How do we welcome newcomers without discouraging them?

[ThePlayzr]

Apologies if some of this sounds a little brutal, but it's intended constructively.

* The thread asks how not to discourage newcomers, but many proposed solutions amount to more rules, greater visibility of rules, or compulsory exposure to rules.
(irrelevant text)

On points about "rules", I would consider this... academic research may have methodological rules, but it cannot have rules about what you're permitted to find interesting or what questions you're permitted to ask.

I get what you mean, but giving people the basic information and rules that they need isn't 'discouraging' them. It's just a normal and essential part of joining any community.
If newcomers don't see the rules, they break them, not because they mean to, but because they don't know. We have to expose them to the rules at some point, and it's much easier to make them read them before they make a bunch of posts then afterwards. Otherwise, the forum gets cluttered and progress towards notable discoveries actually slows, because people have to spend time dealing with this.
A simple redirect or even tasks you have to do before joining isn't discouraging newcomers, it's just making sure they know the rules before joining. And with many newcomers being very young, expecting them to hunt down a small link is unrealistic. Visibility prevents problems before they happen.

To put that more bluntly, "make up your (removed profanity) mind"...

Because “we desperately want new people” and “please learn what we've already decided is worthy of being interested in” are in fairly obvious tension.

I get the point you’re making, but please avoid profanity, even shortened - although you are frustrated, this is not needed or wanted.

And there’s no contradiction here. Making sure newcomers actually see the rules isn’t discouraging them; it's just what makes sure newcomers start off the right foot.

Nothing, diddley squat. No rebuke or rebuttal, just a resounding silence. A few nudges asking for a little introspection being met with silence looks increasingly like classical ossification.

The forum rules repeatedly describe this as an “academic forum”.

So, genuinely: enlighten me.

If this really is an academic forum, where are the reproducibility standards?

Reproducibility and shared expectations all rely on people actually knowing the guidelines in the first place. Making those guidelines more visible isn’t discouraging newcomers, it’s making sure everyone starts with the same baseline.
Also, please avoid necroposting. If you have something to add, please edit your previous post.

[Marrow16]

Thank you for proving every point I made

[R2INT]

I agree with your statement that the LifeWiki is currently being overfiltered. Ideally, the LifeWiki should contain enough information so that newcomers can find enough information to meaningfully contribute, instead of having that information hidden away in user pages or forum posts that eventually get buried away. Otherwise, you end up with problems where newcomers repost patterns that were already discovered but not documented, or newcomers running a pattern search that someone else already completed.

For example, currently there's some work on a quadratic growth pattern in Travelling Ts (B3/S23-a5) using a p20 oscillator to redirect the spaceships to produce an over-unity reaction. Both the OCA:Travelling_Ts page and the user page do not contain information on the synthesis of the p20 or the high-period rake that's being used as a source for the gliders, despite listing quadratic growth as one of the goals and both of the patterns being known and posted on the forums.

EDIT: Another problem that I found on the wiki is that if you're the only person exploring your CA, how are you supposed to document your own discoveries? Storing it locally on your computer works, but then if another person wants to explore your rule, they will need your files to know what's new vs. what's already known. Then the LifeWiki becomes the next major place for you to document that information, but there are rules against self-reporting on the wiki, and the other user is unlikely to document your discoveries unless they find the rule very interesting.

[TheWayOfTheCon]

I agree with your statement that the LifeWiki is currently being overfiltered. Ideally, the LifeWiki should contain enough information so that newcomers can find enough information to meaningfully contribute, instead of having that information hidden away in user pages or forum posts that eventually get buried away. Otherwise, you end up with problems where newcomers repost patterns that were already discovered but not documented, or newcomers running a pattern search that someone else already completed.

I don't think its feasible to have every notable pattern in CA have its place on the LifeWiki, because there's just so much. And even then I doubt people will want to write/read all of that.

I have a different idea to filter some of the "is this known?" posts on the forums. Maybe the website can detect RLEs for notable patterns during writing and link them to said pattern's page on LifeWiki or Catagolue.

[R2INT]

I agree that not all notable patterns (like the individual 5S spaceships) need their own place in the wiki. However, the wiki should have enough information to meaningfully change what people can do with the rule (e.g. periodic reaction for Glider 317, synthesis for the p20 reflector, the Snark) without overloading the wiki with lots of notable but low-utility discoveries.

Additionally, CA patterns are relatively small by modern standards--most notable patterns range from a few hundred bytes to the low kilobytes, while even the large engineered patterns like silverfish only take up hundreds of kilobytes. The entire uncompressed pattern collection on the wiki is about 24 MB. By comparison, forum attachments such as large random patterns and screenshots can consume considerably more space.

So I don't think raw storage is a compelling reason to strongly filter technical information. The harder problem is organizing it well enough that people can find the information they actually need.

[Marrow16]

I'm sorry, but "Is this pattern known?" just shouldn't be a forum question in 2026. It's silly easy to reduce a pattern to a canonical form (i.e. trimmed of surrounding dead cells) and then hash it. The only thing to then do is agree the hashing (that's the reproducibility!). Just for example (Golang)...

Code: Select all

func (p Pattern) Hash() [32]byte {
	// 8 bytes for dimensions + packed alive cell bits...
	data := make([]byte, 8+(len(p.Cells)+7)/8)
	binary.LittleEndian.PutUint32(data[0:4], uint32(p.Width))
	binary.LittleEndian.PutUint32(data[4:8], uint32(p.Height))
	for i, alive := range p.Cells {
		if alive {
			data[8+i/8] |= 1 << (i % 8)
		}
	}
	return sha256.Sum256(data)
}

And running that across the current patterns catalogue you'll see there's already hundreds of duplicates
A searchable corpus would save everyone's pain - and eliminate endless rules on what's notable or not

[TheWayOfTheCon]

It's silly easy to reduce a pattern to a canonical form (i.e. trimmed of surrounding dead cells) and then hash it. The only thing to then do is agree the hashing (that's the reproducibility!). Just for example (Golang)...

Code: Select all

func (p Pattern) Hash() [32]byte {
	// 8 bytes for dimensions + packed alive cell bits...
	data := make([]byte, 8+(len(p.Cells)+7)/8)
	binary.LittleEndian.PutUint32(data[0:4], uint32(p.Width))
	binary.LittleEndian.PutUint32(data[4:8], uint32(p.Height))
	for i, alive := range p.Cells {
		if alive {
			data[8+i/8] |= 1 << (i % 8)
		}
	}
	return sha256.Sum256(data)
}

A searchable corpus would save everyone's pain - and eliminate endless rules on what's notable or not

That's pretty much what the website Catagolue does, and you can search objects with it too. But not everyone's gonna look at the website first before posting their "is this known?" post, many probably wont know Catagolue can be used for that.

[Marrow16]

I'm outta here... but at least I leave enlightened... I now have a pretty good idea why this community has declined over the years... far too many flipping people (oops, sorry - i forgot, no words beginning with "f" - or ucks sake) who want "more rules" (probably so they can police them), revel in their folklore and dismiss anything/everything with "because that's how we've always done it".

And always nice to repay enlightenment, so I leave y'all with a quote...

Nothing is so fatal to the progress of the human mind as to suppose that our views of science are ultimate; that there are no mysteries in nature; that our triumphs are complete, and that there are no new worlds to conquer.
- Sir Humphry Davy

Caio

[rabbit]

Your tone in this thread and the fact that you are leaving so soon make it seem like you are more interested in scolding us than working to make this community a better place. In addition, one of the problems you highlighted (LifeWiki being out of date) can be solved by editing LifeWiki and working to improve it. It bears repeating that conwaylife.com is a community of volunteers and cellular automata enthusiasts, not an institution or company.

No, collecting every single one of the patterns devised over the course of the 50+ years of CA research into a single searchable database is not "silly easy", which is why nobody has done it. We've gotten a lot closer, though, with the modern versions of jslife collecting oscillators and spaceships, and Catagolue collecting glider syntheses and guns. The properties that are considered valuable in different classes of patterns vary, and organizationally it's better to have multiple targeted databases rather than just one. This very forum is as close to a complete pattern collection as we have, and while PHPBB's search feature has its downsides, things like conduit notation make it easier to find out whether something is a rediscovery or not.

Pointing users to the rules is important for keeping the community clean, but is perhaps overemphasized due to its simplicity.

----------------

Back to the intended topic of this thread, I think that the best way to engage with incoming Lifenthusiasts is to be understanding, and interact with their patterns, which I try to do when I can. However, I have some potential ways to improve the onboarding experience as well:

1. Make the pinned threads stick out more. Currently, the only difference between a pinned thread and the rest of the threads appears to be the shape of the post icon, which is easy to overlook. Perhaps giving pinned threads their own section, or highlighting them more obviously would make it easier for users to find the appropriate megathread.
2. Direct new users toward helpful resources. I'm extremely tempted to try and write a "Starter Guide" myself on the wiki to assist with this mission, but I do not have much knowledge of search programs.
3. Explain concepts to new users. I have occasionally seen more experienced users reply that something is merely not notable, which feels too blunt. Instead, it might be worthwhile to point out that there are a lot of patterns that satisfy the supplied property, or encourage them to find similar patterns that are more notable. (e.g. smaller methuselah with a longer lifespan)

EDIT: Another thing that would be helpful is renaming/replacing the "Thread for your Useless Discoveries" thread with something more neutral. The OCA board is a good model for this with its "Miscellaneous Discoveries in Other Cellular Automata" thread.

(Also changed some verbiage to be more precise because my ideas are still in flux somewhat)

[TheWayOfTheCon]

Direct the user toward helpful resources. I'm extremely tempted to try and write a "Starter Guide" myself on the wiki to assist with this mission, but I do not have much knowledge of search programs.

There are several tutorials for using search programs on the wiki, so you could use that.

[ThePlayzr]

Another idea: what about a 'Welcome bot' that sends a PM to users when they first make an account, that gives them a simplified version of the rules, things to look at, a link to the forum rules, and make them feel welcomed?
Also, I agree with the idea to make the forum rules 'stand out' more than they do when someone first makes an account.

[otismo]

Nothing, diddley squat. No rebuke or rebuttal, just a resounding silence. A few nudges asking for a little introspection being met with silence looks increasingly like classical ossification.

https://conwaylife.com/forums/viewtopic.php?t=2092

The forum rules repeatedly describe this as an “academic forum”.

So, genuinely: enlighten me.

If this really is an academic forum, where are the reproducibility standards?

i was feeling a bit ossified so i thot i wud weigh in on this discussion

this guy joined the forum in 2017 - he posted a handfull of times

nobody thought to ask him about his experience as a "member"

or even exactly what was his interest in being here to begin with

i understand that to operate efficiently and effectively as an academic forum

certain rules need to be followed by everyone or the whole thing will get gummed up and stuck

now the only reason cgol got huge was because of conway himself and a few expert mathematicians and programmers

this forum and this community are the goto resource for anyone who is seriously or even casually interested in ca & cgol

but to assume that everyone who joins is a "life enthusiast" is i feel an assumption that is way out-of-line -

people who do join however have the best potential of becoming Life Enthusiasts - myself included...

<3

[pcallahan]

You've got a point. IMO I wouldn't consider this place an "academic forum". A place where formality is expected, sure. But most of us aren't actual researchers putting out studies on this stuff, because at the end of the day cellular automata isn't a very complex subject. It can just be made complex.

I would avoid conflating "academic forum" with "peer reviewed journal" or "university department", which this website obviously is not. I would also avoid using the word "academic" (and do as far as I know), and consider it a hobbyist site that happens to include academics. Note: I am referring specifically to forum threads, not LifeWiki, which has a much higher threshold.

The precedent I would go back to is Usenet in the 1990s, for those who remember comp.theory.cell-automata. It was unmoderated but, like other newsgroups, could elicit testy replies for posts that were deemed uninteresting or irrelevant. There were no official standards, nor should there have been.

The "welcoming" element has to be self-motivated. If you post something and don't get traction, you can decide the website not for you (which is legitimate) and leave. Alternatively, you can try to promote the same idea and persuade others that it's interesting. Finally, you can spend more time reading and getting up to the state of the art before posting further.

I am not sure if I get a warmer reception because people recognize my name and old results. I haven't done anything that novel in over 25 years. If newcomers are being treated worse than I am for similar-quality posts, that is a problem. I haven't noticed, but I don't have that perspective.

I did want to comment on "reproducibility standards" because that part is largely built in. If you claim a pattern does something, you post it in a code block and everyone can see that it does that.

If you make some broader claim e.g. that a particular search algorithm is more effective for some class of problems than another, that is not as obviously reproducible. I have also never seen claims like that presented as more than conjecture.

If you claim a mathematical theorem, you back it up with a proof, just like any mathematician would.

If it's a calculation, e.g. frequency of small oscillators from random soups, then you present the table and ideally the code you used to generate it. You can suggest extrapolations but should not present them as mathematical theorems without a proof.

In my experience, most claims are presented with caution, barring a proof or an executable pattern. Reproducibility is the least of our problems.

We don't have the standards a journal would have, such as peer review. So it is not an "academic forum" in some formal sense (though I don't think that phrase itself is well-defined). We do adhere informally to standards of rigor and relevance. We also have Sandbox for anything else.

Personally I try to be welcoming and would like to see more newcomers, and would recommend not shooting anyone down, though I do not feel there is a need to feign interest either.

[dvgrn]

Another problem that I found on the wiki is that if you're the only person exploring your CA, how are you supposed to document your own discoveries? Storing it locally on your computer works, but then if another person wants to explore your rule, they will need your files to know what's new vs. what's already known. Then the LifeWiki becomes the next major place for you to document that information, but there are rules against self-reporting on the wiki, and the other user is unlikely to document your discoveries unless they find the rule very interesting.

This is the kind of point that's definitely worth taking a fresh look at now and then, to see if the current set of rules is -- on balance -- producing something close to the best results we can expect.

Rule-writing and rule enforcement are always, inevitably, a difficult balancing act. Forum and LifeWiki rules only get added when we collectively notice that there's a problem, and we make a guess that a particular rule will help address that problem without causing more trouble than it solves. Maybe sometimes that judgment is right, and maybe sometimes it's wrong -- but if we notice that something is wrong, we can always change the rules again.

Re: documenting one's own discoveries, I'd say that the next major place to store discoveries after a local hard drive is not the LifeWiki. It's a dedicated thread on the forums.

Once someone puts enough interesting discoveries into forum posts, especially with a well-maintained and active "status/intro" first post and some work put in to showing off recent progress, it becomes very easy for someone else to put together a LifeWiki page that links to those forum posts. The LifeWiki article probably won't happen as fast as any given enthusiastic researcher might want, but it's not an impossibly high bar for an article to make an appearance eventually.

We have quite a few years of experience with enthusiastic newcomers showing up and wanting to immediately document their own discoveries, before they've been in the community long enough to be able to judge reliably what most people are going to want to see showing up in LifeWiki articles.

The Perfect Is the Enemy of the Best-We-Can-Do
It would be great to be able to encourage more people to document every discovery from every corner of CA space. But in practice the LifeWiki doesn't seem to be the best place for that. If we open the notabiilty doors too wide, the LifeWiki ends up getting cluttered with questionable trivia -- and then (from what I've seen) people are less likely to use the LifeWiki as a resource. More important, people are also significantly less likely to want to contribute their own careful editing work, if the LifeWiki doesn't look like it's being maintained adequately.

So the notability rule definitely tries to encourage very careful "academic-style" contributions. Unfortunately that comes somewhat at the expense of enthusiastic-newbie contributions. It seems really important to me to conserve the energy and attention span of long-term dedicated LifeWiki maintainers -- and unfortunately that's a very finite resource. It's very easy for the current round of editors to get exhausted by explaining notability to too many newcomers at once... and it's very difficult to recruit a new round of equally competent editors, when an existing cadre of editors runs out of energy and gives up.

The existing rules are rather conservative. That's not because it's really the theoretical ideal situation for the LifeWiki not to change much. It's just because we're all volunteers around here, and the existing rules tend to make the LifeWiki change at a reasonably slow rate that our collective volunteer energy can pretty reliably keep up with, or at least keep an eye on.

[pcallahan]

Re: documenting one's own discoveries, I'd say that the next major place to store discoveries after a local hard drive is not the LifeWiki. It's a dedicated thread on the forums.

If someone has a large body of work of any kind that they want to make available, not just Life or CAs, the first thing that comes to my mind is a GitHub repo. It's easy to set up and it's a standard place to share code and other files. I also agree that LifeWiki should have a high threshold of acceptance, effectively a form of peer review.

A dedicated thread is also a good start, but I can't think of any reason today for anyone to believe their work is going to languish on their local drive. It may languish in obscurity on the Internet but at least it is available in principle.

[Hunter]

As a beginner: I feel like the main problem is that non beginners don't remember what its like to have no knowledge of any form of cellular automata.

With your knowledge now you would never ask if something like a "pulsar" has been discovered yet because you know that they are very common and that something like apgsearch exist that can search soups a billion times faster then a human ever could. (Which you wouldn't even know that its called a soup anyway). Beginners also don't have the tools to discover interesting things anyways. So the things they first share in the forum will be things they likely found manually or seen in a video.

And to the answer the question "How do we welcome newcomers without discouraging them?" welcome them to the server, answer there question/tell them what they found and then if they broke a rule just tell them so they can fix it next time. They will learn with time what is worthy of a forum post, what everything means etc.

[TheWayOfTheCon]

And to the answer the question "How do we welcome newcomers without discouraging them?" welcome them to the server, answer there question/tell them what they found and then if they broke a rule just tell them so they can fix it next time. They will learn with time what is worthy of a forum post, what everything means etc.

I agree that all of this will work, we've done this stuff and it has worked. But I think the point of this thread is to somewhat automate the process. Changes to the website that make it more newcomer-friendly and elaborate on what's known and how things work around here.

The home page is pretty lacking with that stuff as of now, just a few FAQs and a randomly-selected pattern. I think if the home page was more robust/helpful, people and especially newcomers would pay more attention to it.