Catagolue Oddities

[muzik]

"Uploads" is an unofficial symmetry, as opposed to a troll symmetry:

https://catagolue.appspot.com/census/b3s23

[muzik]

I don't suspect this spaceship is rendering properly https://catagolue.appspot.com/object/xq ... /g143b3s23

[calcyman]

I don't suspect this spaceship is rendering properly https://catagolue.appspot.com/object/xq ... /g143b3s23

Thanks -- that's because I was using 1 byte instead of 2 per cell. Now I've amended that and given a more aesthetically pleasing colour scheme compatible with arbitrary rules:

https://catagolue.appspot.com/haul/x3xb ... 0af65fb79c

[muzik]

The contrasts between the different cell states in Generations still feels too hard to discern in my opinion, even at lower ranges like 17.

Also, why aren't LifeViewer previews given for such Generations patterns, anyway (and indeed, why are they shown on some custom ruletable objects such as this?

[calcyman]

Completely unrelated oddity: Saka and Anonymous are uploading hauls to b3s23/Saka_Test in order to exchange pleasantries embedded in apgcodes, and it's causing the census to distractingly rise up the left-hand column of the main census page.

This isn't quite what I intended when I wrote the following:
http://www.conwaylife.com/wiki/Catagolu ... _inventory

[77topaz]

Maybe just tag it as a troll symmetry so it doesn't show up in the main page list? It sort of was that already from the beginning.

[muzik]

Completely unrelated oddity: Saka and Anonymous are uploading hauls to b3s23/Saka_Test in order to exchange pleasantries embedded in apgcodes, and it's causing the census to distractingly rise up the left-hand column of the main census page.

This does raise a question: should unofficial symmetries even be included in the left-hand column in the first place? Since almost all unofficial symmetries are test symmetries of some sort, it does seem weird for them to be there, since they're technically not "true" censuses.

[77topaz]

This does raise a question: should unofficial symmetries even be included in the left-hand column in the first place? Since almost all unofficial symmetries are test symmetries of some sort, it does seem weird for them to be there, since they're technically not "true" censuses.

Personally, I don't have a problem with benevolent unofficial symmetries like EE9's gutter tests being on that list - the issue is that Saka_Test is becoming more of a troll symmetry than a benevolent unofficial symmetry like the aforementioned examples.

[muzik]

Some old "still lifes" here:
https://catagolue.appspot.com/census/g3b2s/C1

[mniemiec]

While looking for natural predecessors for as-yet-unsynthesized oscillators, I found this tantalyzing tidbit: Almosymmetric
https://catagolue.appspot.com/object/xp ... 0c02/b3s23
comes up in one soup; unfortunately, the soup (from symmetry Saka_test) doesn't actually produce it. Is there a way to manually de-convolute the Catagolue soup to recover the original soup that produces it? I've been able to do so for some of the rotational symmetries.
EDIT:
Also 22P4.1: https://catagolue.appspot.com/object/xp ... 1023/b3s23
Also 24P4.1: https://catagolue.appspot.com/object/xp ... 5011/b3s23

It appears that all of these are exactly the same soup. I'm curious how troll symmetries can pass verification and upload fraudulent data to Catagolue. Clicking on the symmetry opens up a page with a "Sakagolue" logo rather than the original Catagolue logo. From there, the "xs" and "xs<textarea>" category areas include objects that injects HTML. These demonstrate a vector for possible malware, much more dangerous than merely polluting Catagolue census with bogus data.

[Ian07]

While looking for natural predecessors for as-yet-unsynthesized oscillators, I found this tantalyzing tidbit: Almosymmetric
https://catagolue.appspot.com/object/xp ... 0c02/b3s23
comes up in one soup; unfortunately, the soup (from symmetry Saka_Test) doesn't actually produce it. Is there a way to manually de-convolute the Catagolue soup to recover the original soup that produces it? I've been able to do so for some of the rotational symmetries.

Unfortunately, that's actually due to an anonymous user trolling Catagolue by submitting hauls with fake objects. Here's an example: https://catagolue.appspot.com/haul/b3s2 ... 4be2e24502

EDIT: Agree with muzik's post below.

[muzik]

I personally believe that troll symmetries should be excluded from object pages entirely;

- they are submitted with the intent to either test or troll Catagolue and are seldom used for actual soup searching,
- they are not officially supported by Catagolue and will not give functional soups unless C1,
- and they can lead to confusion like in this case.

[calcyman]

I personally believe that troll symmetries should be excluded from object pages entirely;

- they are submitted with the intent to either test or troll Catagolue and are seldom used for actual soup searching,
- they are not officially supported by Catagolue and will not give functional soups unless C1,
- and they can lead to confusion like in this case.

That's a good suggestion; I've now implemented it.

[muzik]

On the topic of object pages, the rule the object runs in is reported as a generic non-standard rule even if said rule has a name recorded in Catagolue:

https://catagolue.appspot.com/object/xp ... x11/b36s23

https://catagolue.appspot.com/census/b36s23

For b3s23, the rule is reported as "standard life" instead of "Conway's Game of Life":

https://catagolue.appspot.com/object/xp ... zx11/b3s23

https://catagolue.appspot.com/census/b3s23

[77topaz]

It appears that all of these are exactly the same soup. I'm curious how troll symmetries can pass verification and upload fraudulent data to Catagolue. Clicking on the symmetry opens up a page with a "Sakagolue" logo rather than the original Catagolue logo. From there, the "xs" and "xs<textarea>" category areas include objects that injects HTML. These demonstrate a vector for possible malware, much more dangerous than merely polluting Catagolue census with bogus data.

• Censuses that have less than one trillion objects or are marked as "unofficial" or "troll" don't have a verification process in the first place.
• The "Sakagolue" logo was placed there deliberately by Calcyman as a form of humour.
• Calcyman is aware of the HTML injection, and according to Saka (the original person uploading to the "Saka_Test" has already restricted what can be done with it.

[muzik]

There exist some censuses that don't fit into any rule family (evidenced by the lack of a navigation bar), and as such remain unlisted; can a separate rules category be made for these?

https://catagolue.appspot.com/census/tlife/C1

[77topaz]

That's just a duplicate of an isotropic non-totalistic rule (B3/S2-i34q), though.

[muzik]

That's just a duplicate of an isotropic non-totalistic rule (B3/S2-i34q), though.

It's still a separate census; it'd be helpful to have a page list out all such unfitting censuses so they cans be viewed (and potentially cleaned up somehow, as well).

[muzik]

https://catagolue.appspot.com/census/g1000b3s23/C1

[77topaz]

Yeah, I think I've seen censuses with nonzero hauls but zero objects look like that before. But those should be such rare cases that it may not be really necessary to fix this bug.

EDIT: I've now apgsearched the rule, so it won't have 0 objects anymore. Surprisingly, there actually appears to be a natural spaceship that has occurred 5 times in 10000 soups, but I can't view it because it's marked as ov_q4 and Golly doesn't support Generations with 1000 states.

EDIT 2: LifeViewer can't do it either (here's one of the spaceship's sample soups):

Code: Select all

x = 16, y = 16, rule = S23/B3/1000
obbbbobbbobobobo$
bbbbbbobooobobbb$
bobbobobboboboob$
oobooboboboboooo$
bobbbbboboooobbo$
obbbboobbbbobboo$
oboobbobbooobbob$
bbboooobbobobboo$
bobbbooobbobobob$
bbbooobobooooobb$
bobboobobooooboo$
oboboobbbooobooo$
bboobboboboooobb$
obooobooobbobbbb$
bboobbbobobbbbbo$
obobbbbbbbbbobob!

But by lowering the states to 100, I think I know what the spaceship must look like:

Code: Select all

x = 16, y = 16, rule = S23/B3/100
obbbbobbbobobobo$
bbbbbbobooobobbb$
bobbobobboboboob$
oobooboboboboooo$
bobbbbboboooobbo$
obbbboobbbbobboo$
oboobbobbooobbob$
bbboooobbobobboo$
bobbbooobbobobob$
bbbooobobooooobb$
bobboobobooooboo$
oboboobbbooobooo$
bboobboboboooobb$
obooobooobbobbbb$
bboobbbobobbbbbo$
obobbbbbbbbbobob!

[muzik]

Deficient rules don't seem to show up here:

https://catagolue.appspot.com/rules/deficient

[calcyman]

Deficient rules don't seem to show up here:

https://catagolue.appspot.com/rules/deficient

Those /rules pages are only refreshed every 6 hours, because it involves running a query against the complete list of censuses.

[muzik]

I forgot about that.

There's something a bit more concerning, though: catagolue seems to want to animate deficient rules, and declares almost all deficient rule object pages as invalid: https://catagolue.appspot.com/census/b3s23d/C1

[Hunting]

Invalid pattern... What the heck?
http://catagolue.appspot.com/object/xp6 ... -ak3ais12a

[77topaz]

Invalid pattern... What the heck?
http://catagolue.appspot.com/object/xp6 ... -ak3ais12a

The page is loading normally for me, I don't see any "invalid pattern" errors.