We can create pattern generators which will create any finite number of individuals. These can be quite insipid and one problem is that the generators will tend to be longer than the individuals. But could we (hypothetically) create a generator which could produce the variations of the individuals under the guise of a true generalities? The answer is probably but the generalizations would appear very insipid. We could, for example, just pick some particle of each of the series of individuals and develop a generalization for each particle. However, the left over particles might present some difficulties and the generalizations used in their generation would look pretty stretched. So what we really want to do, even if we cannot find a generator which is more compact than the individuals generated, is to have the generators interact in different kinds of ways so that the generalizations are different. They aren't all, for example, drawing one pixel in a series of images. Some are concerned with shapes some are with color others with texture and so on.
On Sat, Nov 16, 2013 at 7:28 PM, Jim Bromer <[email protected]> wrote: > Russ Hurlbut <[email protected]> wrote > "patterns can (not) generate patchwork (*instances*)...patterns are > totally non-generative" > "Patchworks are continually and infinitely generative, like the real, > everchanging, ever-evolvable world." > > > > I think that the question of whether a patchwork instance can be generated > by some pattern is not disputable (except from someone who has never seen > how computational pattern generators can produce unexpected results, or by > someone who demands that the word 'pattern' can only refer to the most > common traditional meaning of the word). But whether any 3, 4,... > patchwork instances can be produced by some pattern generator seems beyond > our understanding. The other issue is whether it is possible to produce > 'continuously' novel ways of modifying a pattern using a computational > pattern generator. > Jim Bromer > > > On Sat, Nov 16, 2013 at 12:53 PM, Russ Hurlbut <[email protected]> wrote: > >> Tinter: >> >>> It is commonplace for an ordinary person to look at those mag. covers >>> and say they "share/follow a pattern". (Note these covers are exceptions - >>> most patchworks in any collection are very different from each other). >>> >> ... >>> >> The argument simply cannot be made that patterns can generate >>> patchworks. They are OPPOSITES. A pattern can only have ONE >>> part/shape/colour at any given point. It is set form. A patchwork can have >>> INFINITE parts/shapes.colours at any given point. It is FREE form. One is >>> always the same. One is always different.One is rational and routine, the >>> other is creative and always out-of-the-routine. Patterns are totally >>> non-generative - totally non-related to AGI, which is about how to deal >>> with and produce new objects, new forms, new scenes - how to deal with a >>> new face, scene, text, argument. ... how to deal not with a neatly >>> patterned, toy blocks world, but a messy patchwork real world.Patchworks >>> are continually and infinitely generative, like the real, everchanging, >>> ever-evolvable world. >>> >> >> Thank you for your clarifications - and as always, your links to examples >> are appreciated. It appears that a key distinction that you are making is >> that of "generative versus non-generative". To at least this reader, the >> remaining differences appear to involve semantics. Perhaps introducing a >> new term into the conversation - that of "instance" - may help to clarify >> one's understanding. From the snippets above, it may appear to some that >> "patchwork" and "pattern" have been overloaded with multiple contexts. >> >> For "instances": >> "Most patchwork *instances* in any collection are very different from >> each other" >> "A pattern *instance *can only have ONE part/shape/colour at any given >> point" >> >> For the [non-instance | mental construct | general concept | abstract >> notion | whatever you want to call it]: >> "patterns can (not) generate patchwork (*instances*)...patterns are >> totally non-generative" >> "Patchworks are continually and infinitely generative, like the real, >> everchanging, ever-evolvable world." >> >> Using "pattern" in the non-instance context, one could argue that >> evolution is a pattern - which can even have meta-patterns (e.g >> http://en.wikipedia.org/wiki/Patterns_of_evolution ). You obviously >> disagree, so there is no point in further debate on this point. >> >> However, perhaps you can elucidate further the non-instance context of >> "patchwork", if in fact you agree there is a distinction and assuming each >> of the links provided are examples of patchwork *instances*. It is not >> apparent from you discourse what the non-instance context of "patchwork" >> would >> constitute. >> >> *AGI* | Archives <https://www.listbox.com/member/archive/303/=now> >> <https://www.listbox.com/member/archive/rss/303/24379807-f5817f28> | >> Modify<https://www.listbox.com/member/?&;>Your Subscription >> <http://www.listbox.com> >> > > > > -- > Jim Bromer > -- Jim Bromer ------------------------------------------- AGI Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/21088071-f452e424 Modify Your Subscription: https://www.listbox.com/member/?member_id=21088071&id_secret=21088071-58d57657 Powered by Listbox: http://www.listbox.com
