Talk:Garden of Eden (cellular automaton)

Archives (Index)

2021

This page is archived by ClueBot III.

RfC on which version of lead to use

The consensus is that the version that starts more simply and is deliberately redundant is preferable to the the irredundant version. Cunard (talk) 22:27, 3 July 2016 (UTC)

The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

---

Which version of the lead (specifically, the first part of the lead, giving the definitions of a Garden of Eden and of an orphan) is better: the irredundant version or the version that starts more simply and is deliberately redundant? Secondarily, it would also be useful to get more opinions on whether the other cellular automaton terminology introduced in the lead and primarily covered by this article ("orphan", "twin", and/or "Garden of Eden theorem") should be boldfaced. —David Eppstein (talk) 20:19, 31 May 2016 (UTC)

• My own preference (per the above discussion) is for the deliberately redundant version. —David Eppstein (talk) 20:25, 31 May 2016 (UTC)

Double definitions don't help anyone. The previous version, which included one double definition introduced with "more precisely", and another introduced with "alternatively", was not helpful at all in explaining the concepts concerned. The current version of the text which the user now claims to prefer does not contain these redundancies, and thus the choice the user seeks comment on is not between an intro with redundancies and one without, but simply between different wordings of the intro. I do not see any issue with this introduction.

As for bold text, more opinions are not required to decide which words should be in bold. We have very clear guidelines, which tell us that only the first occurrence of the title and significant alternative titles (which should usually also redirect to the article) are placed in bold. Neither "orphan", nor "twin", nor "garden of eden theorem" are significant alternative titles, and thus they clearly shouldn't be in bold face. 79.158.212.99 (talk) 20:54, 31 May 2016 (UTC)

• The longer lead is better -- it does not read as redundant to me, and a non-mathematician might have some small chance of understanding it. I have no opinion on the bold. --JBL (talk) 13:44, 1 June 2016 (UTC)
• The shorter lead is not very comprehensible to me (and I am a mathematician, albeit one that doesn't specialize in this area). The longer lead is a vast improvement. I have no strong opinions about the bold issue, but I lean slightly towards not bolding them. Sławomir Biały (talk) 14:38, 1 June 2016 (UTC)

I also think the Eppstein version is better than the one by "Staszek Lem". Lem's proposed revision brings some discussion of the lattice into the very first sentence, and overall dwells too much on the concept of lattices, which I do not think is very essential for describing the subject of the article. Sławomir Biały (talk) 12:47, 2 June 2016 (UTC)

• (Answering the RFC call) - Definitions of both versions are bad. And I don't see why "long" version is "redundant", only a bit overly verbose in some places. I will not nitpick the problems which decrease comprehension of the text even for a person who vaguely remembers something about cellular automata; simply here is my version (with "modified" redundancy):

In a cellular automaton, a Garden of Eden is a pattern of the whole lattice of its cells that has no predecessor in any possible evolution of the automaton. John Tukey, who first conjectured existence of these patterns, named them after the Garden of Eden in Abrahamic religions, which was created out of nowhere.[2]

An finite subpattern of the whole (infinite) lattice is called an orphan if any pattern that contains the subpattern is a Garden of Eden. And conversely, it is proven that each Garden of Eden contains at least one orphan.

For one-dimensional cellular automata, orphans and Gardens of Eden can be found by an efficient algorithm, but for higher dimensions this is an undecidable problem. Computer searches have succeeded in finding these Gardens of Eden in Conway's Game of Life. The Garden of Eden theorem of Moore and Myhill states that a cellular automata on the square grid, or on a tiling of any higher dimensional Euclidean space, has a Garden of Eden if and only if it has twins, two finite patterns that have the same successors whenever one is substituted for the other.

Staszek Lem (talk) 17:12, 1 June 2016 (UTC)

• Of the two alternatives, the longer lead is clearer and easier to understand for this non-expert. A little redundancy is fine if it helps explain concepts more clearly. Clarity and ease of understanding in the lead are priorities for technical articles like these, per WP:TECHNICAL. I am with Staszek in that I think the prose could be made even more explicit for the non-expert. Boldface should only be used to indicate alternative names for the main topic. Per MOS:BOLD, it is better to emphasize jargon or words being defined by using the HTML <em>...</em> markup. --Mark viking (talk) 21:15, 1 June 2016 (UTC)

• I like Staszek's lead the best. The current one is okay, too. However, MOS:BOLD does suggest bolding terms in the first couple paragraphs if they are the names of redirects. So if Orphan (celluar automation) or something is a significant redirect here, it might make sense to bold it. That probably doesn't mean we should bold all possible terms which redirect here, though. Nat2 (talk) 04:22, 2 June 2016 (UTC)

• Second, longer version It's just better, Like Sławomir, I am a mathematician, and found the first needlessly terse and an impediment to easy understanding. My preference is to keep this RfC about the two proposed alternatives, and sort out any further modifications later. SemanticMantis (talk) 14:40, 2 June 2016 (UTC)

• Longer lead is clearly better. Can't decide about the bolding. EEng 04:30, 5 June 2016 (UTC)

• I much prefer the irredundant version. It says everything that needs saying, clearly, concisely, and once. The redundant version rabbits on without ever coming to the point – it nowhere says what a Garden of Eden configuration is, it uses "pattern", which it does not define. The irredundant version also uses "pattern", but in a (different) sense that is clear from the context. Maproom (talk) 06:51, 14 June 2016 (UTC)

On further consideration ... a source of confusion, with both versions, is that a Garden of Eden is a state of the whole lattice, but the reader is led astray by seeing, right at the top of the article, an image captioned "A Garden of Eden in Conway's Game of Life ..." and showing something finite. The second image is much better – its cropping style hints that it shows only part of something. Maproom (talk) 07:21, 15 June 2016 (UTC)

In the corresponding French article, it's the finite component, the "orphan", that's said to be named "jardin d'Éden". Maybe we should reach agreement on what the term means, before deciding how best to express it. Maproom (talk) 22:01, 17 June 2016 (UTC)

• Longer version. As someone who came here summoned by a bot, I had no idea what a Garden of Eden configuration was before reading the two versions of the lead. Understanding the shorter version was considerably harder. I suppose that for a reader who is well familiar with the subject the longer text will be redundant, but they will still be able to scan it easily (say, looking for a specific fact or link).“WarKosign” 07:26, 17 June 2016 (UTC)
• Dissatisfied either way. Though I am in fact interested in cellular automata and do know what a garden of Eden is, I had not seen this article. I think the entire article needs some serious editing, and neither lede so far presented for approval is satisfactory. Given the current intensity of disagreement however, I am not inclined to interfere, as I do not expect to be able to satisfy all parties. If, against all expectations, I am requested to participate, I'll think again, but as things stand, I am not holding my breath. Good luck all. JonRichfield (talk) 17:58, 18 June 2016 (UTC)

JonRichfield: please don't let the discussion above deter you from improving the article. There's disagreement, but it has certainly not descended to anything personal. The article needs work, that's why there's a discussion. Maproom (talk) 22:48, 18 June 2016 (UTC)

Yes, I agree. More people identifying problems and finding ways to improve the article is surely better than fewer. —David Eppstein (talk) 22:53, 18 June 2016 (UTC)

Fair enough. I'll be back in a couple of days. JonRichfield (talk) 10:59, 19 June 2016 (UTC)

Maproom Sorry to be so long. (RL). I have made a start and will present something maybe tonight. Question: I cannot find any general use of teh term "quiescent" and the description in the body is opaque. Help anyone? JonRichfield (talk) 09:07, 21 June 2016 (UTC)

For Conway's Life it is the same as "dead". Basically, if you have a region of cells all in this state, then nothing happens there. Google scholar finds many uses of this term. —David Eppstein (talk) 12:28, 21 June 2016 (UTC)

Maproom and David Eppstein Thanks for assistance and patience, and apologies for being so long getting underway. I have begun some proposed material that, to avoid cluttering the present talk page, I have saved in User:JonRichfield/Garden of Eden (cellular automaton). You are welcome to visit and to edit if desired, or take parts of it, modified according to taste. Note that some of the changes amount to the same material reworded or redistributed. For example, the lede would be much smaller because it currently contains material that would be of no value to a reader unfamiliar with the topic who first wishes to know whether it looks worthwhile to read on. I have only the introductory sections so far (and as yet needing a lot of editing; I am a slow writer) because there is no point creating essentially a new article without knowing what its reception would be. JonRichfield (talk) 19:29, 22 June 2016 (UTC)

---

The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.