ConwayLife.com

The page 754P7 keeps crashing on my browser because it contains a (nonstandard) LifeViewer-implemented gallery.

LaundryPizza03 wrote: December 2nd, 2021, 6:16 pm The page 754P7 keeps crashing on my browser because it contains a (nonstandard) LifeViewer-implemented gallery.

Nothing looks particularly non-standard except for the theme. I've gone ahead and fixed that. I'm not seeing any browser crashes, but I suspect that you'll still see the same crashes as before.

(?)

dvgrn wrote: December 2nd, 2021, 7:46 pm Nothing looks particularly non-standard except for the theme. I've gone ahead and fixed that. I'm not seeing any browser crashes, but I suspect that you'll still see the same crashes as before.

(?)

I'm pretty sure he's referring to the fact that the viewers on this page used LV:Viewer directly rather than the more typical Template:EmbedViewer, something which I've also gone and fixed. Not sure why this would result in crashing, though, because the latter contains the former.

Can anyone check my suggestion in the last section of LifeWiki_talk:Did_you_know?

Tangential note, maybe this old thread on pi+block/blinker collisions can provide material for a couple of DYKs.

The page for www.conwaylife.com/wiki/Honey_farm_hassler, the p9 HFH's top eater is missing a cell. Anyone else notice it?

Wyirm wrote: December 9th, 2021, 1:41 pm The page for www.conwaylife.com/wiki/Honey_farm_hassler, the p9 HFH's top eater is missing a cell. Anyone else notice it?

got fixed by user Ihatecorderships

Book wrote: December 9th, 2021, 2:13 pm

Wyirm wrote: December 9th, 2021, 1:41 pm The page for www.conwaylife.com/wiki/Honey_farm_hassler, the p9 HFH's top eater is missing a cell. Anyone else notice it?

got fixed by user Ihatecorderships

Yep, thanks for pointing it out!
Also, you can do it yourself if you have a wiki account.

Should there be a wiki article about lifecompetes?

Is it at all possible to exclude certain patterns from certain categories, such as the blockade page from "Periodic objects with minimum population 16" as it isn't a strict still life, or bi-block from the respective 8 category as it isn't either?

muzik wrote: December 14th, 2021, 2:41 pm Is it at all possible to exclude certain patterns from certain categories, such as the blockade page from "Periodic objects with minimum population 16" as it isn't a strict still life, or bi-block from the respective 8 category as it isn't either?

Should be done now. I did have to mess around a bit with ifeq statements, so let me know if you notice any issues.

the pages F/I, F/L, L/I, and L/MCPS should be merged into the "Measuring methuselahs" section of the Methuselah page.

Are these acorn siblings not mentioned because they are larger in terms of the minimal covering polyplet size, or should I mention them on acorn's page?

Is it possible to make pattern templates natively accept upper and/or lower bounds for discovery years, rather than only absolutes?

muzik wrote: December 23rd, 2021, 7:54 pm Is it possible to make pattern templates natively accept upper and/or lower bounds for discovery years, rather than only absolutes?

Sure, there's no code anywhere that requires that the contents of the #O line contain just one date with a day, month, and four-digit year. You can say whatever you need to say in the #O line -- "between 2005 and 2009" won't cause any trouble. It's just that usually we can figure out a more exact date than that.

EDIT: Oops, sorry -- for some reason I thought this was in a different thread, part of the discussion about additions to Golly's pattern collection.

dvgrn wrote: December 23rd, 2021, 9:06 pm

muzik wrote: December 23rd, 2021, 7:54 pm Is it possible to make pattern templates natively accept upper and/or lower bounds for discovery years, rather than only absolutes?

Sure, there's no code anywhere that requires that the contents of the #O line contain just one date with a day, month, and four-digit year. You can say whatever you need to say in the #O line -- "between 2005 and 2009" won't cause any trouble. It's just that usually we can figure out a more exact date than that.

I don't think pattern infoboxes are capable of reading the #O section of an RLE, and even if they were, it would end up being redundant since the "discoverer" and "discoveryear" parameters are filled out manually in pretty much every case I've seen so far.

The Omniperiodic page wrote:Dean Hickerson proved that a variety of rules were omniperiodic, by running his drifter searcher to find signals that can be manipulated like Herschels.

Can someone provide a reference for the statement? It needs some examples, at least linked to if not direcly included in the page, but what I can see for now is a related forums post by Sokwe.

GUYTU6J wrote: December 26th, 2021, 11:54 am Can someone provide a reference for the statement? It needs some examples, at least linked to if not direcly included in the page...

I guess Dean's signal-injectors page is all vanilla-Life-only stuff. Maybe the best reference will be to this post: here's the only example I found on a quick survey of Dean showing a rule to be omniperiodic using the 'drifter' search program.

On 13 Nov 1997, Dean Hickerson wrote: Subject: Re: B3/S0123


Mostly to Allan [Wechsler]:

This rule is omniperiodic. The following is a drifter loop of length 104,
which can be increased by multiples of 20:

Code: Select all

#C p104 oscillator, in which a drifter moves around a square
x = 55, y = 55, rule = B3/S0123
30b2o$28bo$28bob2o$24bobobo$24bobo3b2o$22bobob2o$24bo5b2o2b2o$22bo3b2o
$21bo7bob5o$21bobo2b2obobo$17bobobobo5bobo2b2o$17bobobob5obo$15bobo11b
o4b4o$15bobo4b4o3b2obo5bobo$13bobob3o10bobob3o$15bo5bobo4bo5bo4b4o$13b
o3b4o7bobobobo$12bo16bo4bobobob5o$12bobo2b2o15bobobobo$6bobobobobo23bo
bo2b2o$6bobobobob5o19bo$8bo30bo3b4o$8bo4b2obo24bo5bobo$obobobob3o28bob
ob3o$obobobo6b2obo24bobo4b4o$8b6o3bo23bobo$b3o11b2o26bobobob3o$5bobobo
bo31bobobobo$3b3obobobo26b2o11b3o$11bobo23bo3b6o$3b4o4bobo24bob2o6bobo
bobo$9b3obobo28b3obobobobo$5bobo5bo24bob2o4bo$8b4o3bo30bo$16bo19b5obob
obobo$10b2o2bobo23bobobobobo$16bobobo15b2o2bobo$10b5obobobo4bo16bo$20b
obobobo7b4o3bo$12b4o4bo5bo4bobo5bo$18b3obobo10b3obobo$14bobo5bob2o3b4o
4bobo$17b4o4bo11bobo$25bob5obobobo$19b2o2bobo5bobobobo$23bobob2o2bobo$
19b5obo7bo$27b2o3bo$19b2o2b2o5bo$27b2obobo$23b2o3bobo$26bobobo$23b2obo
$26bo$23b2o!

Successive drifters can be as close as 7 gens apart, so we can get all
nonmultiples of 5 from 7 on. To get multiples of 5 we need to use more
than 4 corners in the loop. Fortunately, the diagonal track is glide-
symmetric, so successive turns can go in opposite directions as well as
in the same direction. A symmetric loop with 20 corners works; I haven't
checked to see if an asymmetric 10 would do the job. For example:

Code: Select all

#C p25 oscillator, in which 12 drifters moves around a path of length
#C 300 gens with 20 corners
x = 128, y = 64, rule = B3/S0123
21b2o14b2o14b2o14b2o14b2o$19bo15bo15bo15bo15bo$19bob2o12bob2o12bob2o
12bob2o12bob2o$15bobobo11bobobo11bobobo11bobobo11bobobo$15bobo3b2o4bo
3bobo3b2o4bo3bobo3b2o4bo3bobo3b2o4bo3bobo3b2o$13bobob2o10bobob2o10bobo
bo11bobob2o10bobob2o$15bo5b4o3b2obo5b4o3b2obo3b6o3b2obo3bob4o3b2obo5b
2o2b2o$13bo3b2o9bo3b3o9bo4b4o7bo4b4o7bo4b2o$12bo7bob5obo3bo3bob5obo7bo
b5obo7bob5obo7bob5o$12bobo2b2obobo5bobo2b2obobo5bobo2b2obobo5bobo2b2ob
obo5bobo2b2obo$6bobobobobo5bobo2b2obobo5bobo2b2obobo5bobo2b2obobo5bobo
2b2obobo5bobo2b2o$6bobobob7obo7bob5obo7bob5obo7bob5obo7bob5obo$8bo11bo
4b2o9bo4b2o9bo4b2o9bo4b2o9bo4b4o$8bo4b2obo3b2obo5b4o3b2obo5b4o3b2obo5b
4o3b2obo5b4o3b2obo5bobo$obobobob3o10bobob2o10bobob2o10bobob2o10bobob2o
10bobob3o$obobobo6b2obo2bo3bobo3b2o4bo3bobo3b2o4bo3bobo3b2o4bo3bobo3b
2o4bo5bo4b4o$8b6o3bo5bobobo11bobobo11bobobo11bobobo7bobobobo$b3o11b2o
10bob2o12bob2o12bob2o12bob2o5bo4bobobob5o$5bobobobo15bo15bo15bo15bo13b
obobobo$3b3obobobo17b2o14b2o14b2o14b2o14bobo2b2o$11bobo79bo$3b4o4bobo
80bo3b4o$9b3obobo80bo5bobo$5bobo5bo80bobob3o$8b4o3bo80bobo4b4o$16bo79b
obo$10b2o2bobo81bobobob5o$14bobobobo77bobobobo$10b5obobobo81bobo2b2o$
20bobo79bo$12b4o4bobo80bo3b4o$18bobobobo80bo5bobo$14bobo5bo80bobobobo$
17b4o3bo80bobo4b4o$25bo79bobo$19b2o2bobo81bobobob5o$23bobobobo77bobobo
bo$19b5obobobo81bobo2b2o$29bobo79bo$21b4o4bobo80bo3b4o$27b3obobo80bo5b
obo$23bobo5bo80bobob3o$26b4o3bo80bobo4b4o$34bo79bobo$28b2o2bobo14b2o
14b2o14b2o14b2o17bobobob3o$32bobobobo13bo15bo15bo15bo15bobobobo$28b5ob
obobo4bo5b2obo12b2obo12b2obo12b2obo10b2o11b3o$38bobobobo7bobobo11bobob
o11bobobo11bobobo5bo3b6o$30b4o4bo5bo4b2o3bobo3bo4b2o3bobo3bo4b2o3bobo
3bo4b2o3bobo3bo2bob2o6bobobobo$36b3obobo10b2obobo10b2obobo10b2obobo10b
2obobo10b3obobobobo$32bobo5bob2o3b4o5bob2o3b4o5bob2o3b4o5bob2o3b4o5bob
2o3bob2o4bo$35b4o4bo9b2o4bo9b2o4bo9b2o4bo9b2o4bo11bo$43bob5obo7bob5obo
7bob5obo7bob5obo7bob7obobobo$37b2o2bobo5bobob2o2bobo5bobob2o2bobo5bobo
b2o2bobo5bobob2o2bobo5bobobobobo$43bob2o2bobo5bobob2o2bobo5bobob2o2bob
o5bobob2o2bobo5bobob2o2bobo$37b5obo7bob5obo7bob5obo7bob5obo3bo3bob5obo
7bo$45b2o4bo7b4o4bo7b4o4bo9b3o3bo9b2o3bo$37b2o2b2o5bob2o3b4obo3bob2o3b
6o3bob2o3b4o5bob2o3b4o5bo$45b2obobo10b2obobo11bobobo10b2obobo10b2obobo
$41b2o3bobo3bo4b2o3bobo3bo4b2o3bobo3bo4b2o3bobo3bo4b2o3bobo$44bobobo
11bobobo11bobobo11bobobo11bobobo$41b2obo12b2obo12b2obo12b2obo12b2obo$
44bo15bo15bo15bo15bo$41b2o14b2o14b2o14b2o14b2o!

Finally, we need examples of oscillators with periods <7. Mark gave
periods 2, 3, and 5; here are 4 and 6:

Code: Select all

x = 21, y = 10, rule = B3/S0123
14bo$16bobo$7bo6bobobo$bobobo8bo$bobob2o5bobo5bo$2o10bo5bo$3bobobo6bob
obobo$2o3bobo4bobo3bo$2b3o9bo3bo$2bo10bo!
#C [[ GPS 2 AUTOSTART THUMBNAIL THUMBSIZE 3 LABEL 3 11 10 "p4" LABEL 16 11 10 "p6" ]]

(Obviously the format of the patterns has been modernized a little, but the text is otherwise unchanged.)

Regarding muzik's changes, here's what I agree and disagree with:

Agree:
* Hook → bookend when it refers to a bookend (actual hooks stay the way they are)
* R-bee, R-loaf, and R-wing → bun, wing, and dove. Particularly R-bee; R-bee should refer to generation 2 of the R-pentomino that's also a beehive plus one cell, but it refers to something else entirely.
* Marking for deletion of speeds such as 2c/21 orthogonal. I saved two by redirecting them to the HighLife (c/98) and Pedestrian Life ((5,2)c/190) pages, I listed an alternative for deletion for 2c/7 diagonal to whichever page explains that some rules allow faster diagonal ships (2c/7 is too fast for Life), and I feel like 2c/9o and 2c/11o could go to Sokwe's search status page except that it's in userspace, but I didn't edit those two pages.
* Krake badly needed a page. It has one now.
* Marking 20P2 through 26P2 for moving to another name, since they don't have 20 through 26 cells.

Disagree:
* Adding all the combinations of how snakes, buns, wings, doves, houses, and a few other things can act as induction coils. Many of them are nowhere near common enough, especially the bridge ones. Maybe 5 of them are worth it for being in the list of common still lifes (which goes up to about 200).
* Renaming dead spark coil to mirrored house.
* Renaming likely more than one third of the 12-cell still lifes and a decent number of those with 10 and 11. And with redirects suppressed.
* Placing unknown frequency templates and boilerplate glider synthesis templates on pretty much every still life and small enough oscillator. For the boilerplate glider synthesis template, this even includes the block.
* Making nine, block and glider, and wing three separate pages again.

Another complaint towards muzik's recent moves...

dvgrn wrote: December 23rd, 2021, 6:49 pm

muzik wrote: December 23rd, 2021, 5:06 pm Do we have a list of when still lifes of each population count were enumerated and by who?

How about https://conwaylife.com/wiki/Still_life# ... till_lifes ? That seems fairly comprehensive, though dates aren't given for the first few people on the list. Conway's enumeration was 1969-70; Bob Wainwright's, David Buckingham's and Peter Raynham's work was done by 1975 or so; and Mark Niemiec said in 1994 that he had enumerated up to 18 bits "a couple of years ago" (around 1992).

Mark posted the counts up to 21 bits on 28 February 1997, and the counts up to 23 bits on 21 October 1997. The count for 24 bits doesn't seem to show up until 29 March 2004, but was probably calculated years before that -- Mark, and/or the Internet Archive, might remember more.

You see, now these years as well as a link to this post end up showing in the infoboxes of some small still lives. I'd say that is somewhat misleading; for example, muzik have added several 8-cell still lives to the list of patterns created in 1975, despite them having shown up on Lifeline_Volume_3#Page_2 which was published in September 1971.
Honestly, I don't like adding discovery info for the sake of "consistence" in having discovery info as other still lives do in this manner. It does not improve quality.

Could pattern infoboxes have an option added to disable linking to (likely nonexistent) pattern images, possibly adding such pages to a hidden tracking category? I'd prefer this to having an open wound in every page that doesn't yet have an image uploaded.

muzik wrote: January 3rd, 2022, 2:02 pm Could pattern infoboxes have an option added to disable linking to (likely nonexistent) pattern images, possibly adding such pages to a hidden tracking category? I'd prefer this to having an open wound in every page that doesn't yet have an image uploaded.

it is now quick and easy to add static images using RLE2PNG http://localhost/philbookmandotcom/tools/rle2png.htm.

Book wrote: January 3rd, 2022, 2:42 pm

muzik wrote: January 3rd, 2022, 2:02 pm Could pattern infoboxes have an option added to disable linking to (likely nonexistent) pattern images, possibly adding such pages to a hidden tracking category? I'd prefer this to having an open wound in every page that doesn't yet have an image uploaded.

it is now quick and easy to add static images using RLE2PNG http://localhost/philbookmandotcom/tools/rle2png.htm.

What are the parameters for creating static pattern images with your method? Per Help:Images, the LifeWiki convention is to have 15×15 pixels for a single cell and 1 pixel for a borderline in color (192, 192, 192) (#C0C0C0).

I still don't know why we're using PNGs and not SVGs. This type of image seems perfect for SVGs.

Would it be possible to add support for kinetic symmetry to still life, oscillator and spaceship infoboxes? Any pages that don't have it defined should be added to a dedicated tracking category so that they can all be added later.

This should help to replace the microscopic, silent templates added to the bottom of each page, which only seem to categorise the page and don't show the symmetry clearly at all.

What about a Tutorials:Corderizing?