Left-sorted Bongard Problems have the the keyword "left-narrow" on the OEBP.

Call a rule "narrow" if it is likely to be noticed in a large collection of examples, without any counterexamples provided.

A collection of triangles will be recognized as such; "triangles" is a narrow rule. A collection of non-triangular shapes will just be seen as "shapes"; "not triangles" is not narrow.

Intuitively, a narrow rule seems small in comparison to the space of other related possibilities. Narrow rules tend to be phrased positively ("is [property]"), while non-narrow rules opposite narrow rules tend to be phrased negatively ("is not [property]").

Both sides of a Bongard Problem can be narrow, e.g. BP6.

Even a rule and its conceptual opposite can be narrow, e.g. BP20.

A Bongard Problem such that one side is narrow and the other side is the non-narrow opposite reads as the narrow side being a subset of the other. See BP881.

What seems like a typical example depends on expectations. (See the keyword assumesfamiliarity for Bongard Problems that require the solver to go in with special expectations.)

A person might notice the absence of triangles in a collection of just polygons, because a triangle is such a typical example of a polygon. On the other hand, a person will probably not notice the absence of 174-gons in a collection of polygons.

Typically, any example fitting a narrow rule can be changed slightly to no longer fit. (This is not always the case, however. Consider the narrow rule "is approximately a triangle".) See the keyword stable.

It is possible for a rule to be "narrow" (communicable by a properly chosen collection of examples) but not clearly communicated by a particular collection of examples satisfying it, e.g., a collection of examples that is too small to communicate it.

Note that this is not just BP514 (right-narrow) flipped.

Is it possible for a rule to be such that some collections of examples do bring it to mind, but no collection of examples unambiguously communicates it as the intended rule? Perhaps there is some border case the rule excludes, but it is not clear whether the border case was intentionally left out. The border case's absence would likely become more conspicuous with more examples (assuming the collection of examples naturally brings this border case to mind).

See BP830 for a version with pictures of Bongard Problems (miniproblems) instead of links.

Adjacent-numbered pages:
BP508 BP509 BP510 BP511 BP512  *  BP514 BP515 BP516 BP517 BP518

dual, meta (see left/right), links, keyword, side

bp [smaller | same | bigger]

Aaron David Fairbanks

Left-sorted Bongard Problems have the the keyword "right-narrow" on the OEBP.

This sorts Bongard Problems based on how BP513 (left-narrow) would sort them if they were flipped; see that page for a description.

Adjacent-numbered pages:
BP509 BP510 BP511 BP512 BP513  *  BP515 BP516 BP517 BP518 BP519

dual, meta (see left/right), links, keyword, side

bp [smaller | same | bigger]

Aaron David Fairbanks

Left examples have the keyword "teach" on the OEBP.

Sometimes instead of gauging somebody's ability to guess the pattern, a Bongard Problem might teach the pattern.

Consider a Bongard Problem whose left examples are images of a specific person's face; after seeing that Problem, one might be able to recognize that person.

A "teach" Bongard Problem (with a huge number of examples) could be taken as a training set for machine learning.

"Teach" BPs tend to be convoluted, arbitrary, cultural-knowledge-based (keyword culture), or they illustrate some insight that might be overlooked, perhaps mathematical (keyword math).

Adjacent-numbered pages:
BP853 BP854 BP855 BP856 BP857  *  BP859 BP860 BP861 BP862 BP863

meta (see left/right), links, keyword

bp [smaller | same | bigger]

Aaron David Fairbanks

Left-sorted Bongard Problems have the keyword "creativeexamples" on the OEBP.

Be encouraged to contribute new interesting examples to Bongard Problems with this keyword.

There is much overlap with the keyword hardsort.

This is what it usually means to say examples fit on (e.g.) the left of a Bongard Problem in various creative ways: there is no (obvious) general method to determine a left-fitting example fits left.

There is a related idea in computability theory: a "non recursively enumerable" property is one that cannot in general be checked by a computer algorithm.

But keep in mind the tag "creativeexamples" is supposed to mean something less formal. For example, it requires no ingenuity for a human being to check when a simple shape is convex or concave (so BP4 is not labelled "creativeexamples"). However, it is not as if we use an algorithm to do this, like a computer. (It is not even clear what an "algorithm" would mean in this context, since it is ambiguous both what class of shapes the Bongard Problem sorts and how that would be encoded into a computer program's input. There are usually many options and ambiguities like this whenever one tries to formalize the content of a Bongard Problem.)

Adjacent-numbered pages:
BP861 BP862 BP863 BP864 BP865  *  BP867 BP868 BP869 BP870 BP871

notso, meta (see left/right), links, keyword

bp [smaller | same | bigger]

Aaron David Fairbanks

Adjacent-numbered pages:
BP881 BP882 BP883 BP884 BP885  *  BP887 BP888 BP889 BP890 BP891

meta (see left/right), links, metaconcept

bp [smaller | same | bigger]

Aaron David Fairbanks

Left examples have the keyword "contributepairs" on the OEBP.

When this keyword is added to a Problem, OEBP users are advised to add a corresponding right example for every left example they add and vice versa.

It is common for Bongard Problems to present left examples on the left side and corresponding altered versions of those examples on the right side, tweaked only slightly, to highlight the difference and make the solution easier to see (see keyword help).

This is common in more abstract Bongard Problems that admit a wide range of examples, a variety of different styles or types (e.g. BP360). Showing two versions of the same thing, one on the left and one on the right, helps a person interpret what that thing is meant to be in the context of the Bongard Problem; whatever qualities vary between the two in the pair must be relevant.

If a person cannot sort an example according to the solution property without seeing its corresponding opposite example, the Bongard Problem is invalid (see https://www.oebp.org/invalid.php ). There is no one rule dividing the sides; the solution is not a method to determine whether an arbitrary example fits left or right. See also Bongard Problems with the keyword collective, which are similarly borderline-invalid.

A BP in which each left example corresponds to a right example and vice versa could be remade as a Bongard Problem in which the left examples are the pairs. For example BP360 would turn into "a pair consisting of the ordered version of something and the chaotic version of the same thing vs. a pair of things not satisfying this relationship." This process would turn a Bongard Problem that is invalid in the sense described above into a valid one.

(See keyword orderedpair.)

In some "contributepairs" Bongard Problems there really is a natural choice of left version for every right example and vice versa (see keyword dual); in others the choice is artificially imposed by the Bongard Problem creator.

When "contributepairs" Bongard Problems are laid out in the format with a grid of boxes on either side of a dividing line, the boxes may be arranged so as to highlight the correspondence: either

A B | A B

E F | E F

G H | G H

or

A B | B A

E F | F E

G H | H G.

Adjacent-numbered pages:
BP914 BP915 BP916 BP917 BP918  *  BP920 BP921 BP922 BP923 BP924

meta (see left/right), links, keyword, oebp, right-self, instruction

bppage [smaller | same | bigger]
zoom in left (correspondence_bp)

Aaron David Fairbanks

Adjacent-numbered pages:
BP1172 BP1173 BP1174 BP1175 BP1176  *  BP1178 BP1179 BP1180 BP1181 BP1182

meta (see left/right), links

Leo Crabbe

Left-sorted Problems have the keyword "miniworlds" on the OEBP.

All examples in this Problem are visual Bongard Problems with multiple objects in most panels. This is key as an intuitive set of allowable objects needs to be communicated by any one sorted image.

There is a decent degree of overlap between rules and "miniworlds", but BP1049 is an example of a "miniworlds" problem where the rule is constant across examples, and BP1155 is an example of a "rules" Problem that would not be tagged "miniworlds".

Although this Problem does sort any BP whose examples are images of Bongard Problems left, it is probably best not to consider them to avoid clutter and more unnecessary keywords being attached to them.

Adjacent-numbered pages:
BP1175 BP1176 BP1177 BP1178 BP1179  *  BP1181 BP1182 BP1183 BP1184 BP1185

meta (see left/right), links, keyword

visualbp [smaller | same | bigger]

Leo Crabbe