Regarding the name of the """logarithmic""" replicator rule

[DroneBetter]

The creation of this thread may be regarded by all those who have been involved in its events thus far as an instigation of something that we had thus far mostly left settled.
I am of course incapable of impartiality, but will give as full a chronology as I can.

• No later than 2001-01-14: David Eppstein creates a page in his page, "Replicators: B36/S245"
• 2009-08-07: apg adds this to the Gun LifeWiki page, under the name "Logarithmic replicator rule"
• 2018-12-17: Muzik creates a page on the wiki under the same name, redirecting to Eppstein's page
• 2019-06-02: Ian07 turns this into a then-stubby, but full-fledged page of its own.
• 2019-06-26, 2020-04-12 (AforAmpere finds a 3-cell-wide pattern in an INT rule, emulating the replicator), 2020-08-19, 2021-05-22, 2021-10-21, 2022-09-25: Various uses of the name "Logarithmic replicator rule" appear in the forums
• 2023-09-29 (talk page): I find exact forms (in terms of bitwise and special functions) for the functions that give the number of iterations for the left and right edges to reach specific distances from the origin, in AforAmpere's emulator (equivalent to the 4-state Wolfram rule 0x190e002061040c0b86d0010e5980). This provides that the cell-length in the t'th iteration, l(t), has the asymptotic bounds 2*√t < l(t) < 2*√(5*t/3). As such, this prevailing name is incorrect, since the replicator's asymptotic width is Θ(√t).
• 2023-09-30 (Tiki bar), 2023-10-21 (my talk page): Disagreement ensues

We have had many discussions since then, which were mostly not lucrative, and some disagreement and argumentation that almost became an edit war, before Confocal and I (being both highly reasonable people as we are) agreed to a truce.

At present, the name of the page is "OCA:Logarithmic replicator rule" and the replicator embed is labelled "The namesake logarithmic replicator," yet my result is shown in the section OCA:Logarithmic replicator rule#Growth_bounds.

At present, for the page's title, we have three main options before us:
1. Leave it as-is forevermore (this is incorrect, but retains consistency with the many existing mentions. Proposed by Confocal)
2. Move it to "OCA:B36/S245" (more clinical, doesn't note the reason for its interestingness. Proposed by Galoomba)
3. Move it to "OCA:sqrt replicator rule" or (in keeping with O(sqrt(log(t)))) "OCA:Θ(√t) replicator rule" (Proposed by me)

Note that an administrator could also go through and amend all past mentions of the rule in the forums, probably negating the advantage of option 1.

If anyone has any opinions, they may express them here, with the intention of coming to a consensus in a civil (albeit perhaps arbitrarily protracted :-) manner, that doesn't involve any back-and-forth deleterious sparring edits.

[confocaloid]

Googling the existing name shows that it reached some other places outside this website, e.g.:
https://www.researchgate.net/publicatio ... _Character

3. Move it to "OCA:sqrt replicator rule" or (in keeping with O(sqrt(log(t)))) "OCA:Θ(√t) replicator rule" (Proposed by me)

I disagreed with this option because it basically would amount to documenting one's own terminology on the wiki, to replace an existing name. (Even if you did not do it yourself, it would still be a newly coined name replacing a name that was actually used.)

(Added later:)

2023-09-29 (talk page): I find exact forms (in terms of bitwise and special functions) for the functions that give the number of iterations for the left and right edges to reach specific distances from the origin, in AforAmpere's emulator (equivalent to the 4-state Wolfram rule 0x190e002061040c0b86d0010e5980). This provides that the cell-length in the t'th iteration, l(t), has the asymptotic bounds 2*√t < l(t) < 2*√(5*t/3). As such, this prevailing name is incorrect, since the replicator's asymptotic width is Θ(√t).

I think it is better to post such content somewhere on the forum (rather than on a talk page on the wiki). If you post your findings on the forum, then it is easier to link to that post specifically if/when needed, and distinguish it from other posts. OTOH talk pages of wiki articles are meant to be reserved for discussing content of those articles.

[Haycat2009]

3. Move it to "OCA:sqrt replicator rule" or (in keeping with O(sqrt(log(t)))) "OCA:Θ(√t) replicator rule" (Proposed by me)

I support this option. People are just going to make up random and even less accurate names (eg. Sussy rule, Toll booth, Air Jordan Rule) if we move it to B36/S245.

EDIT by dvgrn: Edited to remove a rule-violating personal attack. I had already PMed Haycat2009 to ask that the attack be deleted, but the PM hasn't been opened yet, and the post has since been reported, very appropriately. Be polite, y'all.

[confocaloid]

keeping inaccurate info

The name of something (a rule, a pattern, a corporation, etc.) does not generally provide any particularly accurate information. This is normal. Names do not have to make sense.
The existing name is commonly used and there are no other commonly used names (unless you count the rulestring as a name, which you should not).
Keeping the existing pagename provides accurate information "this rule is commonly referred to by this name".

[hotdogPi]

So why does "commonly used" matter here but not for "signal" or "transition"?

[confocaloid]

So why does "commonly used" matter here but not for "signal" or "transition"?

Reply:

• For the 'condition' vs. 'transition' issue, both variants are commonly used (I posted multiple quotes). So the question becomes "what is a good way to explain these topics on LifeWiki for newcomers / for interested-but-not-invested readers". I believe 'condition' is more explanatory / intuitive than 'transition' (and is in common use).
• Basically the same holds in the other case you mention. 'active object' is in common use, it was not invented by me, other people have used it. So the question becomes "what is a good way to explain these topics on LifeWiki for newcomers / for interested-but-not-invested readers". I believe that 'signal' is unnecessary distraction, unless you actually want to talk communication stuff.

The short version is that neither of these issues is relevant here. Those are different issues, and in both cases the alternatives are already in common use.
In this case, the only common name is 'logarithmic replicator rule', as far as I know.

[Haycat2009]

keeping inaccurate info

The name of something (a rule, a pattern, a corporation, etc.) does not generally provide any particularly accurate information. This is normal. Names do not have to make sense.

So are you implying that I can name several unnamed rules weird things, like Sus Rule for B38/S234c, Air Jordan for B34567/S456 and Chicken nugget for B3/S23/8? Even rule names do not get so far. They are supposed to symbolise the essence of a rule. And log replicator rule is meaningless by this defination.

[confocaloid]

Names of things are not required to mean or "symbolise" anything at all. If several people other than you used the same name consistently over time to refer to the same thing, that suffices to keep the name on the wiki.

As a pseudorandom example, the recently-coined name 'Grounded Life' for B35/S23 does not really make a lot of sense (especially if you just google it, read through search results and think about it). Personally I'm not sure it is common enough to be kept on LifeWiki. Maybe a simple rulestring 'B35/S23' (which is very short in this case) would do much better for the many readers who don't really know anything yet about the rule.

[Haycat2009]

keeping inaccurate info

The name of something (a rule, a pattern, a corporation, etc.) does not generally provide any particularly accurate information. This is normal. Names do not have to make sense.
The existing name is commonly used and there are no other commonly used names (unless you count the rulestring as a name, which you should not).
Keeping the existing pagename provides accurate information "this rule is commonly referred to by this name".

Let me give you an example. In 2022 Cribbage does not exist, so the p19 category was useless, and attempts to add cribbage was spam. In 2023, Cribbage was found, so we can say that p19 oscillators exist, and we can add it. Likewise, we only found out that the replicator was a sqrt replicator recently. Why not correct it?

[Haycat2009]

Names of things are not required to mean or "symbolise" anything at all. If several people other than you used the same name consistently over time to refer to the same thing, that suffices to keep the name on the wiki.

As a pseudorandom example, the recently-coined name 'Grounded Life' for B35/S23 does not really make a lot of sense (especially if you just google it, read through search results and think about it). Personally I'm not sure it is common enough to be kept on LifeWiki. Maybe a simple rulestring 'B35/S23' (which is very short in this case) would do much better for the many readers who don't really know anything yet about the rule.

Highlife has the same amount of sense as Grounded Life, but it is still preferred. Anyways, the alias has already been added to Lifeviewer, so it is obviously sensible enough.

[confocaloid]

Highlife has the same amount of sense as Grounded Life, but it is still preferred.

HighLife was investigated/discussed by more people who used the name over time in discussions. B35/S23 was named recently.

In 2022 Cribbage does not exist, so the p19 category was useless, and attempts to add cribbage was spam.

In 2022, "attempts to add cribbage" were nonexistent AFAIK https://conwaylife.com/w/index.php?titl ... e=Cribbage

In 2023, Cribbage was found, so we can say that p19 oscillators exist, and we can rename it.

Rename what?

[Haycat2009]

Highlife has the same amount of sense as Grounded Life, but it is still preferred.

HighLife was investigated/discussed by more people who used the name over time in discussions. B35/S23 was named recently.

If B36/S23 was named recently, you would still be doing this.

In 2022 Cribbage does not exist, so the p19 category was useless, and attempts to add cribbage was spam.

In 2022, "attempts to add cribbage" were nonexistent AFAIK https://conwaylife.com/w/index.php?titl ... e=Cribbage

That does not matter - it is a demonstration

[confocaloid]

If B36/S23 was named recently, you would still be doing this.

I would still be doing what, exactly?

In a parallel universe where B36/S23 was only named a few weeks or months ago and only few people looked into it, it would not make a lot of sense to go and replace all occurrences of the rulestring 'B36/S23' with the newly-coined name.

Even here and now, in this particular universe, it still makes sense to keep the rulestring 'B36/S23' in places where HighLife is mentioned (so that a reader can understand what is the rule before/without following the link).

That does not matter - it is a demonstration

That definitely matters if you want people other than yourself to understand your posts here.

You did not explain what you mean by 'attempts to add cribbage was spam'. There were no such attempts in 2022 as far as I know.
Also you did not answer my question --

In 2023, Cribbage was found, so we can say that p19 oscillators exist, and we can rename it.

Rename what?

[Haycat2009]

If B36/S23 was named recently, you would still be doing this.

I would still be doing what, exactly?

In a parallel universe where B36/S23 was only named a few weeks or months ago and only few people looked into it, it would not make a lot of sense to go and replace all occurrences of the rulestring 'B36/S23' with the newly-coined name.

Even here and now, in this particular universe, it still makes sense to keep the rulestring 'B36/S23' in places where HighLife is mentioned (so that a reader can understand what is the rule before/without following the link).

That does not matter - it is a demonstration

That definitely matters if you want people other than yourself to understand your posts here.

You did not explain what you mean by 'attempts to add cribbage was spam'. There were no such attempts in 2022 as far as I know.
Also you did not answer my question --

In 2023, Cribbage was found, so we can say that p19 oscillators exist, and we can rename it.

Rename what?

Demonstrations are not real life, so essencially no one cares - if you still do not understand, think of it as a thought experiment. 'attempts to add cribbage was spam' is an example, much like, say, "Can only see 2-d objects projected" in the case of Plato's Cave. If you do not understand why I used plato's cave, it is to show you what a thought experiment is.

[confocaloid]

https://xkcd.com/2595/

In 2022 Cribbage does not exist, so the p19 category was useless, and attempts to add cribbage was spam.

Such attempts could be considered spam (e.g. if it was some non-CA-related content). Or the page could be moved to another pagename, if it happened to be about a notable topic added under newly coined name. Or the page could be moved to user namespace. Or something else.

In 2023, Cribbage was found, so we can say that p19 oscillators exist, and we can add it.

AFAIK the current name of that p19 was chosen by the discoverer.
Note that they did not add the name to LifeWiki themselves - someone else did.

Likewise, we only found out that the replicator was a sqrt replicator recently. Why not correct it?

Naming the p19 'cribbage' amounts to naming a new clearly notable oscillator that had no previous common name, mostly because it did not exist.
The Logarithmic replicator rule is old, and already named for a long time, and there are no other common names.
Hence your "Likewise" is wrong: the two cases are different.

[Haycat2009]

https://xkcd.com/2595/

In 2022 Cribbage does not exist, so the p19 category was useless, and attempts to add cribbage was spam.

Such attempts could be considered spam (e.g. if it was some non-CA-related content). Or the page could be moved to another pagename, if it happened to be about a notable topic added under newly coined name. Or the page could be moved to user namespace. Or something else.

In 2023, Cribbage was found, so we can say that p19 oscillators exist, and we can add it.

AFAIK the current name of that p19 was chosen by the discoverer.
Note that they did not add the name to LifeWiki themselves - someone else did.

Likewise, we only found out that the replicator was a sqrt replicator recently. Why not correct it?

Naming the p19 'cribbage' amounts to naming a new clearly notable oscillator that had no previous common name, mostly because it did not exist.
The Logarithmic replicator rule is old, and already named for a long time, and there are no other common names.
Hence your "Likewise" is wrong: the two cases are different.

Inaccuracy should be corrected. Especially if we are now learned and aware that it is wrong.

[confocaloid]

There is no inaccuracy in the pagename OCA:Logarithmic replicator rule. New knowledge about the topic (the Logarithmic replicator rule, in this case) does not change the fact that this rule is commonly referred to as 'Logarithmic replicator rule'. The pagename is accurate, exactly because it gives the common name. (And there are no other common alternatives actually used by multiple other people in actual discussions when they are actually talking about the rule rather than about its name.)

Inaccuracy should be corrected

[hotdogPi]

AFAIK the current name of that p19 was chosen by the discoverer.

While mostly irrelevant to this discussion, I was actually the one who named it.

[calcyman]

At present, the name of the page is "OCA:Logarithmic replicator rule" and the replicator embed is labelled "The namesake logarithmic replicator," yet my result is shown in the section OCA:Logarithmic replicator rule#Growth_bounds.

At present, for the page's title, we have three main options before us:
1. Leave it as-is forevermore (this is incorrect, but retains consistency with the many existing mentions. Proposed by Confocal)
2. Move it to "OCA:B36/S245" (more clinical, doesn't note the reason for its interestingness. Proposed by Galoomba)
3. Move it to "OCA:sqrt replicator rule" or (in keeping with O(sqrt(log(t)))) "OCA:Θ(√t) replicator rule" (Proposed by me)

I agree with option 3. DroneBetter's analysis is much more thorough and correct, and I don't think that a rule should be named simply as a result of a mistake that I made in its naming a decade ago.

[confocaloid]

I prefer option 1 (leave the current name). I would agree with option 2 (use the rulestring) if there was a good chance that this will settle the issue and the article title will remain a rulestring.

(Re: option 3) I tried several web search engines, and found only three places where the proposed alternative name appears:

• this forum thread
• LifeWiki:Tiki_bar#"logarithmic"_replicator_rule_and_prospect_of_"OCA:rule_225"_page
• a merge request by DroneBetter: https://gitlab.com/hatsya/open-source/c ... equests/88

I did not find any prior uses of the proposed alternative name in (non-meta) discussions about the rule itself.
In comparison, there are multiple prior uses for the existing name 'logarithmic replicator rule' (not only on this website).

I think one needs either
(a) to provide a convincing explanation for why in this case the existing common recognizable name should be replaced by a new name not in common use,
or (b) to admit an inconsistency in how different issues are handled.

[...]
Maybe you could say something like "fixed-vector" for spaceships where both speed and direction are fixed. But that sounds kind of contrived. Maybe there's a better new term out there to be found... but usually this kind of classification causes fewer terminology wars, when all the terms have already come into common use.

[I've been trying to get the "self-supporting" vs. "self-constructing" distinction into common use for several years now, but I'm not sure that I've actually succeeded yet...!]

[confocaloid]

[...] I did not find any prior uses of the proposed alternative name in (non-meta) discussions about the rule itself.
In comparison, there are multiple prior uses for the existing name 'logarithmic replicator rule' (not only on this website). [...]

I readded the "proposed move" template, after the page was moved. The displayed Catagolue name was changed (in this commit). However, the new name was not used in actual discussions about the rule or about objects in the rule, and is a modification (replacement of a single word) of the previous name 'logarithmic replicator rule' which is in use and is recognizable.

[Nathaniel]

An executive decision (as per Step 5 of LW:DR) has been requested, so here I am to make it.

Executive decision: This page will be moved to OCA:sqrt replicator rule. Its introductory paragraph should mention the common-usage name "logarithmic replicator rule" as well though, and the redirect from OCA:logarithmic replicator rule should be kept.

Please keep in mind the procedure outlined in LW:DR going forward from here: attempts to undo or side-step this executive decision (e.g., moving the page back) without proceeding as in Steps 6 and 7 will result in disciplinary actions (e.g., a temporary ban).

[confocaloid]

https://conwaylife.com/w/index.php?oldid=142715
https://conwaylife.com/w/index.php?diff ... did=142541

I do not understand why the executive decision above is implemented (incorrectly; the title is capitalized incorrectly) by a user who is spamming the wiki with misspellings.

[DroneBetter]

a user who is spamming the wiki with misspellings.

I think Haycat did that due to the page with the correct capitalisation being occupied by a redirect scheduled for speedy deletion. They seem to be editing in good faith like the rest of us, I don't know whether they might be dyslexic or an ESL speaker, remember we were all young and naïve once. You seem to care a great deal about minor things and begin contentions that lead to all parties involved becoming frustrated, calm down somewhat and choose your battles wisely.

[confocaloid]

You seem to care a great deal about minor things

I don't agree these are minor things. Either there are meaningful guidelines (i.e. does LifeWiki aim to document actual common usage or prescribe it? Is it okay to spam pages with the same misspelling again and again, even after one is notified that the spelling is incorrect?), or there are no meaningful guidelines and there be chaos.

(Apparently other sides also don't view these issues as minor things. Otherwise there would be no dispute to begin with.)