Re: [opencog-dev] Storytelling

[William Taylor]

Thank you all for the information you have shared. I have
Probabilistic Logic Networks, and Engineering General Intelligence 1
and 2, and Hidden Pattern. I am reviewing c++ and python. I don't have
a working version of opencog on my computer yet though. I am working
on my masters in System Science which is the study of things with a
relation between them. I just need a termination project which is
working code.

I want to create a model of the program. What is wrong with the
thinking model that I described? All information is stored in the mind
and recursively asks questions, give commands, and makes declarations
all of which are high on the heirarchy that has submodalities and
their deformations through time at the bottom? All of this is prompted
by sensory input. That is my idea in a nutshell.

Thank you again for sharing your ideas and knowledge with me. Is it
possible to mention and describe Opencog on a vblog that I am starting
on youtube? Who do I need permission from? 


http://www.avg.com/email-signature?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail";
target="_blank">https://ipmcdn.avast.com/images/icons/icon-envelope-tick-green-avg-v1.png";
alt="" width="46" height="29" style="width: 46px; height: 29px;"
/>
Virus-free. http://www.avg.com/email-signature?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail";
target="_blank" style="color: #4453ea;">www.avg.com




On 10/2/17, Murilo Saraiva de Queiroz  wrote:
> About post-scarcity economy, uploads and such: the new book from Cory
> Doctorow, Walkaway, is pretty interesting! I didnt' finish it yet, but I'm
> enjoying a lot of stuff there.
>
> On Sun, Oct 1, 2017 at 11:02 PM, Linas Vepstas 
> wrote:
>
>> You might enjoy the AGI mailing list. If you cast your net more widely,
>> there's the extropians, the trans-humanists, the life-extensionists,
>> certain classes of economists, and much much more. There are TED talks
>> about unemployment caused by AI/AGI.  Ray Kurzweil has written half a
>> dozen
>> books, there are others before him, and our very own Ben Geortzel just
>> published his sixth or seventh maybe a few months ago or so, this latest
>> one is non-technical, general readership.
>>
>> And of course, the original Star Trek series was set in a post-scarcity
>> economy.  You would name whatever you wanted, and the replicator would
>> just
>> create it, for nothing, for free.  Earl Grey tea. The recent Star Trek
>> reboots completely lost sight of this utopia, and replaced it by
>> pointless
>> stupid action-adventure.
>>
>> --linas
>>
>> On Mon, Oct 2, 2017 at 9:29 AM, William Taylor 
>> wrote:
>>
>>> "So does everyone. Easy to say, hard to do. You can implement quick
>>> cheesy hacks in a few months or a few years.  You can spend ten or 20
>>> yeas after that, and still not get even close.  Many, many many have
>>> tried, no one has been successful yet.  Not even partly-successful."
>>>
>>> Heh, I have the rest of my life to work on this. I am still learning
>>> c++ python and opencog.
>>>
>>> Lots and lots of people talk about this.  The only problem is that its
>>> actually really hard to do this, technically, politically, socially,
>>> economically, legally.
>>>
>>> I haven't heard anyone talking about this idea. My researching skills
>>> suck. Has anyone tried a recursive function that starts with an
>>> outside stimulus, coordinates and runs thousands of smaller
>>> experiential patterns in an effort to maximize a utility function? In
>>> other words the system would get a string from the outside world such
>>> as: System gets a request for a hamburger. It performs a
>>> transderivational search by running the object of a hamburger through
>>> all of the modalities and transforms such as aroma of hamburgers,
>>> shape and color of hamburgers, texture and resistant pressure of
>>> hamburgers and so forth. It then explores the relations of hamburgers
>>> like where to get them, how much they cost and etc. These give a model
>>> of the hamburger, the thought. I think a thought is a model of reality
>>> and it consists of internal and external sensory perceptions ordered
>>> in a specific way. Our brain has thousands of them. We constantly
>>> model things. The mind uses models to ask questions, declare
>>> statements, and issue commands. English language is a basic model of
>>> human thought and as such suggests there are three kinds of thoughts,
>>> declarations, questions and commands. A thinking "function" could be a
>>> recursive function that tests for an external sense or senses and
>>> builds a model by asking questions, making declarations, or issuing
>>> orders. It can receive a language input asking it to create a love
>>> story. It does a transderivational search for love stories and models
>>> a love story, distorts the results from the search to create a new
>>> story about love.
>>>
>>>
>>> Lots and lots of people talk 

Re: [opencog-dev] Storytelling

[Murilo Saraiva de Queiroz]

About post-scarcity economy, uploads and such: the new book from Cory
Doctorow, Walkaway, is pretty interesting! I didnt' finish it yet, but I'm
enjoying a lot of stuff there.

On Sun, Oct 1, 2017 at 11:02 PM, Linas Vepstas 
wrote:

> You might enjoy the AGI mailing list. If you cast your net more widely,
> there's the extropians, the trans-humanists, the life-extensionists,
> certain classes of economists, and much much more. There are TED talks
> about unemployment caused by AI/AGI.  Ray Kurzweil has written half a dozen
> books, there are others before him, and our very own Ben Geortzel just
> published his sixth or seventh maybe a few months ago or so, this latest
> one is non-technical, general readership.
>
> And of course, the original Star Trek series was set in a post-scarcity
> economy.  You would name whatever you wanted, and the replicator would just
> create it, for nothing, for free.  Earl Grey tea. The recent Star Trek
> reboots completely lost sight of this utopia, and replaced it by pointless
> stupid action-adventure.
>
> --linas
>
> On Mon, Oct 2, 2017 at 9:29 AM, William Taylor 
> wrote:
>
>> "So does everyone. Easy to say, hard to do. You can implement quick
>> cheesy hacks in a few months or a few years.  You can spend ten or 20
>> yeas after that, and still not get even close.  Many, many many have
>> tried, no one has been successful yet.  Not even partly-successful."
>>
>> Heh, I have the rest of my life to work on this. I am still learning
>> c++ python and opencog.
>>
>> Lots and lots of people talk about this.  The only problem is that its
>> actually really hard to do this, technically, politically, socially,
>> economically, legally.
>>
>> I haven't heard anyone talking about this idea. My researching skills
>> suck. Has anyone tried a recursive function that starts with an
>> outside stimulus, coordinates and runs thousands of smaller
>> experiential patterns in an effort to maximize a utility function? In
>> other words the system would get a string from the outside world such
>> as: System gets a request for a hamburger. It performs a
>> transderivational search by running the object of a hamburger through
>> all of the modalities and transforms such as aroma of hamburgers,
>> shape and color of hamburgers, texture and resistant pressure of
>> hamburgers and so forth. It then explores the relations of hamburgers
>> like where to get them, how much they cost and etc. These give a model
>> of the hamburger, the thought. I think a thought is a model of reality
>> and it consists of internal and external sensory perceptions ordered
>> in a specific way. Our brain has thousands of them. We constantly
>> model things. The mind uses models to ask questions, declare
>> statements, and issue commands. English language is a basic model of
>> human thought and as such suggests there are three kinds of thoughts,
>> declarations, questions and commands. A thinking "function" could be a
>> recursive function that tests for an external sense or senses and
>> builds a model by asking questions, making declarations, or issuing
>> orders. It can receive a language input asking it to create a love
>> story. It does a transderivational search for love stories and models
>> a love story, distorts the results from the search to create a new
>> story about love.
>>
>>
>> Lots and lots of people talk about this.  The only problem is that its
>> actually really hard to do this, technically, politically, socially,
>> economically, legally.
>>
>> No matter what I would like to join in on the conversation and learn
>> to shape my idea
>> (model) better so that it fits reality. Please point me to books,
>> threads, anything so that I can learn from the giants before me and
>> stand on their shoulders. That way I may see something I wasn't able
>> to see before. Please excuse the sloppy language in trying to explain
>> my idea. I am excited to share these foolish dreams and I am perhaps
>> writing too quickly. > id="DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2">
>> 
>> 
>> > href="http://www.avg.com/email-signature?utm_medium=email&;
>> utm_source=link&utm_campaign=sig-email&utm_content=webmail"
>> target="_blank">> src="https://ipmcdn.avast.com/images/icons/icon-envelope-tic
>> k-green-avg-v1.png"
>> alt="" width="46" height="29" style="width: 46px; height: 29px;"
>> />
>> Virus-free. > href="http://www.avg.com/email-signature?utm_medium=email&;
>> utm_source=link&utm_campaign=sig-email&utm_content=webmail"
>> target="_blank" style="color: #4453ea;">www.avg.com
>> 
>> 
>> > height="1">
>>
>> On 10/1/17, Linas Vepstas  wrote:
>> > On Sun, Oct 1, 2017 at 11:25 PM, William Taylor <
>> wtaylorjr2...@gmail.com>
>> > wrote:
>> >
>> >> Please excuse the double post. I have an idea that I would like to try
>> >> out. I would like to design modalities, sight sound taste touch and
>> >> smell and general. Each of these are in turn broken into
>> >> submodalities, F

Re: [opencog-dev] Storytelling

[Linas Vepstas]

You might enjoy the AGI mailing list. If you cast your net more widely,
there's the extropians, the trans-humanists, the life-extensionists,
certain classes of economists, and much much more. There are TED talks
about unemployment caused by AI/AGI.  Ray Kurzweil has written half a dozen
books, there are others before him, and our very own Ben Geortzel just
published his sixth or seventh maybe a few months ago or so, this latest
one is non-technical, general readership.

And of course, the original Star Trek series was set in a post-scarcity
economy.  You would name whatever you wanted, and the replicator would just
create it, for nothing, for free.  Earl Grey tea. The recent Star Trek
reboots completely lost sight of this utopia, and replaced it by pointless
stupid action-adventure.

--linas

On Mon, Oct 2, 2017 at 9:29 AM, William Taylor 
wrote:

> "So does everyone. Easy to say, hard to do. You can implement quick
> cheesy hacks in a few months or a few years.  You can spend ten or 20
> yeas after that, and still not get even close.  Many, many many have
> tried, no one has been successful yet.  Not even partly-successful."
>
> Heh, I have the rest of my life to work on this. I am still learning
> c++ python and opencog.
>
> Lots and lots of people talk about this.  The only problem is that its
> actually really hard to do this, technically, politically, socially,
> economically, legally.
>
> I haven't heard anyone talking about this idea. My researching skills
> suck. Has anyone tried a recursive function that starts with an
> outside stimulus, coordinates and runs thousands of smaller
> experiential patterns in an effort to maximize a utility function? In
> other words the system would get a string from the outside world such
> as: System gets a request for a hamburger. It performs a
> transderivational search by running the object of a hamburger through
> all of the modalities and transforms such as aroma of hamburgers,
> shape and color of hamburgers, texture and resistant pressure of
> hamburgers and so forth. It then explores the relations of hamburgers
> like where to get them, how much they cost and etc. These give a model
> of the hamburger, the thought. I think a thought is a model of reality
> and it consists of internal and external sensory perceptions ordered
> in a specific way. Our brain has thousands of them. We constantly
> model things. The mind uses models to ask questions, declare
> statements, and issue commands. English language is a basic model of
> human thought and as such suggests there are three kinds of thoughts,
> declarations, questions and commands. A thinking "function" could be a
> recursive function that tests for an external sense or senses and
> builds a model by asking questions, making declarations, or issuing
> orders. It can receive a language input asking it to create a love
> story. It does a transderivational search for love stories and models
> a love story, distorts the results from the search to create a new
> story about love.
>
>
> Lots and lots of people talk about this.  The only problem is that its
> actually really hard to do this, technically, politically, socially,
> economically, legally.
>
> No matter what I would like to join in on the conversation and learn
> to shape my idea
> (model) better so that it fits reality. Please point me to books,
> threads, anything so that I can learn from the giants before me and
> stand on their shoulders. That way I may see something I wasn't able
> to see before. Please excuse the sloppy language in trying to explain
> my idea. I am excited to share these foolish dreams and I am perhaps
> writing too quickly.  id="DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2">
> 
> 
>  href="http://www.avg.com/email-signature?utm_medium=
> email&utm_source=link&utm_campaign=sig-email&utm_content=webmail"
> target="_blank"> src="https://ipmcdn.avast.com/images/icons/icon-envelope-
> tick-green-avg-v1.png"
> alt="" width="46" height="29" style="width: 46px; height: 29px;"
> />
> Virus-free.  href="http://www.avg.com/email-signature?utm_medium=
> email&utm_source=link&utm_campaign=sig-email&utm_content=webmail"
> target="_blank" style="color: #4453ea;">www.avg.com
> 
> 
>  height="1">
>
> On 10/1/17, Linas Vepstas  wrote:
> > On Sun, Oct 1, 2017 at 11:25 PM, William Taylor  >
> > wrote:
> >
> >> Please excuse the double post. I have an idea that I would like to try
> >> out. I would like to design modalities, sight sound taste touch and
> >> smell and general. Each of these are in turn broken into
> >> submodalities, For example sight has shape, color, position. Shape can
> >> be a collection of points in xyz three dimensional space. Feeling has
> >> temperature, pressure, texture. General has intensity, location,
> >> duration. Sound has pitch, volume, and so forth.
> >
> >
> > Sure, we do some of this already, for the robots. The code is here:
> >
> > https://github.com/opencog/ros-b

Re: [opencog-dev] Storytelling

[William Taylor]

"So does everyone. Easy to say, hard to do. You can implement quick
cheesy hacks in a few months or a few years.  You can spend ten or 20
yeas after that, and still not get even close.  Many, many many have
tried, no one has been successful yet.  Not even partly-successful."

Heh, I have the rest of my life to work on this. I am still learning
c++ python and opencog.

Lots and lots of people talk about this.  The only problem is that its
actually really hard to do this, technically, politically, socially,
economically, legally.

I haven't heard anyone talking about this idea. My researching skills
suck. Has anyone tried a recursive function that starts with an
outside stimulus, coordinates and runs thousands of smaller
experiential patterns in an effort to maximize a utility function? In
other words the system would get a string from the outside world such
as: System gets a request for a hamburger. It performs a
transderivational search by running the object of a hamburger through
all of the modalities and transforms such as aroma of hamburgers,
shape and color of hamburgers, texture and resistant pressure of
hamburgers and so forth. It then explores the relations of hamburgers
like where to get them, how much they cost and etc. These give a model
of the hamburger, the thought. I think a thought is a model of reality
and it consists of internal and external sensory perceptions ordered
in a specific way. Our brain has thousands of them. We constantly
model things. The mind uses models to ask questions, declare
statements, and issue commands. English language is a basic model of
human thought and as such suggests there are three kinds of thoughts,
declarations, questions and commands. A thinking "function" could be a
recursive function that tests for an external sense or senses and
builds a model by asking questions, making declarations, or issuing
orders. It can receive a language input asking it to create a love
story. It does a transderivational search for love stories and models
a love story, distorts the results from the search to create a new
story about love.


Lots and lots of people talk about this.  The only problem is that its
actually really hard to do this, technically, politically, socially,
economically, legally.

No matter what I would like to join in on the conversation and learn
to shape my idea
(model) better so that it fits reality. Please point me to books,
threads, anything so that I can learn from the giants before me and
stand on their shoulders. That way I may see something I wasn't able
to see before. Please excuse the sloppy language in trying to explain
my idea. I am excited to share these foolish dreams and I am perhaps
writing too quickly. 


http://www.avg.com/email-signature?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail";
target="_blank">https://ipmcdn.avast.com/images/icons/icon-envelope-tick-green-avg-v1.png";
alt="" width="46" height="29" style="width: 46px; height: 29px;"
/>
Virus-free. http://www.avg.com/email-signature?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail";
target="_blank" style="color: #4453ea;">www.avg.com




On 10/1/17, Linas Vepstas  wrote:
> On Sun, Oct 1, 2017 at 11:25 PM, William Taylor 
> wrote:
>
>> Please excuse the double post. I have an idea that I would like to try
>> out. I would like to design modalities, sight sound taste touch and
>> smell and general. Each of these are in turn broken into
>> submodalities, For example sight has shape, color, position. Shape can
>> be a collection of points in xyz three dimensional space. Feeling has
>> temperature, pressure, texture. General has intensity, location,
>> duration. Sound has pitch, volume, and so forth.
>
>
> Sure, we do some of this already, for the robots. The code is here:
>
> https://github.com/opencog/ros-behavior-scripting/tree/master/sensors
>
> See for example, "room brightness" or "sound track".  Its pretty stupid,
> but it works and gets used,
>
>
>> An object has each of
>> these qualities. Through time these objects can be deformed. For
>> example: red color, round shape. Smell is apple scent. Feeling is
>> hard, and crunchy and juicy.
>
>
> Sure.  There's a write up for how do do this here:
>
>
> https://github.com/opencog/opencog/blob/master/opencog/eva/architecture/embodiment.pdf
>
> Some of it is implemented; its crude.
>
>
> On a higher level is the idea that this
>> is an apple. It can be deformed as chunks over time are taken out of
>> the apple. The higher level idea is that the apple is being eaten.
>> Higher yet again is the thought, this apple tastes good.
>>
>
> Sure, but this is where the problem gets hard.
>
>>
>> Ultimately I would like to create a program that can generate ideas
>> and put them into text.
>
>
> So does everyone. Easy to say, hard to do. You can implement quick cheesy
> hacks in a few months or a few years.  You can spend ten or 20 yeas af

Re: [opencog-dev] Storytelling

[Linas Vepstas]

On Sun, Oct 1, 2017 at 11:25 PM, William Taylor 
wrote:

> Please excuse the double post. I have an idea that I would like to try
> out. I would like to design modalities, sight sound taste touch and
> smell and general. Each of these are in turn broken into
> submodalities, For example sight has shape, color, position. Shape can
> be a collection of points in xyz three dimensional space. Feeling has
> temperature, pressure, texture. General has intensity, location,
> duration. Sound has pitch, volume, and so forth.


Sure, we do some of this already, for the robots. The code is here:

https://github.com/opencog/ros-behavior-scripting/tree/master/sensors

See for example, "room brightness" or "sound track".  Its pretty stupid,
but it works and gets used,


> An object has each of
> these qualities. Through time these objects can be deformed. For
> example: red color, round shape. Smell is apple scent. Feeling is
> hard, and crunchy and juicy.


Sure.  There's a write up for how do do this here:


https://github.com/opencog/opencog/blob/master/opencog/eva/architecture/embodiment.pdf

Some of it is implemented; its crude.


On a higher level is the idea that this
> is an apple. It can be deformed as chunks over time are taken out of
> the apple. The higher level idea is that the apple is being eaten.
> Higher yet again is the thought, this apple tastes good.
>

Sure, but this is where the problem gets hard.

>
> Ultimately I would like to create a program that can generate ideas
> and put them into text.


So does everyone. Easy to say, hard to do. You can implement quick cheesy
hacks in a few months or a few years.  You can spend ten or 20 yeas after
that, and still not get even close.  Many, many many have tried, no one has
been successful yet.  Not even partly-successful.


> Then it can animate the text creating movies.
> I want to create this system so that in the end, the program can
> create items of value. I also want to create, or help create a
> distribution system with the property of getting the text or movie to
> those who want it.
>
> My dream is to create a program that in turn creates items of value
> and distributes them to those who want it. This I believe is a wealth
> generation system and it will be able to reduce poverty. I know it
> sounds crazy but it is my goal.


Lots and lots of people talk about this.  The only problem is that its
actually really hard to do this, technically, politically, socially,
economically, legally.


> It is the reason I need a storyteller.
> I would like to use opencog to make it.
>
> 
> 
> 
>  href="http://www.avg.com/email-signature?utm_medium=
> email&utm_source=link&utm_campaign=sig-email&utm_content=webmail"
> target="_blank"> src="https://ipmcdn.avast.com/images/icons/icon-envelope-
> tick-green-avg-v1.png"
> alt="" width="46" height="29" style="width: 46px; height: 29px;"
> />
> Virus-free.  href="http://www.avg.com/email-signature?utm_medium=
> email&utm_source=link&utm_campaign=sig-email&utm_content=webmail"
> target="_blank" style="color: #4453ea;">www.avg.com
> 
> 
>  height="1">
>
> On 10/1/17, Linas Vepstas  wrote:
> > Ooof dah.  A master's thesis is not much time, so at best I can suggest
> > only quick hacks.  And you probably mostly would not need opencog to
> > implement them.
> >
> > One quick hack would be to try to discover which bundles or groups of
> words
> > dominate nearby paragraphs, and which groups of words recur throughout
> the
> > story.  You can then replace those words by other words.  Alternately, if
> > two paragraphs are closely tied to one-another, but weakly tied to the
> rest
> > of the text, you can cut them out entirely, and replace them by some
> other
> > paragraphs from elsewhere (after adjusting word-content).
> >
> > By "words", I really mean nouns and entity names, and all the pronouns
> that
> > refer to them (he, she, it ...)  just try to see which ones are near to
> > one-another, maybe that's enough to "understand" what a paragraph is
> > talking about.  Verbs might be fun to play with. but you want to avoid
> > creating sentences like "green ideas sleep furiously".  Thus, cutting and
> > replacing entire paragraphs makes more sense; and then adjust the nouns
> so
> > that the story is still about the same topics.
> >
> > --linas
> >
> > On Sun, Oct 1, 2017 at 9:50 PM, William Taylor 
> > wrote:
> >
> >> Thank you for responding. If you can give me some ideas, I may be able
> >> to create the code myself, or help create it. I am not the greatest
> >> programmer but I can learn and improve. I would like to use this for
> >> my termination project for my Masters at Binghamton University in New
> >> York.
> >>
> >> On 10/1/17, Linas Vepstas  wrote:
> >> > Not at this time. We have ideas about this, but no code. --linas
> >> >
> >> > On Wed, Sep 27, 2017 at 4:21 AM, wtaylorjr2...@gmail.com <
> >> > wtaylorjr2...@gmail.com> wrote:
> >> >
> >> >> Is it possi

Re: [opencog-dev] Storytelling

[William Taylor]

Please excuse the double post. I have an idea that I would like to try
out. I would like to design modalities, sight sound taste touch and
smell and general. Each of these are in turn broken into
submodalities, For example sight has shape, color, position. Shape can
be a collection of points in xyz three dimensional space. Feeling has
temperature, pressure, texture. General has intensity, location,
duration. Sound has pitch, volume, and so forth. An object has each of
these qualities. Through time these objects can be deformed. For
example: red color, round shape. Smell is apple scent. Feeling is
hard, and crunchy and juicy. On a higher level is the idea that this
is an apple. It can be deformed as chunks over time are taken out of
the apple. The higher level idea is that the apple is being eaten.
Higher yet again is the thought, this apple tastes good.

Ultimately I would like to create a program that can generate ideas
and put them into text. Then it can animate the text creating movies.
I want to create this system so that in the end, the program can
create items of value. I also want to create, or help create a
distribution system with the property of getting the text or movie to
those who want it.

My dream is to create a program that in turn creates items of value
and distributes them to those who want it. This I believe is a wealth
generation system and it will be able to reduce poverty. I know it
sounds crazy but it is my goal. It is the reason I need a storyteller.
I would like to use opencog to make it.




http://www.avg.com/email-signature?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail";
target="_blank">https://ipmcdn.avast.com/images/icons/icon-envelope-tick-green-avg-v1.png";
alt="" width="46" height="29" style="width: 46px; height: 29px;"
/>
Virus-free. http://www.avg.com/email-signature?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail";
target="_blank" style="color: #4453ea;">www.avg.com




On 10/1/17, Linas Vepstas  wrote:
> Ooof dah.  A master's thesis is not much time, so at best I can suggest
> only quick hacks.  And you probably mostly would not need opencog to
> implement them.
>
> One quick hack would be to try to discover which bundles or groups of words
> dominate nearby paragraphs, and which groups of words recur throughout the
> story.  You can then replace those words by other words.  Alternately, if
> two paragraphs are closely tied to one-another, but weakly tied to the rest
> of the text, you can cut them out entirely, and replace them by some other
> paragraphs from elsewhere (after adjusting word-content).
>
> By "words", I really mean nouns and entity names, and all the pronouns that
> refer to them (he, she, it ...)  just try to see which ones are near to
> one-another, maybe that's enough to "understand" what a paragraph is
> talking about.  Verbs might be fun to play with. but you want to avoid
> creating sentences like "green ideas sleep furiously".  Thus, cutting and
> replacing entire paragraphs makes more sense; and then adjust the nouns so
> that the story is still about the same topics.
>
> --linas
>
> On Sun, Oct 1, 2017 at 9:50 PM, William Taylor 
> wrote:
>
>> Thank you for responding. If you can give me some ideas, I may be able
>> to create the code myself, or help create it. I am not the greatest
>> programmer but I can learn and improve. I would like to use this for
>> my termination project for my Masters at Binghamton University in New
>> York.
>>
>> On 10/1/17, Linas Vepstas  wrote:
>> > Not at this time. We have ideas about this, but no code. --linas
>> >
>> > On Wed, Sep 27, 2017 at 4:21 AM, wtaylorjr2...@gmail.com <
>> > wtaylorjr2...@gmail.com> wrote:
>> >
>> >> Is it possible to use opencog to input stories and break them apart,
>> >> distort them, and then assemble them in new ways to create "new"
>> stories?
>> >>
>> >> --
>> >> You received this message because you are subscribed to the Google
>> Groups
>> >> "opencog" group.
>> >> To unsubscribe from this group and stop receiving emails from it, send
>> an
>> >> email to opencog+unsubscr...@googlegroups.com.
>> >> To post to this group, send email to opencog@googlegroups.com.
>> >> Visit this group at https://groups.google.com/group/opencog.
>> >> To view this discussion on the web visit https://groups.google.com/d/
>> >> msgid/opencog/85ed2664-5369-41f7-aa2a-5bea845a8f00%40googlegroups.com
>> >> > 41f7-aa2a-5bea845a8f00%40googlegroups.com?utm_medium=
>> email&utm_source=footer>
>> >> .
>> >> For more options, visit https://groups.google.com/d/optout.
>> >>
>> >
>> >
>> >
>> > --
>> > *"The problem is not that artificial intelligence will get too smart
>> > and
>> > take over the world," computer scientist Pedro Domingos writes, "the
>> > problem is that it's too stupid and already has." *
>> >
>> > --
>> > You received this m

Re: [opencog-dev] Storytelling

[Linas Vepstas]

Ooof dah.  A master's thesis is not much time, so at best I can suggest
only quick hacks.  And you probably mostly would not need opencog to
implement them.

One quick hack would be to try to discover which bundles or groups of words
dominate nearby paragraphs, and which groups of words recur throughout the
story.  You can then replace those words by other words.  Alternately, if
two paragraphs are closely tied to one-another, but weakly tied to the rest
of the text, you can cut them out entirely, and replace them by some other
paragraphs from elsewhere (after adjusting word-content).

By "words", I really mean nouns and entity names, and all the pronouns that
refer to them (he, she, it ...)  just try to see which ones are near to
one-another, maybe that's enough to "understand" what a paragraph is
talking about.  Verbs might be fun to play with. but you want to avoid
creating sentences like "green ideas sleep furiously".  Thus, cutting and
replacing entire paragraphs makes more sense; and then adjust the nouns so
that the story is still about the same topics.

--linas

On Sun, Oct 1, 2017 at 9:50 PM, William Taylor 
wrote:

> Thank you for responding. If you can give me some ideas, I may be able
> to create the code myself, or help create it. I am not the greatest
> programmer but I can learn and improve. I would like to use this for
> my termination project for my Masters at Binghamton University in New
> York.
>
> On 10/1/17, Linas Vepstas  wrote:
> > Not at this time. We have ideas about this, but no code. --linas
> >
> > On Wed, Sep 27, 2017 at 4:21 AM, wtaylorjr2...@gmail.com <
> > wtaylorjr2...@gmail.com> wrote:
> >
> >> Is it possible to use opencog to input stories and break them apart,
> >> distort them, and then assemble them in new ways to create "new"
> stories?
> >>
> >> --
> >> You received this message because you are subscribed to the Google
> Groups
> >> "opencog" group.
> >> To unsubscribe from this group and stop receiving emails from it, send
> an
> >> email to opencog+unsubscr...@googlegroups.com.
> >> To post to this group, send email to opencog@googlegroups.com.
> >> Visit this group at https://groups.google.com/group/opencog.
> >> To view this discussion on the web visit https://groups.google.com/d/
> >> msgid/opencog/85ed2664-5369-41f7-aa2a-5bea845a8f00%40googlegroups.com
> >>  41f7-aa2a-5bea845a8f00%40googlegroups.com?utm_medium=
> email&utm_source=footer>
> >> .
> >> For more options, visit https://groups.google.com/d/optout.
> >>
> >
> >
> >
> > --
> > *"The problem is not that artificial intelligence will get too smart and
> > take over the world," computer scientist Pedro Domingos writes, "the
> > problem is that it's too stupid and already has." *
> >
> > --
> > You received this message because you are subscribed to the Google Groups
> > "opencog" group.
> > To unsubscribe from this group and stop receiving emails from it, send an
> > email to opencog+unsubscr...@googlegroups.com.
> > To post to this group, send email to opencog@googlegroups.com.
> > Visit this group at https://groups.google.com/group/opencog.
> > To view this discussion on the web visit
> > https://groups.google.com/d/msgid/opencog/CAHrUA35xb2%
> 3DC0TDZGsfYTM3LH0Yps1JG4F_C%2B14gcfZ5TxxO%2Bg%40mail.gmail.com.
> > For more options, visit https://groups.google.com/d/optout.
> >
>
> --
> You received this message because you are subscribed to the Google Groups
> "opencog" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to opencog+unsubscr...@googlegroups.com.
> To post to this group, send email to opencog@googlegroups.com.
> Visit this group at https://groups.google.com/group/opencog.
> To view this discussion on the web visit https://groups.google.com/d/
> msgid/opencog/CADu5FeH5XtRYQaPfMy1gf8kibzA-wXbayJ83TCxGwEBszDDAZA%40mail.
> gmail.com.
> For more options, visit https://groups.google.com/d/optout.
>



-- 
*"The problem is not that artificial intelligence will get too smart and
take over the world," computer scientist Pedro Domingos writes, "the
problem is that it's too stupid and already has." *

-- 
You received this message because you are subscribed to the Google Groups 
"opencog" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to opencog+unsubscr...@googlegroups.com.
To post to this group, send email to opencog@googlegroups.com.
Visit this group at https://groups.google.com/group/opencog.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/opencog/CAHrUA35-By%3DOtH_2MPT4Cmqv%2BNnL%2BFc%2BmH5hHnE94_%2B%3DoZccYQ%40mail.gmail.com.
For more options, visit https://groups.google.com/d/optout.

Re: [opencog-dev] Storytelling

[William Taylor]

Thank you for responding. If you can give me some ideas, I may be able
to create the code myself, or help create it. I am not the greatest
programmer but I can learn and improve. I would like to use this for
my termination project for my Masters at Binghamton University in New
York.

On 10/1/17, Linas Vepstas  wrote:
> Not at this time. We have ideas about this, but no code. --linas
>
> On Wed, Sep 27, 2017 at 4:21 AM, wtaylorjr2...@gmail.com <
> wtaylorjr2...@gmail.com> wrote:
>
>> Is it possible to use opencog to input stories and break them apart,
>> distort them, and then assemble them in new ways to create "new" stories?
>>
>> --
>> You received this message because you are subscribed to the Google Groups
>> "opencog" group.
>> To unsubscribe from this group and stop receiving emails from it, send an
>> email to opencog+unsubscr...@googlegroups.com.
>> To post to this group, send email to opencog@googlegroups.com.
>> Visit this group at https://groups.google.com/group/opencog.
>> To view this discussion on the web visit https://groups.google.com/d/
>> msgid/opencog/85ed2664-5369-41f7-aa2a-5bea845a8f00%40googlegroups.com
>> 
>> .
>> For more options, visit https://groups.google.com/d/optout.
>>
>
>
>
> --
> *"The problem is not that artificial intelligence will get too smart and
> take over the world," computer scientist Pedro Domingos writes, "the
> problem is that it's too stupid and already has." *
>
> --
> You received this message because you are subscribed to the Google Groups
> "opencog" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to opencog+unsubscr...@googlegroups.com.
> To post to this group, send email to opencog@googlegroups.com.
> Visit this group at https://groups.google.com/group/opencog.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/opencog/CAHrUA35xb2%3DC0TDZGsfYTM3LH0Yps1JG4F_C%2B14gcfZ5TxxO%2Bg%40mail.gmail.com.
> For more options, visit https://groups.google.com/d/optout.
>

-- 
You received this message because you are subscribed to the Google Groups 
"opencog" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to opencog+unsubscr...@googlegroups.com.
To post to this group, send email to opencog@googlegroups.com.
Visit this group at https://groups.google.com/group/opencog.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/opencog/CADu5FeH5XtRYQaPfMy1gf8kibzA-wXbayJ83TCxGwEBszDDAZA%40mail.gmail.com.
For more options, visit https://groups.google.com/d/optout.

Re: [opencog-dev] Storytelling

[Linas Vepstas]

Not at this time. We have ideas about this, but no code. --linas

On Wed, Sep 27, 2017 at 4:21 AM, wtaylorjr2...@gmail.com <
wtaylorjr2...@gmail.com> wrote:

> Is it possible to use opencog to input stories and break them apart,
> distort them, and then assemble them in new ways to create "new" stories?
>
> --
> You received this message because you are subscribed to the Google Groups
> "opencog" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to opencog+unsubscr...@googlegroups.com.
> To post to this group, send email to opencog@googlegroups.com.
> Visit this group at https://groups.google.com/group/opencog.
> To view this discussion on the web visit https://groups.google.com/d/
> msgid/opencog/85ed2664-5369-41f7-aa2a-5bea845a8f00%40googlegroups.com
> 
> .
> For more options, visit https://groups.google.com/d/optout.
>



-- 
*"The problem is not that artificial intelligence will get too smart and
take over the world," computer scientist Pedro Domingos writes, "the
problem is that it's too stupid and already has." *

-- 
You received this message because you are subscribed to the Google Groups 
"opencog" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to opencog+unsubscr...@googlegroups.com.
To post to this group, send email to opencog@googlegroups.com.
Visit this group at https://groups.google.com/group/opencog.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/opencog/CAHrUA35xb2%3DC0TDZGsfYTM3LH0Yps1JG4F_C%2B14gcfZ5TxxO%2Bg%40mail.gmail.com.
For more options, visit https://groups.google.com/d/optout.

[opencog-dev] Storytelling

[wtaylorjr2...@gmail.com]

Is it possible to use opencog to input stories and break them apart, 
distort them, and then assemble them in new ways to create "new" stories?

-- 
You received this message because you are subscribed to the Google Groups 
"opencog" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to opencog+unsubscr...@googlegroups.com.
To post to this group, send email to opencog@googlegroups.com.
Visit this group at https://groups.google.com/group/opencog.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/opencog/85ed2664-5369-41f7-aa2a-5bea845a8f00%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.