Talk:Period-89 oscillator engines
Jump to navigation
Jump to search
Notability
I decided to make an article about this reaction after seeing that there are multiple posts about it on the forum, and several people contributed to it -- sufficient to write an article. I don't think there are very many such partials.
I'm sorry if my removal of notability template was received as impolite. At least I did attempt to explain in the edit summary why I consider the topic notable. When the notability template was added, there was no explanation, and the article did already contain forum links to support notability (I did not notice that some links were broken, sorry about that). Confocal (talk) 12:28, 5 April 2023 (UTC)
- Not a big deal -- after all, I did an equally impolite revert-of-the-revert -- it's just good to move to a discussion rather than a revert war at this point!
- This is a very thoroughly documented article, with a lot of work put into it. But I **believe** that there's not a lot of support for the topic being notable at all. So -- let's discuss!
- I'm thinking that many dozens of partials along these lines could be completed in similar ways -- and that the only reason that there aren't very many such partials is that there's a general consensus that they're not notable.
- Seems like most of the contributions that led to the creation of 768P89 were just out of curiosity, like "what's the best known way to supply a domino spark here?".
- So May13's contribution might get linked to from somewhere, say a hypothetical LifeWiki article about supplying sparks from Herschel conduits -- or a table of known oscillators of various periods, or something along those lines.
- It seems to me that we definitely don't need the clutter of the redirects from "2116P89" and "902P89", though -- those seem like non-notability squared.
- Regarding redirects and the current page name, I think these are more short systematic descriptions (Pentadecathlon IDs), rather than names. I agree that there might be a better way to name this page. Confocal (talk) 15:14, 5 April 2023 (UTC)
What's Implied By The Article's Existence
- The article is written as if all of the contributors are building up to the creation of something significant... and yet one of the premises of the last few years of oscillator-collecting has been, "Don't use glider guns for partial support (because that makes things too easy)".
- The idea behind the notability tag here is that it's actually kind of important to discourage the creation of more articles like this one, that make something non-notable look like it must be something notable somehow. Dvgrn (talk) 12:34, 5 April 2023 (UTC)
- I wonder whether the premise of not using glider guns was actually written somewhere. I do agree that glider guns make things easier. On the other hand, there is still a difference between partials that were actually completed as oscillators by people, and partials that were never completed -- and this difference is interesting at least from an editor's viewpoint. And from another viewpoint, even if you use glider streams, you still have to find a compatible way that works.
I do not think the article itself implies anything about significance -- for example, the article does not make any claims about the reaction beating any record. A reason for this reaction to be notable is that it did receive attention; it was completed as an oscillator, and was further reduced.
I think the only part in the current version that actually does mention significance of any kind is the notability message on the top -- the article itself does not claim any more significance than an article about a spaceship or an oscillator. Confocal (talk) 13:03, 5 April 2023 (UTC)
- I wonder whether the premise of not using glider guns was actually written somewhere. I do agree that glider guns make things easier. On the other hand, there is still a difference between partials that were actually completed as oscillators by people, and partials that were never completed -- and this difference is interesting at least from an editor's viewpoint. And from another viewpoint, even if you use glider streams, you still have to find a compatible way that works.
- That all seems true enough, as far as it goes. What I'm saying is that the existence of the article implies notability (just as it does for spaceship and oscillator articles) so it's probably better if the article is removed from the wiki.
- Creating 768P89 and 2116P89 and 902P89 articles seems to me similar to creating ... just to take a random example, it seems like there must be thousands ... articles for "71P12" and "40P12", on the grounds that Scorbie completed the p12 oscillator here, and then Sokwe reduced it:
- The fact that that discussion and reduction happened once upon a time doesn't warrant a LifeWiki article. Even if half a dozen people were involved in progressive reductions, it still wouldn't make the oscillator notable.
- One key consideration is that nobody is talking about "71P12" or "40P12", so nobody is ever likely to want to look up "71P12" on the LifeWiki and be redirected to "40P12". I would say that a Life object doesn't become notable just because multiple people did something with it once upon a time (and then dropped the subject and never brought it up again). Notability has more to do with an object being commonly discussed and/or used, such that there's a reasonable expectation that someone might know the name of it and want to look it up.
- In the case of this article, you've created not just one but three terms, "768P89" and "2116P89" and "902P89", that nobody has ever used before. Ultimately that looks to me like a dangerous step in the wrong direction, just like a "71P12" article redirecting to a "40P12" article would be.
- Regarding making up names: "36P8" (on the period-8 sparkers page) has entered common usage, even though it was made up well after discovery and even after (not during) the sparker compilation wiki page creation. However, "59P8" and some of the p7s on the p7 page are entirely unused even though they're there. I believe there's even an ambiguous one somewhere in there (121P7, maybe? I can't remember). I don't expect any of the three p89 names to ever be used.
- Well, yes -- in this case I think half a dozen people involved in progressive reductions did actually make this period-89 reaction notable. (Are there many known p89 partials out there? The category "Oscillators with period 89" did not exist at all before I made this page, even though p89 oscillators were already known -- doesn't non-existence of a category imply non-existence of oscillators?)
On the other hand, for the proposed example p12 oscillator, I would not create a dedicated page solely based on the given links -- but there might be some further discussion of the same oscillator that would give it notability (I don't know if there is any). Confocal (talk) 15:14, 5 April 2023 (UTC)
- Well, yes -- in this case I think half a dozen people involved in progressive reductions did actually make this period-89 reaction notable. (Are there many known p89 partials out there? The category "Oscillators with period 89" did not exist at all before I made this page, even though p89 oscillators were already known -- doesn't non-existence of a category imply non-existence of oscillators?)
- Okay, opinions can certainly differ. Personally I've always been skeptical about the proliferation of categories where categories aren't needed. In this case -- not surprisingly -- my opinion is that an "Oscillators with period 89" category isn't needed at the moment, because there aren't any notable oscillators of period 89.
- So... non-existence of a category implies non-existence of **notable** oscillators. I'm not sure if that answers your question in a useful way, though. 89 is well above the limit where glider/Herschel loops are possible, as Entity Valkyrie's completion shows. If we make a category page for that, but presumably we don't make a category page for (say) "Oscillators of period 7243", then where exactly is the dividing line?
- I don't see that we're in danger of cluttering up the LifeWiki if we push that dividing line as low as possible. But there does seem to be some serious danger of clutter, when these previously unknown "768P89" and "2116P89" and "902P89" terms are taken as precedents.
- There's an implied "251P89" where you just turn one of the p89 guns in Entity Valkyrie's completion into an oscillator. But that's not notable either! It would seem like a very dangerous idea to spend any time documenting "251P89" as a term, because *nobody has ever used that term* or spent much of any time discussing that p89 signal loop specifically. There's nothing particularly unique or interesting about it to discuss, so it doesn't need its own article and it's probably better to just leave it where it belongs, in the Catagolue gun and truegun collections.
- I think non-existence of a category implies non-existence of notable oscillators that are documented on dedicated pages -- there are undocumented notable oscillators (and there are undocumented/not well-documented notable topics in general). (Otherwise, "notable" would become more-or-less a synonym for "documented on a separate page on LifeWiki".) Confocal (talk) 17:30, 5 April 2023 (UTC)
For what it's worth, I'm much more worried about lack of good articles, rather than about having way too many of them. Writing an article does take an effort and I don't believe there will be too many of them. There are quite a few topics that definitely deserve dedicated pages (i.e. are notable topics) -- and yet people are unnecessarily worried about the notability of those topics, and refrain from creating pages, due to fear that the pages will get marked as non-notable and/or not knowing what "notable" even means. My opinion is that LifeWiki needs more pages about interesting things with something to write about them, not less. Confocal (talk) 15:14, 5 April 2023 (UTC)
- Notability is always going to be a difficult balancing act, but usually it turns out that it's possible to build some community consensus about whether a given article is notable or not. I understand your worry about people not creating articles because they're worried about notability. On the other hand, the current contributors to the LifeWiki have spent over a decade cleaning up after various tsunami waves of new articles about non-notable topics, by over-enthusiastic creators of articles who didn't really understand what notability meant.
- As a result of all this long-term vigilance and cleanup work, the LifeWiki contains a lot of good information and only a manageable amount of clutter. It's not perfect, of course, but it's actually a very impressive community-built resource.
- It's maybe a little unfortunate that "notability" is a difficult concept and that the resulting uncertainty prevents people from creating a few more good articles -- but at least to me, it's much more important that that same uncertainty also does a pretty darn good job of warding off a potential ocean of clutter-y articles on non-notable topics.
- There's another LifeWiki rule of thumb that comes into play here, for very similar reasons: in general, no one should be creating articles for Life patterns that they themselves created, or terms that they themselves created. That's another uncomfortable rule, but without it it becomes far too easy for new terms to proliferate that no one will ever actually use, like "768P89" and "2116P89" and "902P89". Dvgrn (talk) 16:32, 5 April 2023 (UTC)
- Regarding redirects and the current page name, I think these are more short systematic descriptions (Pentadecathlon IDs), rather than names. I agree that there might be a better way to name this page. Confocal (talk) 05:16, 6 April 2023 (UTC)
Where is the dividing line for oscillator periods?
I think at least one notable oscillator always exists for every known period, as long as the period itself is considered "notable" -- there is always the first known oscillator (which may or may not coincide with the smallest known oscillator of this period).
Not counting the two currently-unknown periods (19 and 41), several lowest periods currently missing from category "Oscillators with specific period" are 67, 69, 81, 95, 97, ... If the community decides to delete this page, then 89 will probably become missing again, making the number of two-digit known periods without categories equal to 6 (and the number of prime periods among them equal to 3).
Of course the dividing line will always be arbitrary -- one possibility for oscillator periods is "below 100". Confocal (talk) 16:54, 5 April 2023 (UTC)
- One possible dividing line is whatever the current minimum period is for universally adjustable signal loops that can include a signal splitter. Currently that's 61, for the speed tunnel. I'm not saying that's the "right" answer, but it seems reasonable. Oscillator engines below that period are significantly more difficult to find supports for, and so they end up being more interesting when they can be built anyway. (?) Dvgrn (talk) 20:06, 5 April 2023 (UTC)
- A problem with deriving notability criteria from current knowledge is that notability becomes unstable, and new discoveries may convert notable patterns into non-notable patterns. I think notability should be more permanent -- if a topic is considered notable today, and someone is brave enough to go ahead and document it today, then future discoveries should not invalidate the notability of this topic. (What if tomorrow someone discovers a Faster Tunnel, with repeat time 58 or 48 or 43?). Confocal (talk) 20:34, 5 April 2023 (UTC)
- If someone discovers a Faster Tunnel, then Life goes on.
- This kind of thing happens a lot, and it's perfectly okay. For the most part, notability **is** permanent -- in the sense that nobody is going around and putting notability tags on Garden of Eden 4 just because Garden of Eden 5 came along. On the other hand, nobody is creating articles called "knightlife's 10-block 14G splitter" just because pure glider generators were considered notable in the 1970s. And nobody creates an article for a newly discovered period-89 gun just because odd-period glider guns used to be incredibly rare. Old notability criteria don't make a newly created pattern interesting. The definition of "notable" *has* to change over time.
- To get back to the topic at hand here, I really like the style of the current 768P89 article -- it's a really good fit for the LifeWiki, with all kinds of information and links. But when I follow the links, I don't find anything that explains to me why "768P89" is notable. There's been no mention of this string of patterns since a brief flash of experimentation happened almost two years ago.
- We can't document every brief flash of experimentation, because there are way too many of them. They happen all the time, and mostly they don't lead anywhere. If we try to make the LifeWiki a repository for all information that someone might find interesting, then I think we inevitably create so much clutter that the LifeWiki becomes, on average, much less interesting and less useful.
- -- Let's see, the next possible candidate is just a few posts down in the same thread. There's a chain consisting of Sphenocorona, Entity Valkyrie, and me, experimenting with a "flip-flop reflector" / "resettable merge circuit" that's 'potentially' useful in some hypothetical circumstances, but hasn't actually gotten much of any use... so it's not notable. If an article were created for it, it would be likely to end up with a notability tag.
- Does that make (what I think is) the community consensus on notability any clearer, or no? Some differences of opinion like this are probably inevitable, but of course I and everybody would like to minimize the number of cases like this where a well-written article shows up, but then a notability tag has to be applied. Dvgrn (talk) 00:12, 6 April 2023 (UTC)
- Inclusion/notability criteria do change over time, but this is not an automatic process. I think there should be an observable community consensus behind a change of current inclusion criteria.
For example, I disagree with marking "obsoleted" once-record-holding patterns with notability template only because the record was later beaten by another pattern(s). There should be another reason to consider an once-notable pattern as non-notable and merge/delete the page. For example, a link to a previous discussion where the community did establish the current notability criteria for a class of patterns could be a valid reason. Confocal (talk) 05:00, 6 April 2023 (UTC)
- Inclusion/notability criteria do change over time, but this is not an automatic process. I think there should be an observable community consensus behind a change of current inclusion criteria.
- Speaking about the proposed example, I believe that "flip-flop reflector" does deserve documentation as well. I'm not going to document it myself right now, because I did not collect the necessary information. But certainly I would not object to someone else documenting it. Confocal (talk) 05:00, 6 April 2023 (UTC)
- Yeah, that resettable reflector is an interesting case. My opinion would be that it doesn't belong on the LifeWiki. It hasn't ever been picked up and used anywhere, so nobody is referring to it, so there's no established name for it, so nobody is going to be looking it up in the LifeWiki, so it doesn't need a LifeWiki article.
- It's possible that it hasn't been picked up only because nobody remembers it, and that if it had a LifeWiki article it would have gotten used by now. But that's unlikely, because A) the trigger system is awkward, crossing the lane that the new glider merges onto; B) this merge mechanism is expensive, because needs a separate glider from the opposite direction to reset it; C) the entire device is just a little bit too big to be useful. When people want this functionality they just use a syringe or speed tunnel plus a BNE14T30 or RNE-19T84 or
- What that means is that, if you went ahead and created an article called "flip-flop reflector" or "resettable merge circuit", you would be providing people with bad information. The existence of such an article would be misleading for two reasons.
- One, it would imply that "flip-flop reflector" or "resettable merge circuit" is a widely used and recognized term, which it isn't -- there are actually other candidate mechanisms that could be given those names.
- Two, it would imply that this mechanism is likely to be a good solution for some class of pattern construction problem, which it isn't (as far as we know) -- there are other much simpler mechanism that do the same thing, for any situation except for hypothetical highly specialized cases that nobody has run into yet.
- My conclusion would be that that particular merge mechanism is already documented just about as well as it should be -- on the forums -- and that adding a lot of words about it in a LifeWiki article would basically just decrease the signal-to-noise ration on the LifeWiki. If someone wanted to create a well-organized thread on the forums to collect all the hopeful specialized signal mechanisms that aren't ever used but might be ... then that thread might possibly merit a link added to the External Links section of Signal, or something like that -- though even the whole thread still wouldn't need a dedicated article, I wouldn't think. Dvgrn (talk) 13:15, 6 April 2023 (UTC)
- This might be obvious -- a wiki page does have a practical advantage over a forum thread, as a place to collect all known [X]. A forum thread where the first post is continually edited/updated relies on a single maintainer (the author of the first post), or possibly few maintainers (counting moderators). A wiki page can be maintained by multiple people, removing the critical dependency on a single maintainer (which is hard to change for a forum thread -- unless there's some technical way to transfer "authorship" of a forum post to different contributor).
In general, I do not think mere existence of a wiki page implies too much about usefulness. (Even though existence of a default-formatted encyclopedic-style LifeWiki article might have some implicit implications, there are still several other kinds of pages on LifeWiki that are not default-formatted encyclopedic articles). Confocal (talk) 15:17, 6 April 2023 (UTC)
- This might be obvious -- a wiki page does have a practical advantage over a forum thread, as a place to collect all known [X]. A forum thread where the first post is continually edited/updated relies on a single maintainer (the author of the first post), or possibly few maintainers (counting moderators). A wiki page can be maintained by multiple people, removing the critical dependency on a single maintainer (which is hard to change for a forum thread -- unless there's some technical way to transfer "authorship" of a forum post to different contributor).
- Sure! And this might also be obvious: conversely, forum threads have a big practical advantage over wiki pages, in that it's a whole lot easier to figure out how to contribute to a forum thread without breaking something. If you have a pattern in Golly, you can pretty much paste it into a forum message and click Submit (hopefully after adding code tags). With LifeWiki pages there are all kinds of interesting things that can happen with formatting, up to and including edit conflicts.
- We do have embedded LifeViewers with inline RLE now, which is way better than the old default method of creating a separate page in the RLE namespace. Now we can choose whether or not we want the RLE in question to end up as a named pattern in the downloadable LifeWiki pattern collection. But still, there's a nontrivial learning curve here, and relatively few people go to the trouble of getting past it.
- I suspect that that's a large part of the reason why, in practice, there are lots of "collection threads" on the forums, and not so many "collect all known X" articles on LifeWiki. Quite possibly the LifeWiki is just never going to be a very good place to do that kind of collecting.
- It's certainly absolutely fine to set up a "collect all known X" somewhere in the user namespace. That kind of thing does happen from time to time, and certainly nobody is going to be putting unfriendly non-notability tags on anything in the user namespace. The idea is that people can collaboratively work on an article, and eventually get it into a form where it can be moved over to the main namespace. That way there's never any incomplete or "temporarily wrong" info showing up in the main namespace.
- I certainly agree that it's easier to figure out how to contribute by posting on the forum, and it's harder to learn wiki markup and conventions. (I think it might be good to have a forum-based equivalent of LifeWiki:Tiki bar on the forum -- then people could contribute, even without a LifeWiki account/without knowledge of wiki markup.) Confocal (talk) 21:46, 6 April 2023 (UTC)
- Regarding user-namespace pages, I think there should be some visible indication that the author is fine with other people editing the page. Personally, I do not want to rely on implicit unstated assumptions for userspace pages. Sometimes it is reasonably clear that an userspace page is indeed meant to be edited by everyone -- e.g. when it says that explicitly, or when the talk page says that, or when the edit history shows how multiple people did in fact already contribute to the page from the beginning.
The page User:Confocal/R is partially autogenerated from time to time, and is not really formatted in a form suitable for editing by multiple people. I do not expect contributions to that page from other people. Confocal (talk) 21:46, 6 April 2023 (UTC)
- Regarding user-namespace pages, I think there should be some visible indication that the author is fine with other people editing the page. Personally, I do not want to rely on implicit unstated assumptions for userspace pages. Sometimes it is reasonably clear that an userspace page is indeed meant to be edited by everyone -- e.g. when it says that explicitly, or when the talk page says that, or when the edit history shows how multiple people did in fact already contribute to the page from the beginning.
One of assumptions in this discussion is the assumption "important to discourage the creation of more articles like this one". I question this assumption. I would rather (attempt to) encourage people to go ahead and document things (at least, as long as that documentation work by itself does not cause actual troubles)
Only because I did in fact participate in this discussion up to this point and I did attempt to explain my thoughts, does not automatically mean that every other Lifenthusiast/CA enthusiast/wiki editor will behave in the same way. In reality, I expect that most people will just quickly lose enthusiasm and refrain from further editing.
I believe there are very many notable topics that are not currently documented or are inadequately documented. There is no imaginable way that all of those topics will ever be actually adequately documented (even though every such topic could potentially be adequately documented). I do not accept the "potentially" (as in "potentially would cause clutter") as a valid reason to discourage people from documentation.
I believe that encouraging people to do the documentation could create a much more productive/friendly atmosphere. Tagging new articles as "non-notable" (or worse, redirecting/merging them without any discussion) is unfriendly. Confocal (talk) 05:05, 6 April 2023 (UTC)
- I like your attitude here, and it's fine to question assumptions. On the other hand, when all is said and done, I'm still happy with the current set of rules that the LifeWiki operates under. They don't solve all possible problems, by any means, but they've been tested for over a decade and found to be a good balance between discouraging clutter and encouraging contributions.
- The current small group of moderators seems to generally be able to handle the few rare cases of over-enthusiastic editing, as those cases come up -- and that's a good thing, and shouldn't be taken for granted. The LifeWiki is currently in surprisingly good shape, all in all, and it seems important to keep it that way... such that we should make a point of weighing the possible risks of any given policy change against its possible rewards.
- To recap: this article still has a notability tag added because
- the "768P89" pattern being described doesn't get talked about much, or even at all, mostly because the p89 engine needs glider guns to support it so it doesn't rise to the level of "notable" that's generally agreed on these days by ... you know ... modern oscillator connoisseurs;
- the blow-by-blow style of narration for progressive pattern improvements in this article is only ever used for notable (e.g., record-breaking) patterns, where attribution is important -- so a reader will likely be misled into thinking that this "768P89" is a notable pattern for some reason (and will then be puzzled as to why it's notable);
- Pentadecathlon IDs are pretty much never applied to patterns that contain guns, because Pentadecathlon IDs highlight the population -- and population is clearly important for oscillator SKOPs, but it is not really even considered in glider gun record-keeping;
- you invented the term "768P89" (and "2116P89" and "902P89") yourself. If you've invented a name and nobody else has ever used it, that's a good warning sign that you shouldn't be creating an article with that name.
- People really do try to take that last item seriously. On average, "don't document your own stuff" produces a reliable warning that you're on shaky ground, and it avoids a lot of clutter on the LifeWiki -- not just potential clutter, actual clutter. We have a lot of accumulated collective experience on this point.
- Again, yes, I totally agree, this rule is annoying. I once waited three years for someone to write an article about the orthogonal loopship (because I couldn't write it, because I had invented both the term and the pattern). Eventually another loopship came along, and the outcome of some discussion on Discord was that I'd better write the article since nobody else understood the thing well enough.
- So there are potential exceptions to these rules, for sure! But it's best if exceptions are actually exceptional cases, and to decide that, some perspective from somebody else is going to be needed. In my view, the occasional exception helps to prove the general validity of the rules. Dvgrn (talk) 22:01, 7 April 2023 (UTC)
- I think at least some of these items can be resolved relatively easily, which I'll probably attempt. Even though I'm not sure if a Pentadecathlon ID really counts as a name (I think it is more a partial description for unnamed patterns), I agree that it highlights population. And in this case the population can be changed by changing sources of glider streams. My motivation for listing these descriptions was that they correspond to the patterns that were actually constructed and posted. Confocal (talk) 22:59, 7 April 2023 (UTC)
- I moved the page to a new pagename, and added several forum links that could potentially be used to expand the page to show other known p89 reactions. I still believe that this particular p89 reaction is notable, even though it is supported by glider streams. If there were many such reactions that were actually explored and discussed by several people, with actual patterns posted and further improved on the forum, that would probably convince me that I'm wrong. Confocal (talk) 00:46, 8 April 2023 (UTC)
- This was kind of an odd case, where a bunch of us spent a little time playing around with something that we all knew wasn't particularly notable. It doesn't happen all that often, in good part because (at least by recent tradition) oscillator engine completions using glider guns are not very interesting. I was the one who got this particular series started, just because I didn't know a good way to supply a high-clearance domino spark with just one input glider stream.
- So the only thing that makes the final pattern maybe worth keeping in a LifeWiki article, would be the fact that it showcases a good edgy Herschel-to-domino-spark converter. Maybe a link from domino would be a good idea.
- If it's okay, I'll go ahead and trim down the current article to get rid of the misleading stuff. Seems like it's fine to mention "David Raucci, Dave Greene, Matthias Merzenich, Sphenocorona, May13, and Entity Valkyrie 2" in the caption of the embedded LifeViewer, but I really only want to keep the final link. Five extra links, giving population and bounding box details for each stage, just makes this random p89 engine completion look it ought to be a big record-breaker or something, and it's just not anything special.
- Re: convincing you about this particular case -- it would be easy to use the same glider gun plus H-to-domino converter mechanism in some cases, or single-glider or multi-gun multi-glider collisions in other cases, to complete some of the other p89 engines that you have in the earlier links. E.g., here's a completion of the last engine that I just made up:
| (click above to open LifeViewer) |
- Similar things could be assembled for a large fraction of hotdogPi's incomplete partials, for every period above 13. (For some periods between 14 and 73, you'd have to use pseudo-period guns, which would **really** be missing the point of the exercise). Usually nobody takes the time to build and post these things, just because they're known to be doable but it's too easy and too big-and-ugly -- like bringing a cannon to a shot put competition. Dvgrn (talk) 18:56, 8 April 2023 (UTC)