(cont.) Adding kinetic symmetry to infoboxes

[confocaloid]

Previous discussion: https://conwaylife.com/w/index.php?oldi ... _infoboxes

Would there be any objections to me doing this manually over the coming weeks? - [[User:Muzik|Muzik]] ([[User talk:Muzik|talk]]) 22:40, 13 March 2024 (UTC)

Yes there are objections. At the current point of the discussion, I object to any changes related to adding kinetic symmetry to infoboxes, because those changes were not yet described in detail, and hence were not yet discussed.

• (1) The author of the proposal did not describe details. (What are the exact proposed changes?)
• (2) The author of the proposal did not cover all concerns already raised in previous discussion. (E.g. infoboxes are already quite tall, and significantly reduce the available width for the main content that is located in the beginning of the page. There are already multiple issues with infoboxes. Would the proposed changes (which were not yet described in detail) be an improvement, or they are likely to lead to new issues?
• (3) Why post on a now-inactive Tiki bar, after this message noting that further discussions should happen here on the forum?

[muzik]

Symmetry would be obviously included as an infobox row - I fail to see any other way of including this information in an infobox in any other useful way.

IMG_4267.jpeg (263.54 KiB) Viewed 10908 times

On the topic of height issues, it may be possible to delete the largely-unimportant "Oscillator type" and "Family" sections which provide information which can be communicated much more clearly in prose. It may also be useful to merge "Volatility" and "Strict volatility" into the same row, separated horizontally as in LifeViewer, and possibly likewise for Period and Mod.

[muzik]

Given that there has been no opposition to implementing this over the past four months I will start adding this to pages next week unless any objections arise in the meantime.

[muzik]

This will be added to pages soon as over a week has passed since the previous message.

[muzik]

Kinetic symmetries on oscillator pages should now be complete to my knowledge. At some point in the near future I'll work on adding these to spaceship and still life pages, and possibly static symmetry to other, non-periodic pattern pages.

[confocaloid]

Might be a good idea to collect links:
( *c ) ( *k ) ( +c ) ( +c*c ) ( +e ) ( +k ) ( +k*k ) ( -c ) ( -c+c ) ( -c+e ) ( -e ) ( -e+e ) ( -e+k ) ( .c ) ( .c+c ) ( .crc ) ( .cxc ) ( .e ) ( .e+e ) ( .k ) ( .k+k ) ( .krk ) ( .kxk ) ( / ) ( /xc ) ( /xk ) ( n ) ( n-c ) ( n-e ) ( n.c ) ( n.e ) ( n.k ) ( n/ ) ( nrc ) ( nrk ) ( rc ) ( rc*c ) ( rk ) ( rk*k ) ( xc ) ( xc*c ) ( xk ) ( xk*k )

[confocaloid]

( n-c ) and ( -e+k ) do not appear to have any examples on the wiki so far. Dean Hickerson's oscillator stamp collection lists examples 2.2.5 and 2.5.6 for those, maybe create pages for them (as first known examples)?

One interesting point here is that there are three different properties:

• the symmetry of the oscillator,
• the symmetry of the rotor (ignoring the stator),
• the symmetry of the engine (ignoring the supporting catalysts or sparkers).

In the following example, the symmetry of the engine is identical to the symmetry of the rotor, but the symmetry of the entire oscillator is different:

Code: Select all

x = 40, y = 19, rule = B3/S23
2b3o30b3o2$o5bo2b2o18b2o2bo5bo$o4b2o4bo16bo4b2o4bo$o11bo14bo11bo$11b2o
14b2o3$5b2o26b2o$5bo28bo$6b3o22b3o$9bo21bo$6b3o$5bo$6b3o$9bo$6b3o$5bo$
5b2o!

In the following example, the symmetry of the oscillator is identical to the symmetry of the rotor, but the symmetry of the engine is different:

Code: Select all

x = 42, y = 12, rule = B3/S23
2b3o30b3o2$o5bo2b2o18b2o2bo5bo$o4b2o4bo16bo4b2o4bo$o11bo14bo11bo$11b2o
14b2o3$5b2o31b2o$5bo32bo$6b3o30b3o$8bo32bo!

[muzik]

Still lifes should now also be done, leaving only spaceships and miscellaneous objects awaiting reclassification.

As for rotor stuff: I have been considering a category for oscillators whose rotor is a higher symmetry than the stator is (e.g. champagne glass, Hertz oscillator), though not sure how granular we should be with it.

[muzik]

Spaceships are also now done, excluding the following four:

• Orthogonal loopship

• Orthogonoid

• QuickSilver Demonoid

• Speed Orthogonal Loopship

For #1, #2 and #4, someone needs to,figure out if the line of glide reflection occurs through cell centers (n-c) or cell vertices (n-e). For #3, we don't even have the period of this object, so that page clearly needs further work.

[vilc]

For #1, #2 and #4, someone needs to,figure out if the line of glide reflection occurs through cell centers (n-c) or cell vertices (n-e). For #3, we don't even have the period of this object, so that page clearly needs further work.

This ought to be verified, but I think that the period of the Quicksilver Demonoid is 16,777,218=2^24+2.
Edit :
There was a copy-paste error... Actual period is 2^24, a round number (checked with Golly hashes this time). Full speed is 5923c/16777216d.

[confocaloid]

Still lifes should now also be done, leaving only spaceships and miscellaneous objects awaiting reclassification. [...]

I forgot to mention earlier that "rulespecial" is no longer a thing:
https://conwaylife.com/w/index.php?titl ... _parameter

Hence it would be nice to remove this parameter from infoboxes whenever you're anyway editing a page where it is present (as cleanup), and do not add this parameter to any pages where it is not present (as in https://conwaylife.com/w/index.php?diff ... ldid=98396 ).

[muzik]

Should we classify n/ and n/e as two different symmetries for diagonal spaceships? I originally lumped these together on the basis that it was a simple difference of how the object is displaced, but further research into wick symmetry reveals that the two would probably be better treated as distinct even despite this.

[muzik]

Should we classify n/ and n/e as two different symmetries for diagonal spaceships?

Further research has led to the conclusion that this should be done, and done it has been.

In other news, static symmetries have been added to infoboxes for non-[still life, oscillator, spaceship] patterns.

The next logical question: should we also add kinetic symmetries for pages on puffers, guns and rakes? For example, the Simkin glider gun is clearly 180-degree rotationally symmetric, and the p256 gun 90-degree rotationally symmetric. BLSE and Puffer 2 are glide-symmetric, which can be seen easily in their ash output. The usual kinetic symmetries would presumably apply here.

[confocaloid]

[...] The next logical question: should we also add kinetic symmetries for pages on puffers, guns and rakes? [...]

Can you please provide an one-post overview of all substantial changes on this topic that were already done, preferably including links to [categories containing] created wiki categories?

I think it would be nice to get a better understanding of the situation so far, before trying to implement further changes which may turn out more complicated/nontrivial (e.g. due to the mentioned pattern types being more complicated).

[...] and the p256 gun 90-degree rotationally symmetric. [...]

Some pages (such as period-256 glider gun) aren't about specific patterns, but instead are about types or classes of patterns.
Of course, just because some specific p256 gun has a certain symmetry, doesn't imply anything about symmetry of other notable p256 guns.
I think such pages about classes/types of patterns shouldn't get any infoboxes at all.

[muzik]

[...] The next logical question: should we also add kinetic symmetries for pages on puffers, guns and rakes? [...]

Can you please provide an one-post overview of all substantial changes on this topic that were already done, preferably including links to [categories containing] created wiki categories?

Refer to this category for today's changes: https://conwaylife.com/wiki/Category:Pa ... metry_type

[Haycat2009]

Replace kinetic symmetry in SL infoboxes with static symmetry: It is redundant.

EDIT: This will be done soon.

[muzik]

Replace kinetic symmetry in SL infoboxes with static symmetry: It is redundant.

Still lifes are period-1 oscillators and therefore will be treated identically to other odd-period oscillators in that they are classified by kinetic symmetry rather than static symmetry.

[confocaloid]

Replace kinetic symmetry in SL infoboxes with static symmetry: It is redundant.

Still lifes are period-1 oscillators and therefore will be treated identically to other odd-period oscillators in that they are classified by kinetic symmetry rather than static symmetry.

I think this view is a bit counterintuitive. In many cases, "still life" and "oscillator" are considered to be two separate nonoverlapping pattern types (with "oscillator" automatically implying "period >= 2").

The paperclip is a still life and it feels awkward to discuss its "kinetic symmetry", because that assumes some potential distinction between symmetries of a single generation and symmetries of the evolution of the pattern, and a still life lacks the possibility of any such distinctions.

[muzik]

I think this view is a bit counterintuitive. In many cases, "still life" and "oscillator" are considered to be two separate nonoverlapping pattern types (with "oscillator" automatically implying "period >= 2").

The paperclip is a still life and it feels awkward to discuss its "kinetic symmetry", because that assumes some potential distinction between symmetries of a single generation and symmetries of the evolution of the pattern, and a still life lacks the possibility of any such distinctions.

Much the same could be said for period-3, period-5, etc. oscillators, since on a square grid they can't have time symmetry - but it still makes more sense to classify them by kinetic symmetry than to randomly switch between static and kinetic symmetry for infoboxes depending on period (especially since period-2, 4, etc. oscillators can have spatial symmetry but not temporal symmetry).

[Haycat2009]

Replace kinetic symmetry in SL infoboxes with static symmetry: It is redundant.

Still lifes are period-1 oscillators and therefore will be treated identically to other odd-period oscillators in that they are classified by kinetic symmetry rather than static symmetry.

I think this view is a bit counterintuitive. In many cases, "still life" and "oscillator" are considered to be two separate nonoverlapping pattern types (with "oscillator" automatically implying "period >= 2").

The paperclip is a still life and it feels awkward to discuss its "kinetic symmetry", because that assumes some potential distinction between symmetries of a single generation and symmetries of the evolution of the pattern, and a still life lacks the possibility of any such distinctions.

I am going ahead with the plan now. I do not understand why the kinetic symmetry of a SL is important. Still life's kinetic symmetry is determined by static symmetry regardless of neighbourhood, but not for p3 oscs in hex grid.

Guns, rakes and puffers can have kinetic symmetry. I find it useful.

Also, please do not undo the edits without asking confocaloid or an admin first.

[muzik]

I am going ahead with the plan now.

No consensus has been reached. You shouldn't be going ahead with major changes based on an ongoing conversation with no clear outcome.

[confocaloid]

I am going ahead with the plan now.

No consensus has been reached. You shouldn't be going ahead with major changes based on an ongoing conversation with no clear outcome.

I should point out that no consensus has been reached yet for adding any kind of symmetry information to infoboxes of still life patterns in the first place. Nevertheless, certain large-scale changes were already performed, before reaching any such consensus.

I think this view is a bit counterintuitive. In many cases, "still life" and "oscillator" are considered to be two separate nonoverlapping pattern types (with "oscillator" automatically implying "period >= 2").

The paperclip is a still life and it feels awkward to discuss its "kinetic symmetry", because that assumes some potential distinction between symmetries of a single generation and symmetries of the evolution of the pattern, and a still life lacks the possibility of any such distinctions.

Much the same could be said for period-3, period-5, etc. oscillators, since on a square grid they can't have time symmetry - but it still makes more sense to classify them by kinetic symmetry than to randomly switch between static and kinetic symmetry for infoboxes depending on period (especially since period-2, 4, etc. oscillators can have spatial symmetry but not temporal symmetry).

I'm fine with specifying kinetic symmetry of an actual odd-period oscillator (i.e. p3, p5, p7, etc.) Such oscillators still have kinetic symmetry, even if there are certain restrictions.

I don't support the idea to specify "kinetic symmetry" for still lives. The idea behind the word 'kinetic' in "kinetic symmetry" is incompatible with the idea of a still life.

I don't have an issue with switching between static and kinetic symmetry in infoboxes depending on whether the pattern is a still life or an oscillator. I believe this distinction is intuitively clear, useful and should be preserved, in infoboxes and elsewhere.

[Haycat2009]

There has been no reply for a while. Can we safely declare consensus?

[confocaloid]

There has been no reply for a while. Can we safely declare consensus?

"Declare"? Yes.
"Safely"? No.
Consensus requires discussion. Discussion requires time. Weeks, months, sometimes even years.
Sometimes people who could contribute to the discussion about a certain topic are not here yet. Some people are currently inactive.

If the discussion doesn't remain open for any future comments / feedback / corrections / etc., then there's no possibility of consensus to begin with.

In many relatively recent disagreements, there was very little actual discussion about the matter. (Although there were many tangential distractions, loosely relevant comments, etc.)