BPs sorted left are tagged with the keyword "abstract" on the OEBP. The solution is not an easily-checked or concretely-defined geometrical or numerical property in pictures.

Adjacent-numbered pages:
BP507 BP508 BP509 BP510 BP511  *  BP513 BP514 BP515 BP516 BP517

abstract, meta (see left/right), links, keyword, left-self, sideless

bp [smaller | same | bigger]

Aaron David Fairbanks

Bongard Problems sorted left have the keyword "meta" on the OEBP.

Meta Bongard Problems are Bongard Problems that sort Bongard Problems. Sometimes abbreviated MBPs.

The first meta Bongard Problem was BP200.

Some meta BP pages sort images of Bongard Problems (keyword miniproblems), while other meta BP pages sort other BP pages (keyword links).

BPs that sort meta-BPs are labelled metameta.

Adjacent-numbered pages:
BP532 BP533 BP534 BP535 BP536  *  BP538 BP539 BP540 BP541 BP542

meta (see left/right), links, keyword, world, left-self, sideless, metameta, left-full, feedback

bp [smaller | same | bigger]
zoom in left (linksbp)

Aaron David Fairbanks

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

Examples of "structures": Bongard Problem, Bongard Problem with extra unsorted panel ("Bongard's Dozen"), 4-panel analogy grid, sequence of objects with a quantity changing by a constant amount.

If the solver hasn't become familiar with the featured structure, the Bongard Problem's solution may seem convoluted or inelegant. (See keyword assumesfamiliarity.) Once the solver gets used to seeing a particular structure it becomes easier to read that structure and solve Bongard Problems featuring it.

A Bongard Problem can non-verbally teach someone how a particular structure works, showing valid examples of that structure versus non-examples. E.g., BP968 for the structure of Bongard Problems and BP981 for the structure of analogy grids.

Adjacent-numbered pages:
BP784 BP785 BP786 BP787 BP788  *  BP790 BP791 BP792 BP793 BP794

meta (see left/right), links, keyword

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 examples have the keyword "infodense" on the OEBP.

Consider the amount of data a person has to consciously unpack in each example in the process of determining how it should be sorted. In BP3, it is only necessary to notice the color of the shape. In BP871, however, it is important to read various qualities of every tiny shape shown.

Images of Bongard Problems that are "infodense" typically need to include a large number of examples in order to communicate the solution clearly without admitting unintended solutions. With so much data packed in each example, it becomes more likely that some of the random patterns in the data will happen to distinguish between the two sides in an unintended way. A similar issue appears in convoluted Bongard Problems.

Contrast "infodense" Problems to hardsort Bongard Problems, in which examples are difficult to sort, but perhaps that difficulty does not stem from reading a high amount of information; perhaps there is a small amount of information extracted from the examples, but it is hard to determine whether or not that information fits a rule.

Adjacent-numbered pages:
BP973 BP974 BP975 BP976 BP977  *  BP979 BP980 BP981 BP982 BP983

abstract, spectrum, meta (see left/right), links, keyword

Aaron David Fairbanks

Bongard Problems sorted left have the keyword "links" on the OEBP.

Bongard Problems sorted right have the keyword "miniproblems" on the OEBP.

The keyword "links" is automatically added to a Bongard Problem on the OEBP if a BP number is added as an example.

Meta Bongard problems that sort Bongard Problems purely based on their solutions (keyword presentationmatters) usually have two versions in the database: one that sorts images of Bongard Problems and one that sorts links to pages on the OEBP. If both versions exist, users should make them cross-reference one another.

All the examples of miniature Bongard Problems within any meta Bongard Problem tagged "miniproblems" would fit left on BP1080 (which is a showcase of the various formats for images of Bongard Problems).

Adjacent-numbered pages:
BP1121 BP1122 BP1123 BP1124 BP1125  *  BP1127 BP1128 BP1129 BP1130 BP1131

meta (see left/right), links, keyword, world, left-self, metameta

metabp [smaller | same | bigger]

Aaron David Fairbanks

This Problem is mostly concerned with categorising BPs whose examples aren't necessarily formatted as traditional Bongard Problems.

Adjacent-numbered pages:
BP1149 BP1150 BP1151 BP1152 BP1153  *  BP1155 BP1156 BP1157 BP1158 BP1159

meta (see left/right), links

visualbp [smaller | same | bigger]

Leo Crabbe

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

In the typical "rules" Bongard Problem, it is possible to come up with many convoluted rules that fit each example, but the intended interpretation is the only simple and obvious one.

Since it is difficult to communicate a rule with little detail, "rules" Bongard Problems are usually infodense.

Typically, each example is itself a bunch of smaller examples that all obey the rule. It is the same as how a Bongard Problems relies on many examples to communicate rules; likely just one example wouldn't get the answer across.

On the other hand, in BP1157 for example, each intended rule is communicated by just one example; these rules have to be particularly simple and intuitive, and the individual examples have to be complicated enough to communicate them.

Often, each rule is communicated by showing several examples of things satisfying it. (See keywords left-narrow and right-narrow.) Contrast Bongard Problems, which are more communicative, by showing some examples satisfying the rule and some examples NOT satisfying the rule.

A "rules" Bongard Problem is often collective. Some examples may admit multiple equally plausible rules, and the correct interpretation of each example only becomes clear once the solution is known. The group of examples together improve the solver's confidence about having understood each individual one right.

It is common that there will be one or two examples with multiple reasonable interpretations due to oversight of the author.

All meta Bongard Problems are "rules" Bongard Problems.

Many other Bongard-Problem-like structures seen on the OEBP are also about recognizing a pattern. (See keyword structure.)

"Rules" Bongard Problems are abstract, although the individual rules in them may not be abstract. "Rules" Bongard Problems also usually have the keyword creativeexamples.

Adjacent-numbered pages:
BP1153 BP1154 BP1155 BP1156 BP1157  *  BP1159 BP1160 BP1161 BP1162 BP1163

fuzzy, meta (see left/right), links, keyword, left-self, rules

Aaron David Fairbanks

Any miniproblems BP with all examples in the same format will be sorted left.

See also the keyword structure.

Adjacent-numbered pages:
BP1160 BP1161 BP1162 BP1163 BP1164  *  BP1166 BP1167 BP1168 BP1169 BP1170

notso, meta (see left/right), links

visualbp [smaller | same | bigger]

Aaron David Fairbanks

Not all visual Bongard Problems whose examples all have some amount of black in them are sorted left. In left-sorted BPs, there must be some concrete percentage of black that is exceeded by every sorted example.

Adjacent-numbered pages:
BP1161 BP1162 BP1163 BP1164 BP1165  *  BP1167 BP1168 BP1169 BP1170 BP1171

meta (see left/right), links

Leo Crabbe