiodic, though.)
• Whether or not a cellular automaton is "semiomniperiodic", is
  (a) either defined in terms of current state of knowledge (which would mean it fails to be a mathematical property, and would allow it to change over time),
  (b) or it would be obviously true: for every cellular automaton, and for every period, either there is a nontrivial oscillator of that period, or there is no such oscillator; therefore every CA is "semiomniperiodic".

For this reason, I would prefer some way of much more clearly stating something like

>> "existence of nontrivial oscillators of every period in such-and-such cellular automaton is a solved problem"

(whenever one needs to state something like that), instead of going into the trouble of defining a problematic (misleading) term that would superficially appear to be a term referring to some mathematical property of a CA, but instead would merely refer to the current state of knowledge about the CA.

[R2INT]

[...]
The concept "omnivelocital" needs further discussion.
[...]

In Conway's Life (B3/S23), I think that omnivelocital should refer to at least one spaceship can be mapped to each member of the following set:
{(X,Y,Z)∣X+Y≤2/Z}
Here, the terms X, Y, and Z represent components of a description of the spaceship velocity representation (X,Y)c/Z.

However, in addition to including the set of all possible velocities, this also includes the set of all period-multiplied spaceships. Here is a revision of the set where X and/or Y must be coprime with Z:
{(X,Y,Z)∣X+Y≤2/Z,(gcd(X,Z)=1∨gcd(Y,Z)=1)}

[b-engine]

Recursive breeder: basically a puffer that makes copies of itself. Could be categorized as replicators or failed replicators, but differs from typical replicators in the position of the child relative to the parent: in a XOR-type replicator, there are no distinctive original parent, however there is in a recursive breeder.

An example:

Weird self puffer (a failed infinite recursive MMMMMMMMMMMMMMM.... breeder):

Code: Select all

x = 5, y = 4, rule = B2ek3ijy4ijn5cj/S1c2cek3n5e
2b3o$bo2bo$o3bo$2bo!

[b-engine]

Uncorderization or decorderization: engineer a puffer or a rake using only spaceships.
Recorderization: corderize a puffer or rake obtained via uncorderization. Used to obtain sparky spaceships.


Inverse growing/shrinking spaceship:

…
An inverse growing/shrinking spaceship is a spaceship that:
The back end creates a puffer that's faster than both the front end and back end.
The puffer catches up the front end while the back end clears up the debris.
The front end destroys the puffer and optionally ignites a fuse.
The backend detects that the debris is no more, then creates another puffer to restart the process.

Such spaceships must be necessarily slower than c/2, as the puffer needs to be faster than the front end.

Inverse moving sawtooth: inverse growing/shrinking spaceship, but the front end move faster than the back end, while slower than the puffer.
Inverse sawtooth: the stationary part ignites a fuse (necessary as usual sawtooth shrinking reactions travel slower than spaceships), while the moving part is a rake.

[Quantum-mechanics]

Seems unclear. Is there some sort of measurable/detectable property that is intended here ("sharp", "blunt")? If so, can you define the properties? Or this is supposed to refer to some specific catalyses (in that case, what is important and what is irrelevant)?

Sharp catalysis: [...]
Blunt catalysis: [...]
This is primarily intended for block catalyses but can probably be used for other catalysts.

Sharp catalysis:

Code: Select all

x = 6, y = 5, rule = LifeHistory
A$2A2.D$3A.AC$4.AC$4.D!
#C [[ AUTOSTART GPS 1 STOP 2 ]]

Code: Select all

x = 7, y = 5, rule = LifeHistory
2A$3A2.D$4A.AC$2A3.AC$5.D!
#C [[ AUTOSTART GPS 1 STOP 2 ]]

Blunt catalysis:

Code: Select all

x = 7, y = 3, rule = LifeHistory
.2A$A2.A.AC$.2A2.2C!
#C [[ AUTOSTART GPS 1 STOP 4 ]]

Code: Select all

x = 6, y = 4, rule = LifeHistory
A$2A$2.A.AC$2A2.2C!
#C [[ AUTOSTART GPS 1 STOP 4 ]]

I think.

[b-engine]

True hassler: hassler that hassles a constellation itself rather than a predecessor of it. The constellation may (or must?) complete an oscillation cycle before being hassled again.

[dl-rs]

Uncorderization or decorderization: engineer a puffer or a rake using only spaceships.
Recorderization: corderize a puffer or rake obtained via uncorderization. Used to obtain sparky spaceships.

This is an unclear concept. Does this count as decorderization?

Code: Select all

x = 7, y = 23, rule = B3-c4e7e/S2-n34ci
4b3o$6bo$4bobo$2o2b2o$2o14$2o$2o2b2o$4bobo$6bo$4b3o!

[rutabaga]

pseudo-diehard:
a pattern that eventually reaches population zero within any given finite bounding box.

this could be... a regular diehard:

Code: Select all

#R 2-a356ik8/2-ak34e6a78/3
.A$A.A$.A!

it could be a pattern leaving only spaceships behind: (or just a spaceship to begin with lol)

Code: Select all

#R 2-a34ce57c8/2-ak34c5i8/3
bo$obo$bo!

it could be a nonfiller:

Code: Select all

#R 23-q/2-a3/7
2$2.A$.A.A$2.A!

it could be a combination of those last two options:

Code: Select all

x = 11, y = 11, rule = 23-q/2-a3/7
5.A$4.A.A3$.A3.A3.A$A3.A.A3.A$.A3.A3.A3$4.A.A$5.A!

lemme know if there are any other types of patterns that fall under this category!

[dl-rs]

In fact, there is.

lemme know if there are any other types of patterns that fall under this category!

Code: Select all

x = 18, y = 18, rule = B3-nr4z5y/S23-n4i
8b2o$7bobo$7bo$7bobo$8b2o3$b3o$o3bo8b2ob2o$2ob2o8bo3bo$14b3o3$8b2o$8b
obo$10bo$8bobo$8b2o!

(by yyh_baboon, in one of his rules with an alien 2c/5 pi ship)

[Citation needed]

Weekender: a suggested name for the Weekly Match Character in Rolling Sky. This is because the Character resembles the Weekender in Conway's Game of Life, and the Weekly Match occurs weekly.

[Quantum-mechanics]

Pattern space: A measure of the gaps in the pattern.
It is pattern width + 2 times pattern height + 2 - count of neibourhoods of pattern - count of cells of pattern.

(W + 2)*(H + 2) - N - C

Like this one:

Code: Select all

x = 8, y = 8, rule = LifeHistory
8B$B4D3B$BD2E2D2B$BDEDED2B$B3DE2DB$3BD2EDB$3B4DB$8B!

Width: 6 cells Height: 6 cells Count of neibourhoods of pattern(in life, it is Moore Neighbourhood): 21 Count of cells of pattern: 7
Pattern space: (6+2)*(6+2)-21-7=36
My English is bad.
If you are compare n-of patterns, just use the biggest bounding box for comparing.

[Gustone]

Pressurised: a pattern whose population is bigger than the usual population of ash created by breaking it.

[Quantum-mechanics]

Green light: left one.
Red light: right one.
Yellow light: bottom one.
Back light: top one.

Code: Select all

x = 7, y = 7, rule = LifeHistory
2.3D2$D5.D$D.3A.D$D2.A2.D2$2.3D!

Edit:
T-ship:

Code: Select all

x = 3, y = 2, rule = tlife
3o$bo!

Bullet ship:

Code: Select all

x = 3, y = 2, rule = B3/S23-a6c
3o$bo!

[hotdogPi]

Green light: left one.
Red light: right one.
Yellow light: bottom one.
Back light: top one.

Code: Select all

x = 7, y = 7, rule = LifeHistory
2.3D2$D5.D$D.3A.D$D2.A2.D2$2.3D!

Interestingly, this has a meaning I don't think you intended.

In real life, red is port and green is starboard. Given where you placed "back light", the T is facing downward, so red and green are correct relative to that orientation.

There wouldn't be a yellow light in front, though; those only appear on the road.

[Quantum-mechanics]

...

Interestingly, this has a meaning I don't think you intended.

In real life, red is port and green is starboard. Given where you placed "back light", the T is facing downward, so red and green are correct relative to that orientation.

There wouldn't be a yellow light in front, though; those only appear on the road.

I just don't want to name the light on the top.
The bottom is middle.

[Gustone]

die-easy (die easy, dieasy): a pattern that dies in one generation,

[andrewthelifer]

• A 0-SMOS is a spaceship; an n+1-SMOS is in at least one phase solely a collection of more than one n-SMOS about to collide. (Any n+1-SMOS is a 0-SMOS, ... n-SMOS as well, but one only refers to the largest k for which a pattern is a k-SMOS. Thanks to Macbi for pointing out SMOIs are omega-SMOS.)

• An xMOS is n-diverse if its component spaceships have n different speeds in isolation. Period-diversity aka pd counts speeds different by only a period (3c/6 vs 5c/10), while velocity-diversity aka vd only counts different velocities.

• A self-gun is a gun-MOS for which all component spaceships are of the same type and the gun emits this type, the whole pattern having to be MOS in infinitely many phases so that a glider synthesis of Simkin's glider gun doesn't count alone; a self-rake is similarly a rake-MOS that emits its own component. NimbleRogue has constructed a self-gun example.

• Suppose a failed replicator eventually evolves into N copies of itself and nothing else, and N is finite and maximal for all "k copies of itself and nothing else": N is said to be its crest. For a variant, suppose it eventually becomes Q copies of itself (with arbitrary debris besides) and Q is maximal for "k copies & debris": Q is its dirty crest.

• A SMOS' recipe is the multiset of spaceships within its MOS phase. A two-recipe SMOS has two phases, P and Q, with P being MOS, Q being MOS, and P and Q having different spaceship multisets (e.g. a SMOS which has 4-glider and 7-glider phases, or A+B and A+C phases). NimbleRogue has also exhibited a four-recipe SMOS.

• Suppose an isotropic CA which has n+1 states, one "OFF" (vacuum) and n "active" states, more parsimoniously called "colours". The CA is isochromatic if its transition function is invariant to switching the colours in any way. For example, Immigration cells are born according to the majority colour of their neighbours with no special privileges for Red or Blue, so patterns that are red<->blue-switched will be identical otherwise. 3-state Generations rules are not isochromatic because the older state can only die in the next tick, while the younger state can be kept alive. Day and Night can also be considered isochromatic between its quiescent and active state.

An example of anisochromacy: suppose a 2-colour rule has Red and Blue and turning a vacuum cell into a coloured cell is notated as a Birth transition. Mixed B2i would have to be anisochromatic because the two cells cannot be distinguished in any way that withstands isotropy ("the northwest cell") or isochromacy ("the red one"), but mixed B2a can be isochromatic because "the corner cell" and "the edge cell" retain their descriptions no matter how they're rotated.

[andrewthelifer]

Elaborating existing terminology:

Minimal definition of SMOS: a spaceship which is decomposable as "made of spaceships" and vacuum in at least one phase, and isn't MOS+vacuum in some other phase,

Restrictions on decomposability give increasingly specific and "natural" SMOSes within 2-state isotropic CAs:

• every cell in the neighbourhood of any component must be dead

• the collision must be rewindable: any individual component should be movable arbitrarily far in the opposite direction along the same lane, and when all components are moved back by N generations they should recreate the collision in the SMOS

• every component must make at least one 1-tick "step forward" that it would in isolation (e.g. glider going from its "even" phase to its "odd" phase)

• every component must cycle through its evolution and displacement at least once

Suppose a rule where the glider travels identically to Life and the same glider-collision rules apply: green would be a valid SMOS collision for 1), orange for 2), blue for 3) and yellow for 4), and each category is a subset of the one listed before it.

Code: Select all

x = 27, y = 9, rule = LifeSuper
25.E$4.A7.Q5.I7.E$5.A7.Q5.2I3.3E$.A.3A5.3Q4.2I$A$3A8.3Q4.2I$13.Q5.2I
3.3E$12.Q5.I7.E$25.E!

No SMOI can ever be a 4-type SMOS, as the evolution from the "big" collision state to the "small" single state requires the components of the big phase to fail to evolve fully.

[andrewthelifer]

One can formalise a shooter to be any finite periodic pattern for which Generation N is identical to G0 except that some periodic debris is present at generation N, with catapults being those shooters whose cores undergo 0 displacement from shot to shot, and firearms undergoing some displacement. Guns resp. clean rakes are catapults resp. firearms that fire the same things, and the same relation connects factories and puffers, while clean replicators are self-shooters.

• If a catapult can fire k times and its core is destroyed while trying to fire the k+1st time, the catapult is k-ruptured.

• If a catapult fires k times and this causes a reaction which eventually evolves into an unharmed catapult with vacuum in front of it, the catapult is k-jammed and forms a weird oscillator.

[b-engine]

Pseudonatural: an object that had occurred in a "soup" that's not randomly generated. For example, 2-engine cordership is pseudonatural.

[Resu]

Pseudonatural: an object that had occurred in a "soup" that's not randomly generated. For example, 2-engine cordership is pseudonatural.

Aren’t all soups randomly generated?

[b-engine]

Aren’t all soups randomly generated?

Sometimes you could find soups by finding the predecessor of a reaction via LLS or other search programs. They're not generated by randomization or hash algorithm (for example SHA-256), so not randomly generated.

[dvgrn]

Aren’t all soups randomly generated?

Sometimes you could find soups by finding the predecessor of a reaction via LLS or other search programs. They're not generated by randomization or hash algorithm (for example SHA-256), so not randomly generated.

I think the quotation marks around "soup" were important in the original definition.

Predecessors that aren't randomly generated are not really soups. So there can't be a soup that not's randomly generated -- but there can be a "soup" that's not randomly generated (because it looks just like a soup). Maybe pseudonatural objects come from pseudosoups.

[LuveelVoom]

A rule is omniartilleric (omniartillery) for a spaceship if every possible period of gun for that spaceship is known to exist.

[andrewthelifer]

I think the quotation marks around "soup" were important in the original definition.

Predecessors that aren't randomly generated are not really soups. So there can't be a soup that not's randomly generated -- but there can be a "soup" that's not randomly generated (because it looks just like a soup). Maybe pseudonatural objects come from pseudosoups.

The 2EC (as specified by b-engine) certainly has pseudo-soups, as small as 13x13. Are there any other patterns which are unnatural in any symmetry but have had engineered small predecessors?