What it is like to be a hassler?

[confocaloid]

An incomplete list of questions:

• Which oscillators fall under the general notion of "hassler"?
• Which oscillators should be described as hasslers?
• Which oscillators should be described as honey farm hasslers?
• Which oscillators should be described as "X hasslers", where "X" stands for some unstable object or evolutionary sequence (such as R-pentomino or century)?
• Which oscillators should be described as "X hasslers", where "X" stands for a stable or oscillating object (such as toad, beehive or honey farm constellation)?
• What are possible ways to make the general notion of "hassler" more precise/rigorous?
• Which oscillators are not hasslers?

One of questions that I'm particularly interested is, whether or not oscillators that don't have an external "hassling structure" should be described as hasslers. I think it would be better to distinguish those free-standing reactions from oscillators where a reaction is controlled by catalysts/sparkers.

For a hassler, it is possible to say "this is a case of X hassling Y" or "this is a case of X hassling several copies of Y". For something like Karel's p177, that description does not work, because there is no "X".

---

For some previous discussion, see:

• Talk:Honey farm hasslers
• Talk:Octomino II hasslers
• (likely there are/were related discussions elsewhere that I missed or forgot to link)

While discussing what it means to be a "hassler" on the wiki, I noticed that the p25 pre-pulsar shuttle could be extended to a wick:
[...]

Is it "correct" to refer to an oscillator as an X hassler if it consists of nothing but two or more copies of X interacting with each other? For example, could Gabriel's p138 be said to be a hassler, or that would be too much of a stretch of the idea of hassling?

Code: Select all

x = 21, y = 21, rule = B3/S23
11bo$10b3o$10bob2o$13bo4$2b2o4b2o$b2o10bo$2o11bo$b2o15b2o$7bo11b2o$7b
o10b2o$11b2o4b2o4$7bo$7b2obo$8b3o$9bo!

p3 pi hassler as a joke

Code: Select all

x = 35, y = 61, rule = B3/S23
6b2obo2bob2o5b2obo2bo$5bobob4obobo2bo2bob4o$5bobo5bo2b5obo5b2obo$4b2ob
4o2bobo5bob4obob2o$7bobobobo2bobo2bobobobo$6b2ob3o2bobobob2obobo$4b2o
6bo2b2o5b3obo$2o5bo3bo3bo2bo2bo5bob2o$obo2bo8bobob2obo2bob2ob2o$2bo2bo
3bob2o3bo2b3o2b2obo$obo2b3obobobo3b2o2b3obobo$2o10b2o7bo3bob2o$5bo4bob
3ob2o8bo2bob2o$5b2ob2ob2o2b3o2bob2ob2o2bobo$4bo4bo2bo6bob2o2b2o2bobo$
5b4o3bobobo4b2o5b4ob2o$10bo2b2ob5o7bo3bobo$5b4obo4bo5bob2obobobobo$5bo
2b2o5bob2o2bo2bob2obobobo$16bobo3b2o5bo3bo$29b2o2b2o20$6b2obo2bob2o5b
2obo2bo$5bobob4obobo2bo2bob4o$5bobo5bo2b5obo5b2obo$4b2ob4o2bobo5bob4ob
ob2o$7bobobobo2bobo2bobobobo$6b2ob3o2bobobob2obobo$4b2o6bo2b2o5b3obo$
7bo3bo3bo2bo2bo5bob2o$5bo8bobob2obo2bob2ob2o$5bo3bob2o3bo2b3o2b2obo$5b
3obobobo3b2o2b3obobo$12b2o7bo3bob2o$5bo4bob3ob2o8bo2bob2o$5b2ob2ob2o2b
3o2bob2ob2o2bobo$4bo4bo2bo6bob2o2b2o2bobo$5b4o3bobobo4b2o5b4ob2o$10bo
2b2ob5o7bo3bobo$5b4obo4bo5bob2obobobobo$5bo2b2o5bob2o2bo2bob2obobobo$
16bobo3b2o5bo3bo$29b2o2b2o!

[Scorbie]

Note for context: the p3 "hassler" I posted below was a joke (people in Discord were trying to make the lowest period pi hassler as far as I remember). If that one makes it difficult to settle on a definition feel free to ignore it.

[Scorbie]

Honestly I think some CGoL definitions are inherently hard to well-define, as definitions come out of observing patterns rather than become formalized from the ground up. Just as it's hard to rigorously define a "cat" given an image.

That being said here are some ideas that might help.

For a perturbation to hassle X, some things to consider are:

1. Is X changing? e.g. part of a rotor? (to exclude the p3 above if you want)
2. Is the perturbation small enough for one to identify X as a separate component? (To exclude low period not-so-much-hasslers that happen to contain X. Also I think this could be made to exclude signal circuitry oscs as well.)
3. Is the component after the perturbation different from X without the perturbation? (If you want to exclude things like QB shuttles)

[confocaloid]

Several relevant quotes from earlier discussions were posted on Talk:Octomino II hasslers.

[confocaloid]

Quoting from a duplicate thread:

To start off, here is my definition of a "hassler":

a Hassler is an oscillator that works by manipulating the evolution of one (1) or more unstable patterns using still lifes and/or other oscillators.

It's likely an inadequate definition for this, so i would like to hear any ideas anyone might have.

One issue is that there are certainly hasslers that hassle periodic or stable objects (e.g. toad, loaf, beehive).

Another issue is that it is unclear what counts as manipulation of evolution.

Your proposed definition would exclude things like Karel's p177, because there are no still lifes/oscillators manipulating the evolution a set of unstable patterns. Personally, I believe that Karel's p177 is not a hassler indeed (and shouldn't be described as a hassler), but I'm not everyone, and there is some disagreement on this from the previous discussions on this.

When interactions between components in the oscillator are very frequent (for low-period oscillators), it becomes unclear whether or not it is meaningful to describe the oscillator as a hassler / distinguish between the "hassling" part and the "hassled" part.
(edited to clarify 2023-08-19 00:12 UTC)

[Haycat2009]

My opinion: X is pertubed by another X/something else such in a way that X is [mutally if X pertubs X] rephased backwards but not turned or moved so far away that another set of X/catalysts is needed to move it back.(Or it would be a shuttle)

[confocaloid]

Tanner's p46 is a nice example. It is definitely a pi-heptomino hassler. One set of catalysts (snake, block, fishhook) turns a pi-heptomino into a pi in another location; then the pseudo-still-life catalyst turns that pi back into a pi in the original location.

Code: Select all

x = 13, y = 26, rule = B3/S23
2b2o$2bo$3bo$2b2o2$9b2o$9bo$10bo$9b2o$b2o$b2o6b2o$o7bobo$b2o6bo$b2o7b
3o$12bo7$b2o$b2o2b2o$5bobo$7bo$7b2o!

[Entity Valkyrie 2]

Another one is whether snark loops count as HF hasslers, or rectifier loops as pi hasslers etc.
Or whether a dependent reflector loop counts as a "HF + glider" hassler
Are shuttles (2-way and 4-way) hasslers?

[confocaloid]

I wouldn't want to describe a rectifier loop as a "pi-heptomino hassler". The pi appears temporarily during interaction, and the supporting structure (the rectifier) does not recover if the pi is removed at that point:

Code: Select all

#C p106 rectifier loop; remove the pi-heptomino and it fails
x = 108, y = 86, rule = B3/S23
8bo$7bobo$8bo2$6b5o$5bo4bo20b2o$4bo2bo23bo$bo2bob2o21bobo$obobo5bo18b
2o$bo2bo4bobo8bo$4b2o2bo2bo6b2ob2o$9b2o6bo3bo$16bo4bobo16b2o$15bo3bob
o18bo$15bo2bob2obo14bobo$14bo2bo2b2o8bobo5b2o$15b3o2b2o9b2o$15b2obobo
10bo2$17b3o3$30bo$31bo$30bo5$35b3o$35bo$27b2o7bo$27b2o2b2o$31b2o$29b3o
$29b2o$29bo5$57bo$58b2o$57b2o9$77bo$76bobo$76bobo$62b2o13bo$62bobo$62b
o10$81b3o$81bo$81b3o$68b2o$67bobo$67bo$66b2o$97b2o$96bo2bo2b2o$96bobo
4bo2bo$77b2o18bo5bobobo$76bobo21b2obo2bo$76bo23bo2bo$75b2o20bo4bo$97b
5o2$99bo$98bobo$99bo!

On the other hand, probably it would be uncontroversial to describe the following as a p266 glider hassler --

Code: Select all

x = 75, y = 53, rule = B3/S23
8bo$7bobo$8bo2$6b5o$5bo4bo20b2o$4bo2bo23bo$bo2bob2o21bobo$obobo5bo18b
2o$bo2bo4bobo$4b2o2bo2bo$9b2o$40b2o$40bo$38bobo$38b2o2$20b2o$20b2o$44b
o$43bobo$43bobo$44bo$30b2o$29b2o$31bo5$30bo$29bobo$29bobo$30bo$53b2o$
53b2o2$35b2o$34bobo$34bo$33b2o$64b2o$63bo2bo2b2o$63bobo4bo2bo$44b2o18b
o5bobobo$43bobo21b2obo2bo$43bo23bo2bo$42b2o20bo4bo$64b5o2$66bo$65bobo$
66bo!

[HerscheltheHerschel]

My opinion: X is pertubed by another X/something else such in a way that X is [mutally if X pertubs X] rephased backwards but not turned or moved so far away that another set of X/catalysts is needed to move it back.(Or it would be a shuttle)

This should be the standard definition of hassler

[confocaloid]

My opinion: X is pertubed by another X/something else such in a way that X is [mutally if X pertubs X] rephased backwards but not turned or moved so far away that another set of X/catalysts is needed to move it back.(Or it would be a shuttle)

This should be the standard definition of hassler

Would Tanner's p46 continue to count as a pi-heptomino hassler?

[HerscheltheHerschel]

My opinion: X is pertubed by another X/something else such in a way that X is [mutally if X pertubs X] rephased backwards but not turned or moved so far away that another set of X/catalysts is needed to move it back.(Or it would be a shuttle)

This should be the standard definition of hassler

Would Tanner's p46 continue to count as a pi-heptomino hassler?

Yes, in my opinion.

[Haycat2009]

My opinion: X is pertubed by another X/something else such in a way that X is [mutally if X pertubs X] rephased backwards but not turned or moved so far away that another set of X/catalysts is needed to move it back.(Or it would be a shuttle)

This should be the standard definition of hassler

Would Tanner's p46 continue to count as a pi-heptomino hassler?

Is Simkin's p60 a Herschel hassler? The answer for both will be the same.

[confocaloid]

My opinion: X is pertubed by another X/something else such in a way that X is [mutally if X pertubs X] rephased backwards but not turned or moved so far away that another set of X/catalysts is needed to move it back.(Or it would be a shuttle)

The way I read things, Tanner's p46 would fail to count as a p46 pi-heptomino hassler according to your proposed definition, because the pi-heptomino is "moved so far away that another set of X/catalysts is needed to move it back":

Tanner's p46 is a nice example. It is definitely a pi-heptomino hassler. One set of catalysts (snake, block, fishhook) turns a pi-heptomino into a pi in another location; then the pseudo-still-life catalyst turns that pi back into a pi in the original location.

Code: Select all

x = 13, y = 26, rule = B3/S23
2b2o$2bo$3bo$2b2o2$9b2o$9bo$10bo$9b2o$b2o$b2o6b2o$o7bobo$b2o6bo$b2o7b
3o$12bo7$b2o$b2o2b2o$5bobo$7bo$7b2o!

For me it is obvious that Tanner's p46 is a p46 pi hassler. If a proposed definition does not allow that, well, I suspect so much the worse for the proposed definition.

Would Tanner's p46 continue to count as a pi-heptomino hassler?

Is Simkin's p60 a Herschel hassler? The answer for both will be the same.

I do not understand why the answer has to be the same, given that the oscillators are different. Tanner's p46 features one pi-heptomino; Simkin's p60 features two non-interacting Herschels. Because the two Herschels do not interact in any way, I am not sure whether there is anything close to universal agreement to describe Simkin's p60 as a Herschel hassler. But probably most people who care about this question would agree that a capped (e.g. with eater 1s) Simkin glider gun is a p120 Herschel hassler.

[Haycat2009]

My opinion: X is pertubed by another X/something else such in a way that X is [mutally if X pertubs X] rephased backwards but not turned or moved so far away that another set of X/catalysts is needed to move it back.(Or it would be a shuttle)

The way I read things, Tanner's p46 would fail to count as a p46 pi-heptomino hassler according to your proposed definition, because the pi-heptomino is "moved so far away that another set of X/catalysts is needed to move it back":

Tanner's p46 is a nice example. It is definitely a pi-heptomino hassler. One set of catalysts (snake, block, fishhook) turns a pi-heptomino into a pi in another location; then the pseudo-still-life catalyst turns that pi back into a pi in the original location.

Code: Select all

x = 13, y = 26, rule = B3/S23
2b2o$2bo$3bo$2b2o2$9b2o$9bo$10bo$9b2o$b2o$b2o6b2o$o7bobo$b2o6bo$b2o7b
3o$12bo7$b2o$b2o2b2o$5bobo$7bo$7b2o!

For me it is obvious that Tanner's p46 is a p46 pi hassler. If a proposed definition does not allow that, well, I suspect so much the worse for the proposed definition.

Would Tanner's p46 continue to count as a pi-heptomino hassler?

Is Simkin's p60 a Herschel hassler? The answer for both will be the same.

I do not understand why the answer has to be the same, given that the oscillators are different. Tanner's p46 features one pi-heptomino; Simkin's p60 features two non-interacting Herschels. Because the two Herschels do not interact in any way, I am not sure whether there is anything close to universal agreement to describe Simkin's p60 as a Herschel hassler. But probably most people who care about this question would agree that a capped (e.g. with eater 1s) Simkin glider gun is a p120 Herschel hassler.

The point is that they are shuttles! You might as well call the p81 glider loop a glider hassler when it it obviously a shuttle.

[confocaloid]

The point is that they are shuttles!

What do you mean by 'they'?

[Haycat2009]

The point is that they are shuttles!

What do you mean by 'they'?

Tanner's p46 and CL48C loops, and all conduit loops.

[confocaloid]

The point is that they are shuttles!

What do you mean by 'they'?

Tanner's p46 and CL48C loops, and all conduit loops.

The p192 one maybe would count as a shuttle, but not Tanner's p46. Both are hasslers.

[Haycat2009]

What do you mean by 'they'?

Tanner's p46 and CL48C loops, and all conduit loops.

The p192 one maybe would count as a shuttle, but not Tanner's p46. Both are hasslers.

Don't tell me that you are going to call the p239 rectifier loop a glider hassler. Seriously, Tanner's p46 is a shuttle as another set of catalysts are needed to move the pi back. All conduit loops are shuttles!

[confocaloid]

A shuttle is an oscillator in which an unstable object moves back and forth between stabilizing objects. https://conwaylife.com/w/index.php?titl ... did=127608 In Tanner's p46, the pi sequence reappears in two places and two different orientations, but it does not move back and forth like e.g. in queen bee shuttle.

[Haycat2009]

A shuttle is an oscillator in which an unstable object moves back and forth between stabilizing objects. https://conwaylife.com/w/index.php?titl ... did=127608 In Tanner's p46, the pi sequence reappears in two places and two different orientations, but it does not move back and forth like e.g. in queen bee shuttle.

Oh yes. It does move far enough for other catalysts to be needed.

[confocaloid]

I don't really understand why you suggested that particular definition of 'hassler'. It feels artificial to me. Whether or not the perturbed reaction "X" is "turned or moved so far away that another set of catalysts is needed to move it back", either way, it seems fair to write that "X" is hassled by those catalysts. In a queen bee shuttle, the queen bee is hassled by stabilizing objects (e.g. blocks).

[dvgrn]

A shuttle is an oscillator in which an unstable object moves back and forth between stabilizing objects. https://conwaylife.com/w/index.php?titl ... did=127608 In Tanner's p46, the pi sequence reappears in two places and two different orientations, but it does not move back and forth like e.g. in queen bee shuttle.

That all seems right to me. This also seems like a good place to mention an interesting historical limitation on "shuttle" that hotdogPi pointed out a while back:

On Discord, hotdogPi pointed out that the 2005 p72 quasi-shuttle had to be called "quasi", specifically because "although this looks at first sight like a shuttle, it isn't really, because each side of the oscillator uses a different hassling reaction".

Based on that LifeWiki definition, maybe it would be a good idea to adjust the claim "all conduit loops are shuttles"... maybe "all conduit loops are quasi-shuttles" would fit past usage a little better, but there are still problems.

Possibly Tanner's p46 could be counted as a quasi-shuttle, though?

EDIT 3/25/2024: Some later discussion clarified that this was all a mis-use of "quasi" on my part -- please ignore everything about "quasi". The next sentence is still true, though.

At least, I wouldn't personally ever use "shuttle" for anything with multiple objects going around a loop.

[MEisSCAMMER]

Possibly Tanner's p46 could be counted as a quasi-shuttle, though?

Well, let's work through what happens:

-main pi appears
-junk happens
-secondary pi appears, lasts for ~5 gens before
-getting rotated and becoming main pi
-repeat

My opinion on this is that it is not a quasi-shuttle, if only because it does not meet (my) terms of what a quasi-shuttle should be. Here's the definition of shuttle, as lifted from the Wiki:

A shuttle is an oscillator in which an unstable object moves back and forth between stabilizing objects.

The key here is "between" — what is on the other side of the main pi? A spark. Even if it were a quasi-shuttle the appearance should still be maintained, otherwise there would be no use defining it in terms of the shuttle.

[dvgrn]

The key here is "between" — what is on the other side of the main pi? A spark. Even if it were a quasi-shuttle the appearance should still be maintained, otherwise there would be no use defining it in terms of the shuttle.

Heh, well, no doubt this could be debated endlessly just like anything else. There's a sense in which the pi is pushed rightward by the yellow catalysts on the left, and then leftward again by the white catalyst on the right, shuttling between those two positions.

Code: Select all

x = 23, y = 26, rule = B3/S23Super
12.2E$12.E$13.E$12.2E$11.2B$9.5B5.2C$8.6B5.C$9.6B5.C$8.7B4.2C$5.B.8B.
2B$3.8B3D5B2C$.10BDB3A2BC.C$11B2DCBA2B.C$.12BABAB3.3C$3.15B4.C$6.13B$
7.12B$8.11B$9.9B$9.6B$11.4B$10.B2E2B$11.2E.B2E$14.BE.E$17.E$17.2E!

Now, I have absolutely no interest in trying to draw a bright line between things that are quasi-shuttles and things that are not quasi-shuttles -- if we go by past usage, it's really not a terribly useful term. I kind of think that Tanner's p46 could quasi-reasonably be called a quasi-shuttle; it's more reasonable than calling a still life a "p1 oscillator", but that doesn't mean that it's really a good thing to call it.