Thread for your miscellaneous posts and discussions

[confocaloid]

First things first: I don't know what you mean by "rulespace is the most similar to life". (Reading "life" as referring to B3/S23, it's a specific cellular automaton, while 'rulespace' normally refers to a set of multiple different CA.)

To be able to compare two different cellular automata for "similarity", one would need a clear definition of similarity.

For example, if you evolve many soups in B3/S23 (Conway's Life) and keep track / count how often every isotropic 3-by-3 condition occurs during the evolution, then the condition "S4c" unambiguously would be the least frequent condition, for many reasonable choices of parameters:

Code: Select all

1 0 1
0 1 0
1 0 1

From this, one can expect that many reactions from B3/S23 (Conway's Life) would continue to work in exactly the same way in B3/S234c (where the rule "D4c" is replaced by the rule "S4c", so that the middle cell in the above situation survives instead of dying).

Likewise, one could repeat measurements (evolving many soups and measuring frequencies of 3-by-3 conditions) for B3/S234c, and again observe that "S4c" is the least frequent condition for many reasonable choices of parameters.

The above reasoning suggests that B3/S234c can be said to be "the most similar ruleset to" B3/S23, within the space of all two-state isotropic CA with range-1 Moore neighbourhood.

However, B3/S234c is actually very different from B3/S23 in another sense:

• Switch engine based technology no longer works in B3/S234c. (Contrast with B3/S236e, where switch engine based technology does exist.)
• The sparky alien p62 oscillator doesn't exist in Conway's Life, and can be used for a p62n stationary technology; this can be compared to p30n and p46n technologies in Conway's Life.
• "Checkerboard still lives" exist in B3/S234c, but not in Conway's Life. (3-by-n and 4-by-n checkerboard still lives are glider constructible for all n >= 2. I don't know much about constructibility of wider checkerboards.)

So, I guess, the short answer is "depends on your definition of similar".

Guess this is a change of topic from the previous discussion, but kind of a silly question which I have been brewing over: which Outer-totalistic/Isotropic non-totalistic rulespace is the most similar to life (In terms of the commonness of patterns, spaceships, evolutions of common sequences, etc.)? I have a suspicion that it's Conway++, but won't discount Pedestrian Life or some other variant.

Edit: Also, what would it take to take advantage of Pedestrian Life's numerous infinite growth patterns to construct some sort of alien universal constructor - or at least make universal constructor(s) in other rules?

Edit: Also, what would it take to take advantage of Pedestrian Life's numerous infinite growth patterns to construct some sort of alien universal constructor - or at least make universal constructor(s) in other rules?

Universal constructors exist in OCA, but you'll need more components to make one:

• Reflectors.

• Common spaceship reactions (to actually construct things).

• Recipe (of spaceship reactions).

Or optionally (for universal computer):

• Logic gates.

• Diagram (of the computer).

[Citation needed]

Thanksgiving is coming and I'll be leaving the forum now.

[Vlev2]

which rule is able to hold this p2 spaceship

000101
000010
111000
101000

[b-engine]

which rule is able to hold this p2 spaceship

000101
000010
111000
101000

Use RLE enclosed in code tags or pattern tags to post patterns. To summon pattern tags, click the second rightmost button.

This is the exact pattern you posted:

Code: Select all

3bobo$4bo$3o$obo!

[confocaloid]

which rule is able to hold this p2 spaceship

000101
000010
111000
101000

Not sure whether I understand the request. Do you mean like this?

Code: Select all

x = 10, y = 2, rule = B2e3i/S1c2ce
3o4bobo$obo5bo!

The apgcode of this spaceship is xq2_25.

This spaceship exists in many cellular automata of the same type; all of the following living conditions are optional, independently from each other:

Code: Select all

optional birth conditions:
  B2i B2k B2n B3a B3c B3j B3k B3n B3q B3r B3y
  B4a B4c B4e B4i B4j B4k B4n B4q B4r B4t B4w B4y B4z
  B5a B5c B5e B5j B5k B5n B5q B5r B5y B6a B6c B6e B6i B6k B6n B7c B7e B8
optional survival conditions:
  S0 S1e S2i S2k S2n S3a S3c S3e S3i S3j S3k S3n S3q S3r S3y
  S4a S4c S4e S4j S4k S4n S4q S4r S4t S4w S4y S4z
  S5a S5c S5e S5i S5j S5k S5n S5q S5r S5y S6a S6c S6e S6i S6k S6n S7c S7e S8

See https://www.ibiblio.org/lifepatterns/neighbors2.html for the table explaining the notation.

[confocaloid]

"The local community on these forums" is different from "the CA community".

I certainly disagree re: "general CA theory". Just because some people here don't care much about theory, doesn't in any way imply that people in the wider CA community are not interested in theory. There is definitely interest in theoretical questions, which you can see yourself by looking through the many existing sources and the many discussions on those topics.

I also disagree on several other points, including what makes something "impressive" or "interesting". A solution to some question can be impressive and interesting, even if it doesn't immediately lead to any pattern that would "look impressive".

[hotcrystal0]

How many people have a profile picture that can be synthesized?

[confocaloid]

How many people have a profile picture that can be synthesized?

A 17G synthesis of the alien p62 using a 5G synthesis from Catagolue for the four-block constellation: [...]

Code: Select all

#C [[ GRID THEME MCell ZOOM 6 STOP 200 HARDRESET  ]]
x = 132, y = 132, rule = B3/S234c
107bo$107bobo$107b2o8$113bo$21bo90bo$22bo89b3o$20b3o3$28bobo$12bo16b2o
$10bobo16bo$11b2o$118bo$118bobo$bo116b2o$2bo$3o4$115bo$113b2o$114b2o3$
91bo$91bobo$91b2o20$79bobo$79b2o$80bo2$63bo14bo$64b2o10b2o$63b2o12b2o
31$36b3o$38bo$37bo7$16b2o$17b2o$16bo4$129b3o$129bo$12b2o116bo$11bobo$
13bo$119b2o$102bo16bobo$101b2o16bo$101bobo3$109b3o$17b3o89bo$19bo90bo$
18bo8$23b2o$22bobo$24bo!

[Aleph]

-S8 (sorry to immediately remove your gun, i just wanted to remove transitions that seemed redundant to me)

Code: Select all

x = 3, y = 3, rule = B2i3-cr4ekrw5a6ce7c/S23-a4cq5iq7e8
bo$obo$bo!

You don't need to worry "at all" about destroying "my stuff". I've learned from rulegolfing that we will pass by many "interesting objects": flipping a transition can easily bring a new interesting object to the rule. I often spend about half an hour trying out transitions before deciding on which object I want to highlight: everything usually goes to MDIOCA. Things like rulegolfing and soup searching are all about "sifting" through noise: there are many MANY signals in that noise that one can pull out, but a source of noise is needed to begin with.

[confocaloid]

[...] Things like rulegolfing and soup searching are all about "sifting" through noise: there are many MANY signals in that noise that one can pull out, but a source of noise is needed to begin with.

Speaking about noise, the source of noise should remain fully contained within your patterns and reactions.

It is fine to introduce obscurity and strange things into your patterns and reactions.

However, when you are discussing and explaining those patterns and reactions, whenever you are talking about those patterns and reactions, any unwanted noise, distractions, obscurity should be removed. Explanations should be clear, intuitive, coherent as far as possible at all.

Excessive counterintuitive jargon happens outside the patterns and reactions. It shouldn't happen at all. It is problematic noise, detrimental to understanding.
Excessive low-effort chatting happens outside the patterns and reactions. It shouldn't happen at all. It is problematic noise, detrimental to understanding.

[Aleph]

I just wanted to point out existence of other threads that might be usefully linked from the first post (which could also benefit from having the names of threads written explicitly, instead of just listing bare URLs).
A reader should not be forced to click the link to get an idea of where it leads to.

I have added descriptions and those other additional links to my post.

Also I'm familiar with some history, and in general following one forum rule does not make a good excuse to break another forum rule.
Posting in a known existing thread on the same topic is better than creating a duplicate thread.

Can't I argue the other way, and say that following that other forum rule does not make a good excuse to break the original forum rule?
Either way, I am extremely uncomfortable with setting off a landmine by necroposting, so if I have to break a rule no matter what, I choose to break the other rule instead.

[confocaloid]

When there is an existing forum thread dedicated to some specific topic,
and when the stuff you want to post deserves a new post at all,
and when the stuff you want to post would be ontopic in that existing thread,
and when posting the stuff you want to post does not otherwise break forum rules (e.g. does not cause two or more unnecessary consecutive posts in a row by the same person),

I'd say it's fine to bump the existing thread with an ontopic interesting post (even if the thread is long inactive).

See viewtopic.php?f=3&t=6610 for several examples of unnecessary duplicate threads.

There is no need to create a duplicate thread merely to avoid bumping an existing thread on the same topic.

Relevant:

Making a post in a thread: First of all it is important to note that this is an academic forum, not a chat or microblogging platform if you want a place for looser conversation. If you want to make more random posts, please do so in the Sandbox thread. In all other threads what this means is you should keep conversations relevant, productive, and "of interest to the community"

Some specific examples of discoveries that won't be of interest to this community include:
A new still life.
A methuselah that lasts fewer than 20,000 generations or is larger than a 20-by-20 box.
An alternate stabilization for a reaction (unless it is smaller than the smallest currently known).

In OCA this list also includes:
A rule where patterns die out almost instantly.
A Explosive rule with an uninteresting explosion.
A random still life.
In both OCA and CGoL it is also good practice to DM someone if you are just making a comment or requesting an edit to someone's post.

Posting a new thread: Topics are held to a higher level than forum posts, and are the first things people see when going into a forum so it is important to make them clear, informative, and to be about a topic where having a forum dedicated to it would be useful.

When making a new thread it is important to ask yourself a few things. "
Am I asking a question that just needs a 1 comment answer?" If yes you should ask it in one of the threads dedicated for questions such as
Thread for your website-related questions viewtopic.php?f=3&t=5210
Thread for your basic questions viewtopic.php?f=7&t=2036
Thread for your basic non-CGOL questions viewtopic.php?f=11&t=2238
Does the post I am making have enough meat to have a thread for it? If no then you should post it in the sandbox or in:
Thread for your unsure discoveries viewtopic.php?f=2&t=576
Miscellaneous Discoveries in Other Cellular Automata viewtopic.php?f=11&t=1971

When making a thread it is important to think, does what I am posting about have enough meat for other people to discuss? If it does not then it is better to not make a thread about. For example although the P41 in CGoL is incredibly cool, it would not be a good thing to make a thread about because there is no place for further discussion

Has this thread, or a thread almost identical to this already been made?
Post it in the other thread

Make sure that a user can understand what will be in the thread from reading your topic post. Stuff like "This is rule xyz, post stuff about it here", "This is the Votable challenge", or "here is a pattern I drew" would not be informative enough.
Although the first 2 would be fine as titles, they would need to explain why they are interesting, or what they are.

[Aleph]

I just reread calcyman's rant: you do have a point there since I literally just noticed (yes it took over 7 years) that the landmine was "pointless necroposting" and not "necroposting". However, "creating new threads to avoid necroposting" was not identified as a landmine, and what necroposting is "pointless" or "not pointless" is subjective, so I would still like to play it as safe as possible by not necroposting no matter what.

[confocaloid]

If you would like to play it as safe as possible, then "making posts with a dozen or two of small separate RLE snippets per post and actively participating in chatty forum games" might not be the best possible plan.

[Aleph]

I wasn't clear enough. I meant playing safe with respect to the necroposting issue.

[confocaloid]

I wasn't clear enough. I meant playing safe with respect to the necroposting issue.

Bumping an ontopic thread is a minor issue. Spamming the forums with extremely low-effort posts is a major issue.

In particular, the long strings of small RLE snippets in your posts are really annoying.
Can you instead put these patterns in a single snippet, like this?

Code: Select all

headerless_rle_1 rulestring_1
headerless_rle_2 rulestring_2
headerless_rle_3 rulestring_3
...

Neither of those objects deserves much more attention than the other posted objects.

If you think otherwise, then choose what you believe to be the most interesting object and post it first, and append other RLEs as comments in the same snippet.

[hotdogPi]

LifeViewer won't run it if formatted the way you're suggesting.

[confocaloid]

LifeViewer won't run it if formatted the way you're suggesting.

Sure, but for example those people who contribute to the 5S collection also share their contributions in a format that is not viewable in LifeViewer.
As soon as one's posts become "a dozen or two of small objects in different CA", it's time to think about switching to a different format.
Related: viewtopic.php?f=9&t=6770&p=198442#p198442 "Multiple small patterns in several different CA in a single snippet / Medium-size pattern collections"

[Aleph]

[...] a rule which is the sum of [...]

I prefer type safety. Which means in particular that

• the operations defined for sets (union, intersection, set difference) cannot be applied to integers,
• the operations defined for integers (sum, difference, product, ...) cannot be applied to sets,
• and it is technically incorrect to say "transition" when you actually mean either "rule" or "condition".

It is still possible to define a different notion of "sum" for sets (e.g. for two sets of integers A and B, one could define their sum to be { x + y : x in A, y in B }. But that will be a different operation, and there is no such operation "by default" unless you define it explicitly.

Ambiguity and ill-defined obscure jargon cause many problems, and it is better to catch the errors early, before those errors cause significant damage.

This discussion has reminded me of so-called "operator overloading".

[confocaloid]

This discussion has reminded me of so-called "operator overloading".

Which also must be defined explicitly, and deserves to be defined in some nonarbitrary consistent way, and deserves to be clearly documented.

(edit: in a programming language that is limited to ASCII,)
It could make sense to overload '+' to do set union (edit: or another commutative operation). (Deserves to be documented explicitly.)

It doesn't make much sense to overload '-' to do set union. That would be confusing. People will expect that '-' is something non-commutative at least.

[Aleph]

This discussion has reminded me of so-called "operator overloading".

Which also must be defined explicitly, and deserves to be defined in some nonarbitrary consistent way, and deserves to be clearly documented.

It could make sense to overload '+' to do set union. (Deserves to be documented explicitly.)

It doesn't make much sense to overload '-' to do set union. That would be confusing. People will expect that '-' is something non-commutative at least.

I agree with that: Operator overloading should be made clear. That being said, we *do* have 'U' for set union, which I think makes overloading '+' for set union to be a bit (or a lot) unnecessary here.

[confocaloid]

Does anyone know where "... (blockus or whatever)" comes from? Is this a name actually chosen by the creator of the CA? Or this is invented by someone else?

It might be better to, you know, actually give people opportunity to name stuff they created or discovered.

[CARuler]

Does anyone know where "... (blockus or whatever)" comes from? Is this a name actually chosen by the creator of the CA? Or this is invented by someone else?

It might be better to, you know, actually give people opportunity to name stuff they created or discovered.

check the first post

[confocaloid]

No, it does not explain where the part "(blockus or whatever)" came from.

Did you just invent a name for a CA found by someone else?

Does anyone know where "... (blockus or whatever)" comes from? Is this a name actually chosen by the creator of the CA? Or this is invented by someone else?

It might be better to, you know, actually give people opportunity to name stuff they created or discovered.

check the first post

[CARuler]

LuveelVoom's 8yr old sister made it

came from discord