Right-sorted BPs have the keyword "gap" on the OEBP.

A Bongard Problem with a gap showcases two separate classes of objects.

For example, the Bongard Problem "white vs. black" (BP962) has a gap; there is no obvious choice of shared context between the two sides. One could imagine there is a spectrum of grays between them, or that there is a space of black-and-white bitmap images between them, or any number of other ambient contexts. The BP is about two separate cases.

What "small continuous change" means depends on the context. In BP6, adding a corner seems an abrupt change, even though it may be a small visual change in terms of pixels on the screen. A change in the class of object someone parses is not considered a small change here.

Contrast "allsorted" (left-BP509).

Bongard Problems about comparing quantities on a spectrum (keyword "spectrum" BP507) should not usually be considered gap BPs (including discrete spectra--these are borderline cases); a spectrum establishes a shared context, with examples on both sides of the BP landing somewhere on it. But if it is reasonable to imagine getting the solution without parsing a spectrum, consider it a gap, since that means the ambient context is unclear.

Bongard Problems with gaps seem like completely arbitrary (left-BP950) non-sequiturs when the two classes of objects are particularly unrelated.

Right-sorted BPs have the keyword "gap" on the OEBP.

A Bongard Problem with a gap showcases two separate classes of objects.

For example, the Bongard Problem "white vs. black" (BP962) has a gap; there is no obvious choice of shared context between the two sides. One could imagine there is a spectrum of grays between them, or that there is a space of black-and-white bitmap images between them, or any number of other ambient contexts. The BP is about two separate cases.

What "small continuous change" means depends on the context. In BP6, adding a corner seems an abrupt change, even though it may be a small visual change in terms of pixels on the screen. A change in the class of object someone parses is not considered a small change here.

Contrast "allsorted" (left-BP509).

Bongard Problems about comparing quantities on a spectrum (keyword "spectrum" BP507) should not usually be considered gap BPs (including discrete spectra); a spectrum establishes a shared context, with examples on both sides of the BP landing somewhere on it. But if it is reasonable to imagine getting the solution without parsing a spectrum, consider it a gap, since that means the ambient context is unclear.

Bongard Problems with gaps seem like completely arbitrary (left-BP950) non-sequiturs when the two classes of objects are particularly unrelated.

Right-sorted BPs have the keyword "gap" on the OEBP.

A Bongard Problem with a gap showcases two separate classes of objects.

For example, the Bongard Problem "white vs. black" (BP962) has a gap; there is no obvious choice of shared context between the two sides. One could imagine there is a spectrum of grays between them, or that there is a space of black-and-white bitmap images between them, or any number of other ambient contexts. The BP is about two separate cases.

What "small continuous change" means depends on the context. In BP6, adding a corner seems an abrupt change, even though it may be a small visual change in terms of pixels on the screen. A change in the class of object someone parses is not considered a small change here.

Contrast "allsorted" (left-BP509).

Bongard Problems about comparing quantities on a spectrum (keyword "spectrum" BP507) should not usually be considered gap BPs (including discrete spectra); a spectrum establishes a shared context, with examples on both sides of the BP landing somewhere on it. But if it is reasonable to imagine getting the solution without parsing a spectrum, consider it a gap, since that means the ambient context is unclear.

Some Bongard Problems with gaps seem arbitrary (left-BP950) when the two classes of objects are particularly unrelated, complete non-sequiturs.

Right-sorted BPs have the keyword "gap" on the OEBP.

A Bongard Problem with a gap showcases two separate classes of objects.

For example, the Bongard Problem "white vs. black" (BP962) has a gap; there is no obvious choice of shared context between the two sides. One could imagine there is a spectrum of grays between them, or that there is a space of black-and-white bitmap images between them, or any number of other ambient contexts. The BP is about two separate cases.

What "small continuous change" means depends on the context. In BP6, adding a corner seems an abrupt change, even though it may be a small visual change in terms of pixels on the screen. A change in the class of object someone parses is not considered a small change here.

Contrast "allsorted" (left-BP509).

Bongard Problems about comparing quantities on a spectrum (keyword "spectrum" BP507) should not usually be considered gap BPs (including discrete spectra); a spectrum establishes a shared context, with examples on both sides of the BP landing somewhere on it. But if it is reasonable to imagine getting the solution without parsing a spectrum, consider it a gap, since that means the ambient context is unclear.

Some Bongard Problems with gaps may seem arbitrary (left-BP950), as non-sequiturs, when the two classes of objects are particularly unrelated.

Right-sorted BPs have the keyword "gap" on the OEBP.

A Bongard Problem with a gap showcases two separate classes of objects.

For example, the Bongard Problem "white vs. black" (BP962) has a gap; there is no obvious choice of shared context between the two sides. One could imagine there is a spectrum of grays between them, or that there is a space of black-and-white bitmap images between them, or any number of other ambient contexts. The BP is about two separate cases.

What "small continuous change" means depends on the context. In BP6, adding a corner seems an abrupt change, crossing a boundary, even though it may be a small visual change in terms of pixels on the screen. A change in the class of object someone parses is not considered a small change here.

Contrast "allsorted" (left-BP509).

Bongard Problems about comparing quantities on a spectrum (keyword "spectrum" BP507) should not usually be considered gap BPs (including discrete spectra); a spectrum establishes a shared context, with examples on both sides of the BP landing somewhere on it. But if it is reasonable to imagine getting the solution without parsing a spectrum, consider it a gap, since that means the ambient context is unclear.

Some Bongard Problems with gaps may seem arbitrary (left-BP950), as non-sequiturs, when the two classes of objects are particularly unrelated.

If Bongard Problem has a "gap", and there no potential examples that are hard to decide whether they fit in the problem (there is a clear boundary to the collection of examples the problem sorts), then it should furthermore also be clear which side any potential example fits (keyword "exact" left-BP508).

See BP1140 for the version of this about any (perhaps large) additions of detail instead of context-dependent gradual change.

If Bongard Problem has a "gap", and there no potential examples that are hard to decide whether they fit in the problem (there is a clear boundary to the collection of examples the problem sorts), then it should furthermore also be clear which side any potential example fits (keyword "exact" left-BP508).

See BP1140 for the version of this about any (perhaps large) additions of visual detail instead of context-dependent gradual change.

If Bongard Problem has a "gap", and there no potential examples that are hard to decide whether they fit in the problem (there is a clear boundary to the collection of examples the problem sorts), then it should furthermore also be clear which side any potential example fits (keyword "exact" left-BP508).

See BP1140 for the version of this about additions of visual detail instead of context-dependent gradual change.

Right-sorted BPs have the keyword "gap" on the OEBP.

A Bongard Problem with a gap showcases two separate classes of objects.

For example, the Bongard Problem "white vs. black" (BP962) has a gap; there is no obvious choice of shared context between the two sides. One could imagine there is a spectrum of grays between them, or that there is a space of black-and-white bitmap images between them, or any number of other ambient contexts. The BP is about two separate cases.

What "small continuous change" means depends on the context. In BP6, adding a corner seems an abrupt change, crossing a boundary, even though it may be a small visual change in terms of pixels on the screen. A change in the class of object someone parses is not to be considered a small change.

Contrast "allsorted" (left-BP509).

Bongard Problems about comparing quantities on a spectrum (keyword "spectrum" BP507) should not usually be considered gap BPs (including discrete spectra); a spectrum establishes a shared context, with examples on both sides of the BP landing somewhere on it. But if it is reasonable to imagine getting the solution without parsing a spectrum, consider it a gap, since that means the ambient context is unclear.

Some Bongard Problems with gaps may seem arbitrary (left-BP950), as non-sequiturs, when the two classes of objects are particularly unrelated.

Bongard Problems such that making small successive changes can switch an example's sorting vs. Bongard Problems in which the two sides are so different that it is impossible to cross the gap by making successive small changes to examples (there is no middle-ground between the sides and there is no obvious choice of shared ambient context both sides are part of).

Right-sorted BPs have the keyword "gap" on the OEBP.

A Bongard Problem with a gap showcases two separate classes of objects.

For example, the Bongard Problem "white vs. black" (BP962) has a gap; there is no obvious choice of shared context between the two sides. One could imagine there is a spectrum of grays between them, or that there is a space of black-and-white bitmap images between them, or any number of other ambient contexts. The BP is about two separate cases.

Contrast "allsorted" (left-BP509).

Bongard Problems about comparing quantities on a spectrum (keyword "spectrum" BP507) should not usually be considered gap BPs (including discrete spectra); a spectrum establishes a shared context, with examples on both sides of the BP landing somewhere on it. But if it is reasonable to imagine getting the solution without parsing a spectrum, consider it a gap, since that means the ambient context is unclear.

Some Bongard Problems with gaps may seem arbitrary (left-BP950), as non-sequiturs, when the two classes of objects are particularly unrelated.

Bongard Problems such that making small successive changes can switch an example's sorting vs. Bongard Problems in which the two sides are so different that it is impossible to cross the gap by making successive small changes to examples (there is no middle-ground between the sides and there is no obvious choice of shared ambient world both sides are part of).

Right-sorted BPs have the keyword "gap" on the OEBP.

A Bongard Problem with a gap showcases two completely separate classes of objects.

For example, the Bongard Problem "white vs. black" (BP962) has a gap; there is no obvious choice of shared context between the two sides. One could imagine there is a spectrum of grays between them, or that there is a space of black-and-white bitmap images between them, or any number of other ambient contexts. The BP is about two separate cases.

Contrast "allsorted" (left-BP509).

Bongard Problems about comparing quantities on a spectrum (keyword "spectrum" BP507) should not usually be considered gap BPs (including discrete spectra); a spectrum establishes a shared context, with examples on both sides of the BP landing somewhere on it. But if it is reasonable to imagine getting the solution without parsing a spectrum, consider it a gap, since that means the ambient context is unclear.

Some Bongard Problems with gaps may seem arbitrary (left-BP950), as non-sequiturs, when the two classes of objects are particularly unrelated.

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

A Bongard Problem with a gap showcases two completely separate classes of objects.

For example, the Bongard Problem "white vs. black" (BP962) has a gap; there is no obvious choice of shared context between the two sides. One could imagine there is a spectrum of grays between them, or that there is a space of black-and-white bitmap images between them, or any number of other ambient contexts. The BP is about two separate cases.

Contrast "allsorted" (left-BP509).

Bongard Problems about comparing quantities on a spectrum (keyword "spectrum" BP507) should not usually be considered gap BPs (including discrete spectra); a spectrum establishes a shared context, with examples on both sides of the BP landing somewhere on it. But if it is reasonable to imagine getting the solution without parsing a spectrum, consider it a gap, since that means the ambient context is unclear.

Some Bongard Problems with gaps may seem arbitrary (left-BP950), as non-sequiturs, when the two classes of objects are particularly unrelated.

If Bongard Problem has a "gap", and there no potential examples that are hard to decide whether they fit in the problem (there is a clear boundary to the collection of examples the problem sorts), then it should furthermore also be clear which side any potential example fits (keyword "exact" left-BP508).

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

A Bongard Problem with a gap showcases two completely separate classes of objects.

For example, the Bongard Problem "white vs. black" (BP962) has a gap; there is no obvious choice of shared context between the two sides. One could imagine there is a spectrum of grays between them, or that there is a space of black-and-white bitmap images between them, or any number of other ambient contexts. The BP is about two separate cases.

Contrast "allsorted" (left-BP509).

Bongard Problems about comparing quantities on a spectrum (keyword "spectrum" BP507) should not be considered gap BPs (including discrete spectra); a spectrum establishes a shared context, with examples on both sides of the BP landing somewhere on it. But if it is reasonable to imagine getting the solution without parsing a spectrum, consider it a gap, since that means the ambient context is unclear.

Some Bongard Problems with gaps may seem arbitrary (left-BP950), as non-sequiturs, when the two classes of objects are particularly unrelated.

Bongard Problems with a "gap" in which it is always clear whether an example should fit in the Bongard Problem or not are "exact" (left-BP508).

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

A Problem with a gap showcases two completely separate classes of objects.

For example, the Problem "white vs. black" (BP962) has a gap; there is no obvious choice of shared context between the two sides. One could imagine there is a spectrum of grays between them, or that there is a space of black-and-white bitmap images between them, or any number of other ambient contexts. The BP is about two separate cases.

Contrast "allsorted" (left-BP509).

Problems about comparing quantities on a spectrum (keyword "spectrum" BP507) should not be considered gap BPs (including discrete spectra); a spectrum establishes a shared context, with examples on both sides of the BP landing somewhere on it. But if it is reasonable to imagine getting the solution without parsing a spectrum, consider it a gap, since that means the ambient context is unclear.

Some BPs with gaps may seem arbitrary (left-BP950), as non-sequiturs, when the two classes of objects are particularly unrelated.

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

A Problem with a gap showcases two completely separate classes of objects.

For example, the Problem "white vs. black" (BP962) has a gap; there is no obvious choice of shared context between the two sides. One could imagine there is a spectrum of grays between them, or that there is a space of black-and-white bitmap images between them, or any number of other ambient contexts. The BP is about two separate cases.

Contrast "wholesort" (left-BP509).

Problems about comparing quantities on a spectrum (keyword "spectrum" BP507) should not be considered gap BPs (including discrete spectra); a spectrum establishes a shared context, with examples on both sides of the BP landing somewhere on it. But if it is reasonable to imagine getting the solution without parsing a spectrum, consider it a gap, since that means the ambient context is unclear.

Some BPs with gaps may seem arbitrary (left-BP950), as non-sequiturs, when the two classes of objects are particularly unrelated.

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

A Problem with a gap showcases two completely separate classes of objects.

For example, the Problem "white vs. black" (BP962) has a gap; there is no obvious choice of shared context between the two sides. One could imagine there is a spectrum of grays between them, or that there is a space of black-and-white bitmap images between them, or any number of other ambient contexts. The BP is about two separate cases.

Contrast "wholesort" (left-BP509).

Problems about comparing quantities on a spectrum (keyword "spectrum" BP507) should not be considered gap BPs (including discrete spectra); a spectrum establishes a shared context, with examples on both sides of the BP landing somewhere on it. But if it is reasonable to imagine getting the solution without parsing a spectrum, consider it a gap, since that means the ambient context is unclear.

Some BPs with gaps may seem arbitrary (left-BP950), as non-sequiturs, because the two classes of objects are so unrelated.

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

A Problem with a gap showcases two completely separate classes of objects. Some BPs with gaps may sometimes seem arbitrary (left-BP950), as non-sequiturs, because the two classes of objects are so unrelated.

For example, the Problem "white vs. black" (BP962) has a gap; there is no obvious choice of shared context between the two sides. One could imagine there is a spectrum of grays between them, or that there is a space of black-and-white bitmap images between them, or any number of other ambient contexts. The BP is about two separate cases.

Contrast "wholesort" (left-BP509).

Whether there is a "gap" becomes vague in cases of discrete spectra. For example, with BP6, "triangle vs. quadrilateral," triangles and quadrilaterals can seem like separate cases with no middle ground, but they are also neighbors on the spectrum of polygons with respect side number, and from that perspective there is no gap. Many gaps could even be understood as discrete 0-to-1 spectra. Generally, in these cases, if it is reasonable to imagine understanding the Problem's solution without parsing a spectrum, consider it a gap, since that means the ambient context is unclear.

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

A Problem with a gap showcases two completely separate classes of objects. Some BPs with gaps may sometimes seem arbitrary (left-BP950), as non-sequiturs, because the two classes of objects are so unrelated.

For example, the Problem "white vs. black" (BP962) has a gap; there is no obvious choice of shared context between the two sides. One could imagine there is a spectrum of grays between them, or that there is a space of black-and-white bitmap images between them, or any number of other ambient contexts. The Problem is about two separate cases.

Contrast "wholesort" (left-BP509).

Whether there is a "gap" becomes vague in cases of discrete spectra. For example, with BP6, "triangle vs. quadrilateral," triangles and quadrilaterals can seem like separate cases with no middle ground, but they are also neighbors on the spectrum of polygons with respect side number, and from that perspective there is no gap. Many gaps could even be understood as discrete 0-to-1 spectra. Generally, in these cases, if it is reasonable to imagine understanding the Problem's solution without parsing a spectrum, consider it a gap, since that means the ambient context is unclear.

Bongard Problems in which the two sides are so different that it is impossible to cross the gap by making successive small changes to examples (there is no middle-ground between the sides and there is no obvious choice of shared ambient world both sides are part of) vs. other Bongard Problems.

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

A Problem with a gap showcases two completely separate classes of objects. Some BPs with gaps may sometimes seem arbitrary (left-BP950), as non-sequiturs, because the two classes of objects are so unrelated.

For example, the Problem "white vs. black" (BP962) has a gap; there is no obvious choice of shared context between the two sides. One could imagine there is a spectrum of grays between them, or that there are pixel-by-pixel slightly changed black-and-white bitmap images between them, or any number of other ambient contexts. The Problem is about two separate cases.

Contrast "wholesort" (left-BP509).

Whether there is a "gap" becomes vague in cases of discrete spectra. For example, with BP6, "triangle vs. quadrilateral," triangles and quadrilaterals can seem like separate cases with no middle ground, but they are also neighbors on the spectrum of polygons with respect side number, and from that perspective there is no gap. Many gaps could even be understood as discrete 0-to-1 spectra. Generally, in these cases, if it is reasonable to imagine understanding the Problem's solution without parsing a spectrum, consider it a gap, since that means the ambient context is unclear.

Bongard Problems in which the two sides are so different that it is impossible to cross the gap by making successive small changes to examples (there is no intuitive middle-ground between the sides and there is no obvious choice of shared ambient world both sides are part of) vs. other Bongard Problems.

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

A Problem with a gap showcases two completely separate classes of objects. Problems with gaps may sometimes feel arbitrary, like non-sequiturs, because the two classes of objects are so unrelated.

For example, the Problem "white vs. black" (BP962) has a gap; there is no obvious choice of shared context between the two sides. One could imagine there is a spectrum of grays between them, or that there are pixel-by-pixel slightly changed black-and-white bitmap images between them, or any number of other ambient contexts. The Problem is about two separate cases.

Contrast "wholesort" (left-BP509).

Whether there is a "gap" becomes vague in cases of discrete spectra. For example, with BP6, "triangle vs. quadrilateral," triangles and quadrilaterals can seem like separate cases with no middle ground, but they are also neighbors on the spectrum of polygons with respect side number, and from that perspective there is no gap. Many gaps could even be understood as discrete 0-to-1 spectra. Generally, in these cases, if it is reasonable to imagine understanding the Problem's solution without parsing a spectrum, consider it a gap, since that means the ambient context is unclear.