The above description of the answer is probably not very helpful, so here are some explained examples. In one left example, out of the groups shown, all but one are "all but one are ___". In the corresponding right example, all are "all but one are ___" and all but one of these groups are black.

Other left examples include "every other is 'every other is ___' " and "gradually getting more like 'gradually getting more like ___' ".

Roughly speaking, right examples are meta while left examples are doubly-meta: "meta" is when the whole group imitates its parts; "doubly-meta" is when the whole group imitates its parts with respect to the way it imitates its parts.

This Problem is a "correspondence" (left-BP919) Problem. Examples in this Problem should ideally admit both a doubly-meta version and a singly-meta version. It may be very hard to interpret some examples if this Problem isn't presented helpfully (in a symmetrical format with corresponding squares opposite one another).

All left and right examples show five groups of five objects. (These numbers are unimportant; feel free to change this.) The objects in each group are aligned horizontally and the groups are stacked on top of one another vertically.

Each example showcases some property that depends on other properties: "X with respect to __[property]__".

Currently, in none of the uploaded left examples do all the groups shown follow the same pattern, so a valid alternative solution is "not all groups share some obvious property vs. all do." This is because the relevant group-property usually has to do with not all objects in it satisfying a certain property, and thus not all groups can satisfy that group-property. Some group-properties would allow the whole group to be the same, but these tend to be hard to parse. Indeed, shown sorted ambiguously above are some examples that would fit left interpreted as follows: EX8220 "all are 'all are ___' " and EX8222 "palindrome with respect to being a palindrome with respect to ___", where every example is a palindrome with respect to some property. It may confuse the message to include any examples like these on the left side.

Clarification: EX8231, "palindromic with respect to the property the group is palindromic in" does not quite fit left in this Problem, although it does seem doubly-meta in some sense. Still, it is not palindromic with respect to the property of being palindromic. Instead of having the form "X 'X __' ", this is more like "X [the __appearing in 'X__'] ". Analogously, consider "gradual transition with respect to what the gradual transition is", "odd one out with respect to what property is the odd one out", etc. This idea could make for another Bongard Problem.

Below is a list of left example ideas that would be impossible to make. Please add and sign any more ideas you have.

"Exhaustive list of all exhaustive lists of all ____". - Aaron David Fairbanks, Nov 23 2020

The above description of the answer is probably not very helpful, so here are some worked examples. In one left example, out of the groups shown, all but one are "all but one are ___". In the corresponding right example, all are "all but one are ___" and all but one of these groups are black.

Other left examples include "every other is 'every other is ___' " and "gradually getting more like 'gradually getting more like ___' ".

Roughly speaking, right examples are meta while left examples are doubly-meta: "meta" is when the whole group imitates its parts; "doubly-meta" is when the whole group imitates its parts with respect to the way it imitates its parts.

This Problem is a "correspondence" (left-BP919) Problem. Examples in this Problem should ideally admit both a doubly-meta version and a singly-meta version. It may be very hard to interpret some examples if this Problem isn't presented helpfully (in a symmetrical format with corresponding squares opposite one another).

All left and right examples show five groups of five objects. (These numbers are unimportant; feel free to change this.) The objects in each group are aligned horizontally and the groups are stacked on top of one another vertically.

Each example showcases some property that depends on other properties: "X with respect to __[property]__".

Currently, in none of the uploaded left examples do all the groups shown follow the same pattern, so a valid alternative solution is "not all groups share some obvious property vs. all do." This is because the relevant group-property usually has to do with not all objects in it satisfying a certain property, and thus not all groups can satisfy that group-property. Some group-properties would allow the whole group to be the same, but these tend to be hard to parse. Indeed, shown sorted ambiguously above are some examples that would fit left interpreted as follows: EX8220 "all are 'all are ___' " and EX8222 "palindrome with respect to being a palindrome with respect to ___", where every example is a palindrome with respect to some property. It may confuse the message to include any examples like these on the left side.

Clarification: EX8231, "palindromic with respect to the property the group is palindromic in" does not quite fit left in this Problem, although it does seem doubly-meta in some sense. Still, it is not palindromic with respect to the property of being palindromic. Instead of having the form "X 'X __' ", this is more like "X [the __appearing in 'X__'] ". Analogously, consider "gradual transition with respect to what the gradual transition is", "odd one out with respect to what property is the odd one out", etc. This idea could make for another Bongard Problem.

Below is a list of left example ideas that would be impossible to make. Please add and sign any more ideas you have.

"Exhaustive list of all exhaustive lists of all ____". - Aaron David Fairbanks, Nov 23 2020

The above description of the answer is probably not very helpful, so here are some worked examples. In one left example, out of the groups shown, all but one are "all but one are ___". In the corresponding right example, all are "all but one are ___" and all but one of these groups are black.

Other left examples include "every other is 'every other is ___' " and "gradually getting more like 'gradually getting more like ___' ".

Roughly speaking, right examples are meta while left examples are doubly-meta: "meta" is when the whole group imitates its parts; "doubly-meta" is when the whole group imitates its parts with respect to the way it imitates its parts.

This Problem is a "correspondence" (left-BP919) Problem. Examples in this Problem should ideally admit both a doubly-meta version and a singly-meta version. It may be very hard to interpret some examples if this Problem isn't presented helpfully (in a symmetrical format with corresponding squares opposite one another).

All left and right examples show five groups of five objects. (These numbers are unimportant; feel free to change this.) The objects in each group are aligned horizontally and the groups are stacked on top of one another vertically.

Each example showcases some property that depends on other properties: "X with respect to __[property]__".

Currently, in none of the uploaded left examples do all the groups shown follow the same pattern, so a valid alternative solution is "not all groups share some obvious property vs. all do." This is because the relevant group-property usually has to do with not all objects in it satisfying a certain property, and thus not all groups can satisfy that group-property. Some group-properties would allow the whole group to be the same, but these tend to be hard to parse. Indeed, shown sorted ambiguously above are some examples that would fit left interpreted as follows: EX8220 "all are 'all are ___' " and EX8222 "palindrome with respect to being a palindrome with respect to ___", where every example is a palindrome with respect to some property. It may confuse the message to include any examples like these on the left side.

Clarification: EX8231, "palindromic with respect to the property the group is palindromic in" does not quite fit left in this Problem, although it does seem doubly-meta in some sense. Still, it is not palindromic with respect to the property of being palindromic. Instead of having the form "X 'X __' ", this is more like "X [the __appearing in 'X ___'] ". Here X is being applied to the__ instead of the property X itself. Analogously, consider "gradual transition with respect to what the gradual transition is", "odd one out with respect to what property is the odd one out", etc. This idea could make for another Bongard Problem.

Below is a list of left example ideas that would be impossible to make. Please add and sign any more ideas you have.

"Exhaustive list of all exhaustive lists of all ____". - Aaron David Fairbanks, Nov 23 2020

The above description of the answer is probably not very helpful, so here are some worked examples. In one left example, out of the groups shown, all but one are "all but one are ___". In the corresponding right example, all are "all but one are ___" and all but one of these groups are black.

Other left examples include "every other is 'every other is ___' " and "gradually getting more like 'gradually getting more like ___' ".

Roughly speaking, right examples are meta while left examples are doubly-meta: "meta" is when the whole group imitates its parts; "doubly-meta" is when the whole group imitates its parts with respect to the way it imitates its parts.

This Problem is a "correspondence" (left-BP919) Problem. Examples in this Problem should ideally admit both a doubly-meta version and a singly-meta version. It may be very hard to interpret some examples if this Problem isn't presented helpfully (in a symmetrical format with corresponding squares opposite one another).

All left and right examples show five groups of five objects. (These numbers are unimportant; feel free to change this.) The objects in each group are aligned horizontally and the groups are stacked on top of one another vertically.

Each example showcases some property that depends on other properties: "X with respect to __[property]__".

Currently, in none of the uploaded left examples do all the groups shown follow the same pattern, so a valid alternative solution is "not all groups share some obvious property vs. all do." This is because the relevant group-property usually has to do with not all objects in it satisfying a certain property, and thus not all groups can satisfy that group-property. Some group-properties would allow the whole group to be the same, but these tend to be hard to parse. Indeed, shown sorted ambiguously above are some examples that would fit left interpreted as follows: EX8220 "all are 'all are ___' " and EX8222 "palindrome with respect to being a palindrome with respect to ___", where every example is a palindrome with respect to some property. It may confuse the message to include any examples like these on the left side.

Clarification: EX8231, "palindromic with respect to the property the group is palindromic in" does not quite fit left in this Problem, although it does seem doubly-meta in some sense. Still, it is not palindromic with respect to the property of being palindromic. Instead of having the form "X 'X __' ", this is more like "X [the __appearing in 'X ___'] "; X is now being applied to the__ instead of X itself. Analogously, consider "gradual transition with respect to what the gradual transition is", "odd one out with respect to what property is the odd one out", etc. This idea could make for another Bongard Problem.

Below is a list of left example ideas that would be impossible to make. Please add and sign any more ideas you have.

"Exhaustive list of all exhaustive lists of all ____". - Aaron David Fairbanks, Nov 23 2020

The above description of the answer is probably not very helpful, so here are some worked examples. In one left example, out of the groups shown, all but one are "all but one are ___". In the corresponding right example, all are "all but one are ___" and all but one of these groups are black.

Other left examples include "every other is 'every other is ___' " and "gradually getting more like 'gradually getting more like ___' ".

Roughly speaking, right examples are meta while left examples are doubly-meta: "meta" is when the whole group imitates its parts; "doubly-meta" is when the whole group imitates its parts with respect to the way it imitates its parts.

This Problem is a "correspondence" (left-BP919) Problem. Examples in this Problem should ideally admit both a doubly-meta version and a singly-meta version. It may be very hard to interpret some examples if this Problem isn't presented helpfully (in a symmetrical format with corresponding squares opposite one another).

All left and right examples show five groups of five objects. (These numbers are unimportant; feel free to change this.) The objects in each group are aligned horizontally and the groups are stacked on top of one another vertically.

Each example showcases some property that depends on other properties: "X with respect to __[property]__".

Currently, in none of the uploaded left examples do all the groups shown follow the same pattern, so a valid alternative solution is "not all groups share some obvious property vs. all do." This is because the relevant group-property usually has to do with not all objects in it satisfying a certain property, and thus not all groups can satisfy that group-property. Some group-properties would allow the whole group to be the same, but these tend to be hard to parse. Indeed, shown sorted ambiguously above are some examples that would fit left interpreted as follows: EX8220 "all are 'all are ___' " and EX8222 "palindrome with respect to being a palindrome with respect to ___", where every example is a palindrome with respect to some property. It may confuse the message to include any examples like these on the left side.

Clarification: EX8231, "palindromic with respect to the property the group is palindromic in" does not quite fit left in this Problem, although it does seem doubly-meta in some sense. Still, it is not palindromic with respect to the property of being palindromic.

Below is a list of left example ideas that would be impossible to make. Please add and sign any more ideas you have.

"Exhaustive list of all exhaustive lists of all ____". - Aaron David Fairbanks, Nov 23 2020

The above description of the answer is probably not very helpful, so here are some worked examples. In one left example, out of the groups shown, all but one are "all but one are ___". In the corresponding right example, all are "all but one are ___" and all but one of these groups are black.

Other left examples include "every other is 'every other is ___' " and "gradually getting more like 'gradually getting more like ___' ".

Roughly speaking, right examples are meta while left examples are doubly-meta: "meta" is when the whole group imitates its parts; "doubly-meta" is when the whole group imitates its parts with respect to the way it imitates its parts.

This Problem is a "correspondence" (left-BP919) Problem. Examples in this Problem should ideally admit both a doubly-meta version and a singly-meta version. It may be very hard to interpret some examples if this Problem isn't presented helpfully (in a symmetrical format with corresponding squares opposite one another).

All left and right examples show five groups of five objects. (These numbers are unimportant; feel free to change this.) The objects in each group are aligned horizontally and the groups are stacked on top of one another vertically.

Each example showcases some property that depends on other properties: "X with respect to __[property]__".

Currently, in none of the uploaded left examples do all the groups shown follow the same pattern, so a valid alternative solution is "not all groups share some obvious property vs. all do." This is because the relevant group-property usually has to do with not all objects in it satisfying a certain property, and thus not all groups can satisfy that group-property. Some group-properties would allow the whole group to be the same, but these tend to be hard to parse. Indeed, shown sorted ambiguously above are some examples that would fit left interpreted as follows: EX8220 "all are 'all are ___' " and EX8222 "palindrome with respect to being a palindrome with respect to ___", where every example is a palindrome with respect to some property. It may confuse the message to include any examples like these on the left side.

Clarification: EX8231, "palindromic with respect to the property the group is palindromic in" does not quite fit left in this Problem even though it does seem doubly-meta in some sense. Still, it is not palindromic with respect to the property of being palindromic.

Below is a list of left example ideas that would be impossible to make. Please add and sign any more ideas you have.

"Exhaustive list of all exhaustive lists of all ____". - Aaron David Fairbanks, Nov 23 2020

The above description of the answer is probably not very helpful, so here are some worked examples. In one left example, out of the groups shown, all but one are "all but one are ___". In the corresponding right example, all are "all but one are ___" and all but one of these groups are black.

Other left examples include "every other is 'every other is ___' " and "gradually getting more like 'gradually getting more like ___' ".

Roughly speaking, right examples are meta while left examples are doubly-meta: "meta" is when the whole group imitates its parts; "doubly-meta" is when the whole group imitates its parts with respect to the way it imitates its parts.

This Problem is a "correspondence" (left-BP919) Problem. Examples in this Problem should ideally admit both a doubly-meta version and a singly-meta version. It may be very hard to interpret some examples if this Problem isn't presented helpfully (in a symmetrical format with corresponding squares opposite one another).

All left and right examples show five groups of five objects. (These numbers are unimportant; feel free to change this.) The objects in each group are aligned horizontally and the groups are stacked on top of one another vertically.

Each example showcases some property that depends on other properties: "X with respect to __[property]__".

Currently, in none of the uploaded left examples do all the groups shown follow the same pattern, so a valid alternative solution is "not all groups share some obvious property vs. all do." This is because the relevant group-property usually has to do with not all objects in it satisfying a certain property, and thus not all groups can satisfy that group-property. Some group-properties would allow the whole group to be the same, but these tend to be hard to parse. Indeed, shown sorted ambiguously above are some examples that would fit left interpreted as follows: EX8220 "all are 'all are ___' " and EX8222 "palindrome with respect to being a palindrome with respect to ___", where every example is a palindrome with respect to some property. It may confuse the message to include any examples like these on the left side.

Clarification: EX8231, "palindromic with respect to the property the group is palindromic in" does not quite fit left in this Problem. It is not palindromic with respect to the property of being palindromic.

Below is a list of left example ideas that would be impossible to make. Please add and sign any more ideas you have.

"Exhaustive list of all exhaustive lists of all ____". - Aaron David Fairbanks, Nov 23 2020

The above description of the answer is probably not very helpful, so here are some worked examples. In one left example, out of the groups shown, all but one are "all but one are ___". In the corresponding right example, all are "all but one are ___" and all but one of these groups are black.

Other left examples include "every other is 'every other is ___' " and "gradually getting more like 'gradually getting more like ___' ".

Roughly speaking, right examples are meta while left examples are doubly-meta: "meta" is when the whole group imitates its parts; "doubly-meta" is when the whole group imitates its parts with respect to the way it imitates its parts.

This Problem is a "correspondence" (left-BP919) Problem. Examples in this Problem should ideally admit both a doubly-meta version and a singly-meta version. It may be very hard to interpret some examples if this Problem isn't presented helpfully (in a symmetrical format with corresponding squares opposite one another).

All left and right examples show five groups of five objects. (These numbers are unimportant; feel free to change this.) The objects in each group are aligned horizontally and the groups are stacked on top of one another vertically.

Each example showcases some property that depends on other properties: "X with respect to __[property]__".

Currently, in none of the uploaded left examples do all the groups shown follow the same pattern, so a valid alternative solution is "not all groups share some obvious property vs. all do." This is because the relevant group-property usually has to do with not all objects in it satisfying a certain property, and thus not all groups can satisfy that group-property. Some group-properties would allow the whole group to be the same, but these tend to be hard to parse. Indeed, shown sorted ambiguously above are some examples that would fit left interpreted as follows: EX8220 "all are 'all are ___' " and EX8222 "palindrome with respect to being a palindrome with respect to ___", where every example is a palindrome with respect to some property. It may confuse the message to include any examples like these on the left side.

Notice EX8231, "palindromic with respect to the property the group is palindromic in" does not quite fit left in this Problem. It is not palindromic with respect to the property of being palindromic.

Below is a list of left example ideas that would be impossible to make. Please add and sign any more ideas you have.

"Exhaustive list of all exhaustive lists of all ____". - Aaron David Fairbanks, Nov 23 2020

The above description of the answer is probably not very helpful, so here are some worked examples. In one left example, out of the groups shown, all but one are "all but one are ___". In the corresponding right example, all are "all but one are ___" and all but one of these groups are black.

Other left examples include "every other is 'every other is ___' " and "gradually getting more like 'gradually getting more like ___' ".

Roughly speaking, right examples are meta while left examples are doubly-meta: "meta" is when the whole group imitates its parts; "doubly-meta" is when the whole group imitates its parts with respect to the way it imitates its parts.

This Problem is a "correspondence" (left-BP919) Problem. Examples in this Problem should ideally admit both a doubly-meta version and a singly-meta version. It may be very hard to interpret some examples if this Problem isn't presented helpfully (in a symmetrical format with corresponding squares opposite one another).

All left and right examples show five groups of five objects. (These numbers are unimportant; feel free to change this.) The objects in each group are aligned horizontally and the groups are stacked on top of one another vertically.

Each example showcases some property that depends on other properties: "X with respect to __[property]__".

Currently, in none of the uploaded left examples do all the groups shown follow the same pattern, so a valid alternative solution is "not all groups share some obvious property vs. all do." This is because the relevant group-property usually has to do with not all objects in it satisfying a certain property, and thus not all groups can satisfy that group-property. Some group-properties would allow the whole group to be the same, but these tend to be hard to parse. Indeed, shown sorted ambiguously above are some examples that would fit left interpreted as follows: EX8220 "all are 'all are ___' " and EX8222 "palindrome with respect to being a palindrome with respect to ___", where every example is a palindrome with respect to some property. It may confuse the message to include any examples like these on the left side.

Notice EX8231, "palindromic with respect to the property the group is palindromic in" does not quite fit in this Problem. It is not palindromic with respect to the property of being palindromic.

Below is a list of left example ideas that would be impossible to make. Please add and sign any more ideas you have.

"Exhaustive list of all exhaustive lists of all ____". - Aaron David Fairbanks, Nov 23 2020

The above description of the answer is probably not very helpful, so here are some worked examples. In one left example, out of the groups shown, all but one are "all but one are ___". In the corresponding right example, all are "all but one are ___" and all but one of these groups are black.

Other left examples include "every other is 'every other is ___' " and "gradually getting more like 'gradually getting more like ___' ".

Roughly speaking, right examples are meta while left examples are doubly-meta: "meta" is when the whole group imitates its parts; "doubly-meta" is when the whole group imitates its parts with respect to the way it imitates its parts.

This Problem is a "correspondence" (left-BP919) Problem. Examples in this Problem should ideally admit both a doubly-meta version and a singly-meta version. It may be very hard to interpret some examples if this Problem isn't presented helpfully (in a symmetrical format with corresponding squares opposite one another).

All left and right examples show five groups of five objects. (These numbers are unimportant; feel free to change this.) The objects in each group are aligned horizontally and the groups are stacked on top of one another vertically.

Each example showcases some property that depends on other properties: "X with respect to __[property]__".

Currently, in none of the uploaded left examples do all the groups shown follow the same pattern, so a valid alternative solution is "not all groups share some obvious property vs. all do." This is because the relevant group-property usually has to do with not all objects in it satisfying a certain property, and thus not all groups can satisfy that group-property. Some group-properties would allow the whole group to be the same, but these tend to be hard to parse. Indeed, shown sorted ambiguously above are some examples that would fit left interpreted as follows: EX8220 "all are 'all are ___' " and EX8222 "palindrome with respect to being a palindrome with respect to ___", where every example is a palindrome with respect to some property. It may confuse the message to include any examples like these on the left side.

Notice EX8223, "palindromic with respect to the property the group is palindromic in" does not quite fit in this Problem. It is not palindromic with respect to the property of being palindromic.

Below is a list of left example ideas that would be impossible to make. Please add and sign any more ideas you have.

"Exhaustive list of all exhaustive lists of all ____". - Aaron David Fairbanks, Nov 23 2020

The above description of the answer is probably not very helpful, so here are some worked examples. In one left example, of the groups shown, all but one are "all but one are ___". In the corresponding right example, all are "all but one are ___" and all but one of these groups are black.

Other left examples include "every other is 'every other is ___' " and "gradually getting more like 'gradually getting more like ___' ".

Roughly speaking, right examples are meta while left examples are doubly-meta: "meta" is when the whole group imitates its parts; "doubly-meta" is when the whole group imitates its parts with respect to the way it imitates its parts.

This Problem is a "correspondence" (left-BP919) Problem. Examples in this Problem should ideally admit both a doubly-meta version and a singly-meta version. It may be very hard to interpret some examples if this Problem isn't presented helpfully (in a symmetrical format with corresponding squares opposite one another).

All left and right examples show five groups of five objects. (These numbers are unimportant; feel free to change this.) The objects in each group are aligned horizontally and the groups are stacked on top of one another vertically.

Each example showcases some property that depends on other properties: "X with respect to __[property]__".

Currently, in none of the uploaded left examples do all the groups shown follow the same pattern, so a valid alternative solution is "not all groups share some obvious property vs. all do." This is because the relevant group-property usually has to do with not all objects in it satisfying a certain property, and thus not all groups can satisfy that group-property. Some group-properties would allow the whole group to be the same, but these tend to be hard to parse. Indeed, shown sorted ambiguously above are some examples that would fit left interpreted as follows: EX8220 "all are 'all are ___' " and EX8222 "palindrome with respect to being a palindrome with respect to ___", where every example is a palindrome with respect to some property. It may confuse the message to include any examples like these on the left side.

Notice EX8223, "palindromic with respect to the property the group is palindromic in" does not quite fit in this Problem. It is not palindromic with respect to the property of being palindromic.

Below is a list of left example ideas that would be impossible to make. Please add and sign any more ideas you have.

"Exhaustive list of all exhaustive lists of all ____". - Aaron David Fairbanks, Nov 23 2020

The above description of the answer is not very helpful, so here is an example: of the groups in one of the left boxes, all but one are "all but one are ___". Of the groups in the corresponding right box, all are "all but one are ___" and all but one of these groups are black.

Other left examples include "every other is 'every other is ___' " and "gradually getting more like 'gradually getting more like ___' ".

Roughly speaking, right examples are meta while left examples are doubly-meta: "meta" is when the whole group imitates its parts; "doubly-meta" is when the whole group imitates its parts with respect to the way it imitates its parts.

This Problem is a "correspondence" (left-BP919) Problem. Examples in this Problem should ideally admit both a doubly-meta version and a singly-meta version. It may be very hard to interpret some examples if this Problem isn't presented helpfully (in a symmetrical format with corresponding squares opposite one another).

All left and right examples show five groups of five objects. (These numbers are unimportant; feel free to change this.) The objects in each group are aligned horizontally and the groups are stacked on top of one another vertically.

Each example showcases some property that depends on other properties: "X with respect to __[property]__".

Currently, in none of the uploaded left examples do all the groups shown follow the same pattern, so a valid alternative solution is "not all groups share some obvious property vs. all do." This is because the relevant group-property usually has to do with not all objects in it satisfying a certain property, and thus not all groups can satisfy that group-property. Some group-properties would allow the whole group to be the same, but these tend to be hard to parse. Indeed, shown sorted ambiguously above are some examples that would fit left interpreted as follows: EX8220 "all are 'all are ___' " and EX8222 "palindrome with respect to being a palindrome with respect to ___", where every example is a palindrome with respect to some property. It may confuse the message to include any examples like these on the left side.

Notice EX8223, "palindromic with respect to the property the group is palindromic in" does not quite fit in this Problem. It is not palindromic with respect to the property of being palindromic.

Below is a list of left example ideas that would be impossible to make. Please add and sign any more ideas you have.

"Exhaustive list of all exhaustive lists of all ____". - Aaron David Fairbanks, Nov 23 2020

The above description of the answer is not very helpful, so here is an example: of the groups in one of the left boxes, all but one are "all but one are ___". Of the groups in the corresponding right box, all are "all but one are ___" and all but one of these groups are black.

Other left examples include "every other is 'every other is ___' " and "gradually getting more like 'gradually getting more like ___' ".

Roughly speaking, right examples are meta while left examples are doubly-meta: "meta" is when the whole group imitates its parts; "doubly-meta" is when the whole group imitates its parts with respect to the way it imitates its parts.

This Problem is a "correspondence" (left-BP919) Problem. Examples in this Problem should ideally admit both a doubly-meta version and a singly-meta version. It may be very hard to interpret some examples if this Problem isn't presented helpfully (in a symmetrical format with corresponding squares opposite one another).

All left and right examples show five groups of five objects. (These numbers are unimportant; feel free to change this.) The objects in each group are aligned horizontally and the groups are stacked on top of one another vertically.

Each example showcases some property that depends on other properties: "X with respect to __[property]__".

Currently, in none of the uploaded left examples do all the groups shown follow the same pattern, so a valid alternative solution is "not all groups share some obvious property vs. all do." This is because the relevant group-property usually has to do with not all objects in it satisfying a certain property, and thus not all groups can satisfy that group-property. Some group-properties would allow the whole group to be the same, but these tend to be hard to parse. Indeed, shown sorted ambiguously above are some examples that would fit left interpreted as follows: EX8220 "all are 'all are ___' " and EX8222 "palindrome with respect to being a palindrome with respect to ___", where every example is a palindrome with respect to some property. It may confuse the message to include any examples like these on the left side.

Below is a list of left example ideas that would be impossible to make. Please add and sign any more ideas you have.

"Exhaustive list of all exhaustive lists of all ____". - Aaron David Fairbanks, Nov 23 2020