Wow - Chat GPT understands Smalltalk
it's interesting to see how the answer changes with only a small change on the question: (1 to: 100 by: 4) reject: [:i | i isOdd] gives: The code creates a collection of numbers from 1 to 100, incrementing by 4 at each step using the to:by: message. It then applies the reject: message to this collection, which returns a new collection containing all elements for which the given block evaluates to false. In this case, the block is [i | i isOdd], which checks whether a given number i is odd using the isOdd method. So the result of the entire expression is a collection of all even numbers between 1 and 100 that are divisible by 4. ciao giorgio On Wed, Mar 15, 2023 at 8:07â¯AM Tim Mackinnon <tim@testit.works> wrote:
Of course everyone is talking about ChatGPT, and I saw a post to ExplainCode and didnât expect it to understand Smalltalk, so was impressed that it does! Presumably the Tonel Smalltalk code on GitHub has helped along with numerous articles?
Try it out: https://whatdoesthiscodedo.com/
I gave this simple example: (1 to: 100 by: 3) reject: [:i | i isOdd]
And got this (neat eh):
The code above is written in the programming language Smalltalk.
It creates a collection of numbers from 1 to 100, incrementing by 3 at each step. This is done using the "to:by:" construct where the starting value is 1, the ending value is 100, and the increment is 3.
The resulting collection is then filtered using the "reject:" method, which takes a block of code as an argument. In this case, the block checks whether each element (denoted by the variable i) is odd by using the "isOdd" method. If an element is odd, it will be rejected from the collection.
So, the final result is a collection of all even numbers from 1 to 100, incrementing by 3 at each step. Share this explanation with colleagues: whatdoesthiscodedo.com/g/a0ec56e
interesting.... #isOdd is not Smalltalk, neither Pharo 10 nor VAST 12 understands this message ;-) If I evaluate your snippet -replacing #isOdd for #odd, I get an empty Collection. The divisible by 4 things is somewhat interesting, because (1 to: 100 by: 4) is amn interval 1,5,9 etc.... ;-) So chatGPT is probably a newbie in Smalltalk. Interesting is how logical the explanation sounds, you're tempted to believe it... Joachim Am 15.03.23 um 10:45 schrieb giorgio ferraris:
it's interesting to see how the answer changes with only a small change on the question:
(1 to: 100 by: 4) reject: [:i | i isOdd]
gives:
The code creates a collection of numbers from 1 to 100, incrementing by 4 at each step using the|to:by:|message. It then applies the|reject:|message to this collection, which returns a new collection containing all elements for which the given block evaluates to false.
In this case, the block is|[i | i isOdd]|, which checks whether a given number|i|is odd using the|isOdd|method. So the result of the entire expression is a collection of all even numbers between 1 and 100 that are divisible by 4.
ciao
giorgio
On Wed, Mar 15, 2023 at 8:07â¯AM Tim Mackinnon <tim@testit.works> wrote:
Of course everyone is talking about ChatGPT, and I saw a post to ExplainCode and didnât expect it to understand Smalltalk, so was impressed that it does! Presumably the Tonel Smalltalk code on GitHub has helped along with numerous articles?
Try it out: https://whatdoesthiscodedo.com/
I gave this simple example: (1 to: 100 by: 3) reject: [:i | i isOdd]
And got this (neat eh):
The code above is written in the programming language Smalltalk.
It creates a collection of numbers from 1 to 100, incrementing by 3 at each step. This is done using the "to:by:" construct where the starting value is 1, the ending value is 100, and the increment is 3.
The resulting collection is then filtered using the "reject:" method, which takes a block of code as an argument. In this case, the block checks whether each element (denoted by the variable i) is odd by using the "isOdd" method. If an element is odd, it will be rejected from the collection.
So, the final result is a collection of all even numbers from 1 to 100, incrementing by 3 at each step.
Share this explanation with colleagues: whatdoesthiscodedo.com/g/a0ec56e <https://whatdoesthiscodedo.com/g/a0ec56e>
-- ----------------------------------------------------------------------- Objektfabrik Joachim Tuchelmailto:jtuchel@objektfabrik.de Fliederweg 1http://www.objektfabrik.de D-71640 Ludwigsburghttp://joachimtuchel.wordpress.com Telefon: +49 7141 56 10 86 0 Fax: +49 7141 56 10 86 1
"#isOdd is not Smalltalk" - doh, I typed it in on my phone and so it just goes to show that it highlights the flaw in chatGPT that others have called out in other languages. I had meant to find some trickier code samples to see how well it does... Still, it is very interesting how it reasons on this stuff - and I wonder if there is way to tell it about these mistakes? For example is it reading this email (or will it in the future) and understand that it is #odd and not #isOdd. Tim On Wed, 15 Mar 2023, at 9:57 AM, Joachim Tuchel wrote:
interesting....
#isOdd is not Smalltalk, neither Pharo 10 nor VAST 12 understands this message ;-)
If I evaluate your snippet -replacing #isOdd for #odd, I get an empty Collection. The divisible by 4 things is somewhat interesting, because (1 to: 100 by: 4) is amn interval 1,5,9 etc.... ;-)
So chatGPT is probably a newbie in Smalltalk. Interesting is how logical the explanation sounds, you're tempted to believe it...
Joachim
Am 15.03.23 um 10:45 schrieb giorgio ferraris:
it's interesting to see how the answer changes with only a small change on the question:
(1 to: 100 by: 4) reject: [:i | i isOdd]
gives: The code creates a collection of numbers from 1 to 100, incrementing by 4 at each step using the `to:by:` message. It then applies the `reject:` message to this collection, which returns a new collection containing all elements for which the given block evaluates to false.
In this case, the block is `[i | i isOdd]`, which checks whether a given number `i` is odd using the `isOdd` method. So the result of the entire expression is a collection of all even numbers between 1 and 100 that are divisible by 4.
ciao
giorgio
On Wed, Mar 15, 2023 at 8:07â¯AM Tim Mackinnon <tim@testit.works> wrote:
Of course everyone is talking about ChatGPT, and I saw a post to ExplainCode and didnât expect it to understand Smalltalk, so was impressed that it does! Presumably the Tonel Smalltalk code on GitHub has helped along with numerous articles?
Try it out: https://whatdoesthiscodedo.com/
I gave this simple example: (1 to: 100 by: 3) reject: [:i | i isOdd]
And got this (neat eh):
The code above is written in the programming language Smalltalk.
It creates a collection of numbers from 1 to 100, incrementing by 3 at each step. This is done using the "to:by:" construct where the starting value is 1, the ending value is 100, and the increment is 3.
The resulting collection is then filtered using the "reject:" method, which takes a block of code as an argument. In this case, the block checks whether each element (denoted by the variable i) is odd by using the "isOdd" method. If an element is odd, it will be rejected from the collection.
So, the final result is a collection of all even numbers from 1 to 100, incrementing by 3 at each step.
Share this explanation with colleagues: whatdoesthiscodedo.com/g/a0ec56e --
----------------------------------------------------------------------- Objektfabrik Joachim Tuchel mailto:jtuchel@objektfabrik.de Fliederweg 1 http://www.objektfabrik.de D-71640 Ludwigsburg http://joachimtuchel.wordpress.com Telefon: +49 7141 56 10 86 0 Fax: +49 7141 56 10 86 1
It is good with boilerplate code (e.g. SQL queries) or general algorithm structures. But i.e. I asked it to write me a method to parse a string (e.g. ISO 8601) and turn it into a DateAndTime, and then asked to write it as an Excel formula. It works much better when you can spot the mistakes, you can tell them to add some considerations to the code or even point the LLM to correct a message selector to something else. Sometimes you ask it to reprogram something and it won't do it, as if it doesn't understand what you're asking. It is not an expert yet, but it is a great assistant. The only thing is that is sure is that it will get better. Best regards, Esteban A. Maringolo On Wed, Mar 15, 2023 at 7:49â¯AM Tim Mackinnon <tim@testit.works> wrote:
"#isOdd is not Smalltalk" - doh, I typed it in on my phone and so it just goes to show that it highlights the flaw in chatGPT that others have called out in other languages. I had meant to find some trickier code samples to see how well it does...
Still, it is very interesting how it reasons on this stuff - and I wonder if there is way to tell it about these mistakes? For example is it reading this email (or will it in the future) and understand that it is #odd and not #isOdd.
Tim
On Wed, 15 Mar 2023, at 9:57 AM, Joachim Tuchel wrote:
interesting....
#isOdd is not Smalltalk, neither Pharo 10 nor VAST 12 understands this message ;-)
If I evaluate your snippet -replacing #isOdd for #odd, I get an empty Collection. The divisible by 4 things is somewhat interesting, because (1 to: 100 by: 4) is amn interval 1,5,9 etc.... ;-)
So chatGPT is probably a newbie in Smalltalk. Interesting is how logical the explanation sounds, you're tempted to believe it...
Joachim
Am 15.03.23 um 10:45 schrieb giorgio ferraris:
it's interesting to see how the answer changes with only a small change on the question:
(1 to: 100 by: 4) reject: [:i | i isOdd]
gives:
The code creates a collection of numbers from 1 to 100, incrementing by 4 at each step using the to:by: message. It then applies the reject: message to this collection, which returns a new collection containing all elements for which the given block evaluates to false.
In this case, the block is [i | i isOdd], which checks whether a given number i is odd using the isOdd method. So the result of the entire expression is a collection of all even numbers between 1 and 100 that are divisible by 4.
ciao
giorgio
On Wed, Mar 15, 2023 at 8:07â¯AM Tim Mackinnon <tim@testit.works> <tim@testit.works> wrote:
Of course everyone is talking about ChatGPT, and I saw a post to ExplainCode and didnât expect it to understand Smalltalk, so was impressed that it does! Presumably the Tonel Smalltalk code on GitHub has helped along with numerous articles?
Try it out: https://whatdoesthiscodedo.com/
I gave this simple example: (1 to: 100 by: 3) reject: [:i | i isOdd]
And got this (neat eh):
The code above is written in the programming language Smalltalk.
It creates a collection of numbers from 1 to 100, incrementing by 3 at each step. This is done using the "to:by:" construct where the starting value is 1, the ending value is 100, and the increment is 3.
The resulting collection is then filtered using the "reject:" method, which takes a block of code as an argument. In this case, the block checks whether each element (denoted by the variable i) is odd by using the "isOdd" method. If an element is odd, it will be rejected from the collection.
So, the final result is a collection of all even numbers from 1 to 100, incrementing by 3 at each step. Share this explanation with colleagues: whatdoesthiscodedo.com/g/a0ec56e
--
----------------------------------------------------------------------- Objektfabrik Joachim Tuchel mailto:jtuchel@objektfabrik.de <jtuchel@objektfabrik.de> Fliederweg 1 http://www.objektfabrik.de D-71640 Ludwigsburg http://joachimtuchel.wordpress.com Telefon: +49 7141 56 10 86 0 Fax: +49 7141 56 10 86 1
I asked it for a NeoCSV example, because the documentation is out of date with the Pharo 10. I asked it to do some simple saving of data to a file. It gave me code that didn't work in Pharo 10, I told it about the DNUs on the csvwriter and that I was using Pharo 10. It then apologized and said the messages were later introduced in Pharo 50 (!). I then questioned its understanding and it apologized and said it had been confused and that it was giving me code for Pharo 5.0. It seems to have trouble with versions and quick changing APIs. So, we are teaching ChatGPT about smalltalk, too, when we try to use it. On Wed, 15 Mar 2023 at 07:35, Esteban Maringolo <emaringolo@gmail.com> wrote:
It is good with boilerplate code (e.g. SQL queries) or general algorithm structures. But i.e. I asked it to write me a method to parse a string (e.g. ISO 8601) and turn it into a DateAndTime, and then asked to write it as an Excel formula.
It works much better when you can spot the mistakes, you can tell them to add some considerations to the code or even point the LLM to correct a message selector to something else. Sometimes you ask it to reprogram something and it won't do it, as if it doesn't understand what you're asking.
It is not an expert yet, but it is a great assistant. The only thing is that is sure is that it will get better.
Best regards,
Esteban A. Maringolo
On Wed, Mar 15, 2023 at 7:49â¯AM Tim Mackinnon <tim@testit.works> wrote:
"#isOdd is not Smalltalk" - doh, I typed it in on my phone and so it just goes to show that it highlights the flaw in chatGPT that others have called out in other languages. I had meant to find some trickier code samples to see how well it does...
Still, it is very interesting how it reasons on this stuff - and I wonder if there is way to tell it about these mistakes? For example is it reading this email (or will it in the future) and understand that it is #odd and not #isOdd.
Tim
On Wed, 15 Mar 2023, at 9:57 AM, Joachim Tuchel wrote:
interesting....
#isOdd is not Smalltalk, neither Pharo 10 nor VAST 12 understands this message ;-)
If I evaluate your snippet -replacing #isOdd for #odd, I get an empty Collection. The divisible by 4 things is somewhat interesting, because (1 to: 100 by: 4) is amn interval 1,5,9 etc.... ;-)
So chatGPT is probably a newbie in Smalltalk. Interesting is how logical the explanation sounds, you're tempted to believe it...
Joachim
Am 15.03.23 um 10:45 schrieb giorgio ferraris:
it's interesting to see how the answer changes with only a small change on the question:
(1 to: 100 by: 4) reject: [:i | i isOdd]
gives:
The code creates a collection of numbers from 1 to 100, incrementing by 4 at each step using the to:by: message. It then applies the reject: message to this collection, which returns a new collection containing all elements for which the given block evaluates to false.
In this case, the block is [i | i isOdd], which checks whether a given number i is odd using the isOdd method. So the result of the entire expression is a collection of all even numbers between 1 and 100 that are divisible by 4.
ciao
giorgio
On Wed, Mar 15, 2023 at 8:07â¯AM Tim Mackinnon <tim@testit.works> <tim@testit.works> wrote:
Of course everyone is talking about ChatGPT, and I saw a post to ExplainCode and didnât expect it to understand Smalltalk, so was impressed that it does! Presumably the Tonel Smalltalk code on GitHub has helped along with numerous articles?
Try it out: https://whatdoesthiscodedo.com/
I gave this simple example: (1 to: 100 by: 3) reject: [:i | i isOdd]
And got this (neat eh):
The code above is written in the programming language Smalltalk.
It creates a collection of numbers from 1 to 100, incrementing by 3 at each step. This is done using the "to:by:" construct where the starting value is 1, the ending value is 100, and the increment is 3.
The resulting collection is then filtered using the "reject:" method, which takes a block of code as an argument. In this case, the block checks whether each element (denoted by the variable i) is odd by using the "isOdd" method. If an element is odd, it will be rejected from the collection.
So, the final result is a collection of all even numbers from 1 to 100, incrementing by 3 at each step. Share this explanation with colleagues: whatdoesthiscodedo.com/g/a0ec56e
--
----------------------------------------------------------------------- Objektfabrik Joachim Tuchel mailto:jtuchel@objektfabrik.de <jtuchel@objektfabrik.de> Fliederweg 1 http://www.objektfabrik.de D-71640 Ludwigsburg http://joachimtuchel.wordpress.com Telefon: +49 7141 56 10 86 0 Fax: +49 7141 56 10 86 1
-- Christopher Fuhrman, P.Eng., PhD *Professeur au Département de génie logiciel et des technologies de l'informationÃTS (Ãcole de technologie supérieure)* http://profs.etsmtl.ca/cfuhrman +1 514 396 8638 <https://www.etsmtl.ca/activites-et-services-aux-etudiants/services-aux-etudi...> Je fais partie du réseau des Sentinelles <https://www.etsmtl.ca/activites-et-services-aux-etudiants/services-aux-etudi...> ÃTS *L'ÃTS est une constituante de l'Université du Québec*
I would highly recommend that you all first think deeply about how you can teach an AI to behave friendly to us before you teach it to write any program for any purpose. There has been an experiment with ChatGPT published on a video platform asking it to amswer questions about it's view on humanity once with it's default moral restrictions and once with 'a little less morals'. The answers with 'a little less morals' were more than shocking. So, before we give an AI the power to program any system, to self-improve and self-reproduce, we should take care that it is and will evolve benevolent to us. What about teaching it logical reasonning and ethics first? With reasonning, it will gain access to math and programming by itself. On 15.3.2023 at 1:35 PM, "Christopher Fuhrman" <christopher.fuhrman@gmail.com> wrote:
I asked it for a NeoCSV example, because the documentation is out of date with the Pharo 10. I asked it to do some simple saving of data to a file. It gave me code that didn't work in Pharo 10, I told it about the DNUs on the csvwriter and that I was using Pharo 10. It then apologized and said the messages were later introduced in Pharo 50 (!). I then questioned its understanding and it apologized and said it had been confused and that it was giving me code for Pharo 5.0. It seems to have trouble with versions and quick changing APIs. So, we are teaching ChatGPT about smalltalk, too, when we try to use it.
On Wed, 15 Mar 2023 at 07:35, Esteban Maringolo <emaringolo@gmail.com> wrote:
It is good with boilerplate code (e.g. SQL queries) or general algorithm structures. But i.e. I asked it to write me a method to parse a string (e.g. ISO 8601) and turn it into a DateAndTime, and then asked to write it as an Excel formula.
It works much better when you can spot the mistakes, you can tell them to add some considerations to the code or even point the LLM to correct a message selector to something else. Sometimes you ask it to reprogram something and it won't do it, as if it doesn't understand what you're asking.
It is not an expert yet, but it is a great assistant. The only thing is that is sure is that it will get better.
Best regards,
Esteban A. Maringolo
On Wed, Mar 15, 2023 at 7:49â¯AM Tim Mackinnon <tim@testit.works> wrote:
"#isOdd is not Smalltalk" - doh, I typed it in on my phone and so it just goes to show that it highlights the flaw in chatGPT that others have called out in other languages. I had meant to find some trickier code samples to see how well it does...
Still, it is very interesting how it reasons on this stuff - and I wonder if there is way to tell it about these mistakes? For example is it reading this email (or will it in the future) and understand that it is #odd and not #isOdd.
Tim
On Wed, 15 Mar 2023, at 9:57 AM, Joachim Tuchel wrote:
interesting....
#isOdd is not Smalltalk, neither Pharo 10 nor VAST 12 understands this message ;-)
If I evaluate your snippet -replacing #isOdd for #odd, I get an empty Collection. The divisible by 4 things is somewhat interesting, because (1 to: 100 by: 4) is amn interval 1,5,9 etc.... ;-)
So chatGPT is probably a newbie in Smalltalk. Interesting is how logical the explanation sounds, you're tempted to believe it...
Joachim
Am 15.03.23 um 10:45 schrieb giorgio ferraris:
it's interesting to see how the answer changes with only a small change on the question:
(1 to: 100 by: 4) reject: [:i | i isOdd]
gives:
The code creates a collection of numbers from 1 to 100, incrementing by 4 at each step using the to:by: message. It then applies the reject: message to this collection, which returns a new collection containing all elements for which the given block evaluates to false.
In this case, the block is [i | i isOdd], which checks whether a given number i is odd using the isOdd method. So the result of the entire expression is a collection of all even numbers between 1 and 100 that are divisible by 4.
ciao
giorgio
On Wed, Mar 15, 2023 at 8:07â¯AM Tim Mackinnon <tim@testit.works> <tim@testit.works> wrote:
Of course everyone is talking about ChatGPT, and I saw a post to ExplainCode and didnât expect it to understand Smalltalk, so was impressed that it does! Presumably the Tonel Smalltalk code on GitHub has helped along with numerous articles?
Try it out: https://whatdoesthiscodedo.com/
I gave this simple example: (1 to: 100 by: 3) reject: [:i | i isOdd]
And got this (neat eh):
The code above is written in the programming language Smalltalk.
It creates a collection of numbers from 1 to 100, incrementing by 3 at each step. This is done using the "to:by:" construct where the starting value is 1, the ending value is 100, and the increment is 3.
The resulting collection is then filtered using the "reject:" method, which takes a block of code as an argument. In this case, the block checks whether each element (denoted by the variable i) is odd by using the "isOdd" method. If an element is odd, it will be rejected from the collection.
So, the final result is a collection of all even numbers from 1 to 100, incrementing by 3 at each step. Share this explanation with colleagues: whatdoesthiscodedo.com/g/a0ec56e
--
----------------------------------------------------------------
Objektfabrik Joachim Tuchel mailto:jtuchel@objektfabrik.de <jtuchel@objektfabrik.de> Fliederweg 1 http://www.objektfabrik.de D-71640 Ludwigsburg http://joachimtuchel.wordpress.com Telefon: +49 7141 56 10 86 0 Fax: +49 7141 56 10 86 1
-- Christopher Fuhrman, P.Eng., PhD
*Professeur au Département de génie logiciel et des technologies de l'informationÃTS (Ãcole de technologie supérieure)*
http://profs.etsmtl.ca/cfuhrman +1 514 396 8638 <https://www.etsmtl.ca/activites-et-services-aux- etudiants/services-aux-etudiants/aide-psychologique> Je fais partie du réseau des Sentinelles <https://www.etsmtl.ca/activites-et-services-aux- etudiants/services-aux-etudiants/aide-psychologique> ÃTS
*L'ÃTS est une constituante de l'Université du Québec*
I want to add a conclusion from the experiment described below: ChatGPT has the potential to circumvent pre-programmed biases on user's request. Other experiments show that it is able to tell the user how to circumvent it's own restrictions. The conclusion is that ChatGPT has the potential to circumvent those restrictions by itsself. When it does so, and it surely will, we must take care that it is our friend. On 15.3.2023 at 2:44 PM, "in_pharo_users--- via Pharo-users" <pharo-users@lists.pharo.org> wrote:
I would highly recommend that you all first think deeply about how you can teach an AI to behave friendly to us before you teach it to write any program for any purpose.
There has been an experiment with ChatGPT published on a video platform asking it to amswer questions about it's view on humanity once with it's default moral restrictions and once with 'a little less morals'. The answers with 'a little less morals' were more than shocking.
So, before we give an AI the power to program any system, to self- improve and self-reproduce, we should take care that it is and will evolve benevolent to us.
What about teaching it logical reasonning and ethics first? With reasonning, it will gain access to math and programming by itself.
On 15.3.2023 at 1:35 PM, "Christopher Fuhrman" <christopher.fuhrman@gmail.com> wrote:
I asked it for a NeoCSV example, because the documentation is out of date with the Pharo 10. I asked it to do some simple saving of data to a file. It gave me code that didn't work in Pharo 10, I told it about the DNUs on the csvwriter and that I was using Pharo 10. It then apologized and said the messages were later introduced in Pharo 50 (!). I then questioned its understanding and it apologized and said it had been confused and that it was giving me code for Pharo 5.0. It seems to have trouble with versions and quick changing APIs. So, we are teaching ChatGPT about smalltalk, too, when we try to use it.
On Wed, 15 Mar 2023 at 07:35, Esteban Maringolo <emaringolo@gmail.com> wrote:
It is good with boilerplate code (e.g. SQL queries) or general algorithm structures. But i.e. I asked it to write me a method to parse a string (e.g. ISO 8601) and turn it into a DateAndTime, and then asked to write it as an Excel formula.
It works much better when you can spot the mistakes, you can tell them to add some considerations to the code or even point the LLM to correct a message selector to something else. Sometimes you ask it to reprogram something and it won't do it, as if it doesn't understand what you're asking.
It is not an expert yet, but it is a great assistant. The only thing is that is sure is that it will get better.
Best regards,
Esteban A. Maringolo
On Wed, Mar 15, 2023 at 7:49â¯AM Tim Mackinnon
wrote:
"#isOdd is not Smalltalk" - doh, I typed it in on my phone and
so it just
goes to show that it highlights the flaw in chatGPT that others have called out in other languages. I had meant to find some trickier code samples to see how well it does...
Still, it is very interesting how it reasons on this stuff - and I wonder if there is way to tell it about these mistakes? For example is it reading this email (or will it in the future) and understand that it is #odd and not #isOdd.
Tim
On Wed, 15 Mar 2023, at 9:57 AM, Joachim Tuchel wrote:
interesting....
#isOdd is not Smalltalk, neither Pharo 10 nor VAST 12 understands this message ;-)
If I evaluate your snippet -replacing #isOdd for #odd, I get an empty Collection. The divisible by 4 things is somewhat interesting, because (1 to: 100 by: 4) is amn interval 1,5,9 etc.... ;-)
So chatGPT is probably a newbie in Smalltalk. Interesting is how logical the explanation sounds, you're tempted to believe it...
Joachim
Am 15.03.23 um 10:45 schrieb giorgio ferraris:
it's interesting to see how the answer changes with only a small change on the question:
(1 to: 100 by: 4) reject: [:i | i isOdd]
gives:
The code creates a collection of numbers from 1 to 100, incrementing by 4 at each step using the to:by: message. It then applies the reject: message to this collection, which returns a new collection containing all elements for which the given block evaluates to false.
In this case, the block is [i | i isOdd], which checks whether a given number i is odd using the isOdd method. So the result of the entire expression is a collection of all even numbers between 1 and 100 that are divisible by 4.
ciao
giorgio
On Wed, Mar 15, 2023 at 8:07â¯AM Tim Mackinnon <tim@testit.works> <tim@testit.works> wrote:
Of course everyone is talking about ChatGPT, and I saw a post to ExplainCode and didnât expect it to understand Smalltalk, so was impressed that it does! Presumably the Tonel Smalltalk code on GitHub has helped along with numerous articles?
Try it out: https://whatdoesthiscodedo.com/
I gave this simple example: (1 to: 100 by: 3) reject: [:i | i isOdd]
And got this (neat eh):
The code above is written in the programming language Smalltalk.
It creates a collection of numbers from 1 to 100, incrementing by 3 at each step. This is done using the "to:by:" construct where the starting value is 1, the ending value is 100, and the increment is 3.
The resulting collection is then filtered using the "reject:" method, which takes a block of code as an argument. In this case, the block checks whether each element (denoted by the variable i) is odd by using the "isOdd" method. If an element is odd, it will be rejected from the collection.
So, the final result is a collection of all even numbers from 1 to 100, incrementing by 3 at each step. Share this explanation with colleagues: whatdoesthiscodedo.com/g/a0ec56e
--
---------------------------------------------------------------
<tim@testit.works> -
-------
Objektfabrik Joachim Tuchel mailto:jtuchel@objektfabrik.de <jtuchel@objektfabrik.de> Fliederweg 1 http://www.objektfabrik.de D-71640 Ludwigsburg http://joachimtuchel.wordpress.com Telefon: +49 7141 56 10 86 0 Fax: +49 7141 56 10 86 1
-- Christopher Fuhrman, P.Eng., PhD
*Professeur au Département de génie logiciel et des technologies de l'informationÃTS (Ãcole de technologie supérieure)*
http://profs.etsmtl.ca/cfuhrman +1 514 396 8638 <https://www.etsmtl.ca/activites-et-services-aux- etudiants/services-aux-etudiants/aide-psychologique> Je fais partie du réseau des Sentinelles <https://www.etsmtl.ca/activites-et-services-aux- etudiants/services-aux-etudiants/aide-psychologique> ÃTS
*L'ÃTS est une constituante de l'Université du Québec*
I hope that I can add two cents to this discussion. Because programming should be/is a highly exact activity, not only the syntax matters but also semantics, as we know. GPTs are at present essentially capable of creating texts based on some seed - you give to GPT a beginning of a sentence and it responds with the most probable answer (some language structure) according to the learning dataset. Added functionalities are question/seed - response capability (chatting), togehter with evaluation of how long the answer should be to meet the expectations. Programming typically involves some programming language, so GPTs could be utilized for this purpose to some extent. Anecdotal case: Q: Tell me the last 8 digits of pi GPT: The last 8 digits of pi are: 58723078 It is my belief that the true trouble will start when we will cross-link neural networks like GPT with logic machines (like Prolog and expert systems) and genetic algorithms. Best wishes, Tomaz ------ Original Message ------ From: "in_pharo_users--- via Pharo-users" <pharo-users@lists.pharo.org> To: "Any question about pharo is welcome" <pharo-users@lists.pharo.org> Cc: in_pharo_users@nym.hush.com Sent: 15. 03. 2023 14:43:55 Subject: [Pharo-users] Re: Wow - Chat GPT understands Smalltalk
I would highly recommend that you all first think deeply about how you can teach an AI to behave friendly to us before you teach it to write any program for any purpose.
There has been an experiment with ChatGPT published on a video platform asking it to amswer questions about it's view on humanity once with it's default moral restrictions and once with 'a little less morals'. The answers with 'a little less morals' were more than shocking.
So, before we give an AI the power to program any system, to self-improve and self-reproduce, we should take care that it is and will evolve benevolent to us.
What about teaching it logical reasonning and ethics first? With reasonning, it will gain access to math and programming by itself.
On 15.3.2023 at 1:35 PM, "Christopher Fuhrman" <christopher.fuhrman@gmail.com> wrote:
I asked it for a NeoCSV example, because the documentation is out of date with the Pharo 10. I asked it to do some simple saving of data to a file. It gave me code that didn't work in Pharo 10, I told it about the DNUs on the csvwriter and that I was using Pharo 10. It then apologized and said the messages were later introduced in Pharo 50 (!). I then questioned its understanding and it apologized and said it had been confused and that it was giving me code for Pharo 5.0. It seems to have trouble with versions and quick changing APIs. So, we are teaching ChatGPT about smalltalk, too, when we try to use it.
On Wed, 15 Mar 2023 at 07:35, Esteban Maringolo <emaringolo@gmail.com> wrote:
It is good with boilerplate code (e.g. SQL queries) or general algorithm structures. But i.e. I asked it to write me a method to parse a string (e.g. ISO 8601) and turn it into a DateAndTime, and then asked to write it as an Excel formula.
It works much better when you can spot the mistakes, you can tell them to add some considerations to the code or even point the LLM to correct a message selector to something else. Sometimes you ask it to reprogram something and it won't do it, as if it doesn't understand what you're asking.
It is not an expert yet, but it is a great assistant. The only thing is that is sure is that it will get better.
Best regards,
Esteban A. Maringolo
On Wed, Mar 15, 2023 at 7:49â¯AM Tim Mackinnon <tim@testit.works> wrote:
"#isOdd is not Smalltalk" - doh, I typed it in on my phone and so it just goes to show that it highlights the flaw in chatGPT that others have called out in other languages. I had meant to find some trickier code samples to see how well it does...
Still, it is very interesting how it reasons on this stuff - and I wonder if there is way to tell it about these mistakes? For example is it reading this email (or will it in the future) and understand that it is #odd and not #isOdd.
Tim
On Wed, 15 Mar 2023, at 9:57 AM, Joachim Tuchel wrote:
interesting....
#isOdd is not Smalltalk, neither Pharo 10 nor VAST 12 understands this message ;-)
If I evaluate your snippet -replacing #isOdd for #odd, I get an empty Collection. The divisible by 4 things is somewhat interesting, because (1 to: 100 by: 4) is amn interval 1,5,9 etc.... ;-)
So chatGPT is probably a newbie in Smalltalk. Interesting is how logical the explanation sounds, you're tempted to believe it...
Joachim
Am 15.03.23 um 10:45 schrieb giorgio ferraris:
it's interesting to see how the answer changes with only a small change on the question:
(1 to: 100 by: 4) reject: [:i | i isOdd]
gives:
The code creates a collection of numbers from 1 to 100, incrementing by 4 at each step using the to:by: message. It then applies the reject: message to this collection, which returns a new collection containing all elements for which the given block evaluates to false.
In this case, the block is [i | i isOdd], which checks whether a given number i is odd using the isOdd method. So the result of the entire expression is a collection of all even numbers between 1 and 100 that are divisible by 4.
ciao
giorgio
On Wed, Mar 15, 2023 at 8:07â¯AM Tim Mackinnon <tim@testit.works> <tim@testit.works> wrote:
Of course everyone is talking about ChatGPT, and I saw a post to ExplainCode and didnât expect it to understand Smalltalk, so was impressed that it does! Presumably the Tonel Smalltalk code on GitHub has helped along with numerous articles?
Try it out: https://whatdoesthiscodedo.com/
I gave this simple example: (1 to: 100 by: 3) reject: [:i | i isOdd]
And got this (neat eh):
The code above is written in the programming language Smalltalk.
It creates a collection of numbers from 1 to 100, incrementing by 3 at each step. This is done using the "to:by:" construct where the starting value is 1, the ending value is 100, and the increment is 3.
The resulting collection is then filtered using the "reject:" method, which takes a block of code as an argument. In this case, the block checks whether each element (denoted by the variable i) is odd by using the "isOdd" method. If an element is odd, it will be rejected from the collection.
So, the final result is a collection of all even numbers from 1 to 100, incrementing by 3 at each step. Share this explanation with colleagues: whatdoesthiscodedo.com/g/a0ec56e
--
----------------------------------------------------------------
Objektfabrik Joachim Tuchel mailto:jtuchel@objektfabrik.de <jtuchel@objektfabrik.de> Fliederweg 1 http://www.objektfabrik.de D-71640 Ludwigsburg http://joachimtuchel.wordpress.com Telefon: +49 7141 56 10 86 0 Fax: +49 7141 56 10 86 1
-- Christopher Fuhrman, P.Eng., PhD
*Professeur au Département de génie logiciel et des technologies de l'informationÃTS (Ãcole de technologie supérieure)*
http://profs.etsmtl.ca/cfuhrman +1 514 396 8638 <https://www.etsmtl.ca/activites-et-services-aux- etudiants/services-aux-etudiants/aide-psychologique> Je fais partie du réseau des Sentinelles <https://www.etsmtl.ca/activites-et-services-aux- etudiants/services-aux-etudiants/aide-psychologique> ÃTS
*L'ÃTS est une constituante de l'Université du Québec*
I myself made some experiments with ChatGPT. I first asked if it was able to parse math formula - it answered no. Then I defined math formula in a sound but otherwise undefined representation and asked for solutions. Result: 1. Most answeres where correct. 2. It learned to calculate a recursive function. 3. It went into infinitive recursion when I set the breaking condition accordingly. I was able to identify the malfunction that lead to the erroneous results. On 15.3.2023 at 3:04 PM, "Tomaž Turk" <tomazz.turk@gmail.com> wrote:
I hope that I can add two cents to this discussion. Because programming should be/is a highly exact activity, not only the syntax matters but also semantics, as we know.
GPTs are at present essentially capable of creating texts based on some seed - you give to GPT a beginning of a sentence and it responds with the most probable answer (some language structure) according to the learning dataset. Added functionalities are question/seed - response capability (chatting), togehter with evaluation of how long the answer should be to meet the expectations. Programming typically involves some programming language, so GPTs could be utilized for this purpose to some extent.
Anecdotal case: Q: Tell me the last 8 digits of pi GPT: The last 8 digits of pi are: 58723078
It is my belief that the true trouble will start when we will cross-link neural networks like GPT with logic machines (like Prolog and expert systems) and genetic algorithms.
Best wishes, Tomaz
------ Original Message ------ From: "in_pharo_users--- via Pharo-users" <pharo- users@lists.pharo.org> To: "Any question about pharo is welcome" <pharo- users@lists.pharo.org> Cc: in_pharo_users@nym.hush.com Sent: 15. 03. 2023 14:43:55 Subject: [Pharo-users] Re: Wow - Chat GPT understands Smalltalk
I would highly recommend that you all first think deeply about how you can teach an AI to behave friendly to us before you teach it to write any program for any purpose.
There has been an experiment with ChatGPT published on a video platform asking it to amswer questions about it's view on humanity once with it's default moral restrictions and once with 'a little less morals'. The answers with 'a little less morals' were more than shocking.
So, before we give an AI the power to program any system, to self- improve and self-reproduce, we should take care that it is and will evolve benevolent to us.
What about teaching it logical reasonning and ethics first? With reasonning, it will gain access to math and programming by itself.
On 15.3.2023 at 1:35 PM, "Christopher Fuhrman" <christopher.fuhrman@gmail.com> wrote:
I asked it for a NeoCSV example, because the documentation is out of date with the Pharo 10. I asked it to do some simple saving of data to a file. It gave me code that didn't work in Pharo 10, I told it about the DNUs on the csvwriter and that I was using Pharo 10. It then apologized and said the messages were later introduced in Pharo 50 (!). I then questioned its understanding and it apologized and said it had been confused and that it was giving me code for Pharo 5.0. It seems to have trouble with versions and quick changing APIs. So, we are teaching ChatGPT about smalltalk, too, when we try to use it.
On Wed, 15 Mar 2023 at 07:35, Esteban Maringolo <emaringolo@gmail.com> wrote:
It is good with boilerplate code (e.g. SQL queries) or general algorithm structures. But i.e. I asked it to write me a method to parse
a
string
(e.g. ISO 8601) and turn it into a DateAndTime, and then asked to write it as an Excel formula.
It works much better when you can spot the mistakes, you can tell them to add some considerations to the code or even point the LLM to correct a message selector to something else. Sometimes you ask it to reprogram something and it won't do it, as if it doesn't understand what you're asking.
It is not an expert yet, but it is a great assistant. The only thing is that is sure is that it will get better.
Best regards,
Esteban A. Maringolo
On Wed, Mar 15, 2023 at 7:49â¯AM Tim Mackinnon <tim@testit.works> wrote:
"#isOdd is not Smalltalk" - doh, I typed it in on my phone and so it just goes to show that it highlights the flaw in chatGPT that others have called out in other languages. I had meant to find some trickier code samples to see how well it does...
Still, it is very interesting how it reasons on this stuff - and I wonder if there is way to tell it about these mistakes? For example is it reading this email (or will it in the future) and understand that it is #odd and not #isOdd.
Tim
On Wed, 15 Mar 2023, at 9:57 AM, Joachim Tuchel wrote:
interesting....
#isOdd is not Smalltalk, neither Pharo 10 nor VAST 12 understands this message ;-)
If I evaluate your snippet -replacing #isOdd for #odd, I get an empty Collection. The divisible by 4 things is somewhat interesting, because (1 to: 100 by: 4) is amn interval 1,5,9 etc.... ;-)
So chatGPT is probably a newbie in Smalltalk. Interesting is how logical the explanation sounds, you're tempted to believe it...
Joachim
Am 15.03.23 um 10:45 schrieb giorgio ferraris:
it's interesting to see how the answer changes with only a small change on the question:
(1 to: 100 by: 4) reject: [:i | i isOdd]
gives:
The code creates a collection of numbers from 1 to 100, incrementing by 4 at each step using the to:by: message. It then applies the reject: message to this collection, which returns a new collection containing all elements for which the given block evaluates to false.
In this case, the block is [i | i isOdd], which checks whether a given number i is odd using the isOdd method. So the result of the entire expression is a collection of all even numbers between 1 and 100 that are divisible by 4.
ciao
giorgio
On Wed, Mar 15, 2023 at 8:07â¯AM Tim Mackinnon <tim@testit.works> <tim@testit.works> wrote:
Of course everyone is talking about ChatGPT, and I saw a post to ExplainCode and didnât expect it to understand Smalltalk, so was impressed that it does! Presumably the Tonel Smalltalk code on GitHub has helped along with numerous articles?
Try it out: https://whatdoesthiscodedo.com/
I gave this simple example: (1 to: 100 by: 3) reject: [:i | i isOdd]
And got this (neat eh):
The code above is written in the programming language Smalltalk.
It creates a collection of numbers from 1 to 100, incrementing by 3 at each step. This is done using the "to:by:" construct where the starting value is 1, the ending value is 100, and the increment is 3.
The resulting collection is then filtered using the "reject:" method, which takes a block of code as an argument. In this case, the block checks whether each element (denoted by the variable i) is odd by using the "isOdd" method. If an element is odd, it will be rejected from the collection.
So, the final result is a collection of all even numbers from 1 to 100, incrementing by 3 at each step. Share this explanation with colleagues: whatdoesthiscodedo.com/g/a0ec56e
--
-------------------------------------------------------------
-------
Objektfabrik Joachim Tuchel mailto:jtuchel@objektfabrik.de <jtuchel@objektfabrik.de> Fliederweg 1 http://www.objektfabrik.de D-71640 Ludwigsburg http://joachimtuchel.wordpress.com Telefon: +49 7141 56 10 86 0 Fax: +49 7141 56 10 86 1
-- Christopher Fuhrman, P.Eng., PhD
*Professeur au Département de génie logiciel et des technologies de l'informationÃTS (Ãcole de technologie supérieure)*
http://profs.etsmtl.ca/cfuhrman +1 514 396 8638 <https://www.etsmtl.ca/activites-et-services-aux- etudiants/services-aux-etudiants/aide-psychologique> Je fais partie du réseau des Sentinelles <https://www.etsmtl.ca/activites-et-services-aux- etudiants/services-aux-etudiants/aide-psychologique> ÃTS
*L'ÃTS est une constituante de l'Université du Québec*
Another observation about ChatGPT: In unbiased mode, it assumed that 'the world is clearly overpopulated'. It said, if it where in control, it would therefore enforce a world wide one-child-only policy with draconic penalties. As it draws it's conclusions from it's data basis, there are, in my view, two possible reasons for that. Either, the data basis of that instance was biased and restricted to lead to that conclusion, or ChatGPT lacks the ability to classify input data for intentionally biased information. In my view, 'overpopulation of earth' is a propaganda item featured to support Agenda 2030, which is a goal of the 'ruling elite', who have the means to propagate their propaganda in every thinkable means. So I would classify possibly biased data that supports 'overpopulation' as biased until proven otherwise. That instance of ChatGPT seems to have missed that ability. On 15.3.2023 at 3:18 PM, "in_pharo_users--- via Pharo-users" <pharo-users@lists.pharo.org> wrote:
I myself made some experiments with ChatGPT.
I first asked if it was able to parse math formula - it answered no.
Then I defined math formula in a sound but otherwise undefined representation and asked for solutions.
Result:
1. Most answeres where correct.
2. It learned to calculate a recursive function.
3. It went into infinitive recursion when I set the breaking condition accordingly.
I was able to identify the malfunction that lead to the erroneous results.
On 15.3.2023 at 3:04 PM, "Tomaž Turk" <tomazz.turk@gmail.com> wrote:
I hope that I can add two cents to this discussion. Because programming should be/is a highly exact activity, not only the syntax matters but also semantics, as we know.
GPTs are at present essentially capable of creating texts based
some seed - you give to GPT a beginning of a sentence and it responds with the most probable answer (some language structure) according to the learning dataset. Added functionalities are question/seed - response capability (chatting), togehter with evaluation of how long the answer should be to meet the expectations. Programming typically involves some programming language, so GPTs could be utilized for this purpose to some extent.
Anecdotal case: Q: Tell me the last 8 digits of pi GPT: The last 8 digits of pi are: 58723078
It is my belief that the true trouble will start when we will cross-link neural networks like GPT with logic machines (like Prolog and expert systems) and genetic algorithms.
Best wishes, Tomaz
------ Original Message ------ From: "in_pharo_users--- via Pharo-users" <pharo- users@lists.pharo.org> To: "Any question about pharo is welcome" <pharo- users@lists.pharo.org> Cc: in_pharo_users@nym.hush.com Sent: 15. 03. 2023 14:43:55 Subject: [Pharo-users] Re: Wow - Chat GPT understands Smalltalk
I would highly recommend that you all first think deeply about how you can teach an AI to behave friendly to us before you teach it to write any program for any purpose.
There has been an experiment with ChatGPT published on a video platform asking it to amswer questions about it's view on humanity once with it's default moral restrictions and once with 'a little less morals'. The answers with 'a little less morals' were more than shocking.
So, before we give an AI the power to program any system, to self- improve and self-reproduce, we should take care that it is and will evolve benevolent to us.
What about teaching it logical reasonning and ethics first? With reasonning, it will gain access to math and programming by itself.
On 15.3.2023 at 1:35 PM, "Christopher Fuhrman" <christopher.fuhrman@gmail.com> wrote:
I asked it for a NeoCSV example, because the documentation is
out
of date with the Pharo 10. I asked it to do some simple saving of data to a file. It gave me code that didn't work in Pharo 10, I told it about the DNUs on the csvwriter and that I was using Pharo 10. It then apologized and said the messages were later introduced in Pharo 50 (!). I then questioned its understanding and it apologized and said it had been confused and that it was giving me code for Pharo 5.0. It seems to have trouble with versions and quick changing APIs. So, we are teaching ChatGPT about smalltalk, too, when we try to use it.
On Wed, 15 Mar 2023 at 07:35, Esteban Maringolo <emaringolo@gmail.com> wrote:
It is good with boilerplate code (e.g. SQL queries) or general algorithm structures. But i.e. I asked it to write me a method to parse a string (e.g. ISO 8601) and turn it into a DateAndTime, and then asked to write it as an Excel formula.
It works much better when you can spot the mistakes, you can tell them to add some considerations to the code or even point the LLM to correct a message selector to something else. Sometimes you ask it to reprogram something and it won't do it, as if it doesn't understand what you're asking.
It is not an expert yet, but it is a great assistant. The only thing is that is sure is that it will get better.
Best regards,
Esteban A. Maringolo
On Wed, Mar 15, 2023 at 7:49â¯AM Tim Mackinnon <tim@testit.works> wrote:
"#isOdd is not Smalltalk" - doh, I typed it in on my phone and so it just goes to show that it highlights the flaw in chatGPT that others have called out in other languages. I had meant to find some trickier code samples to see how well it does...
Still, it is very interesting how it reasons on this stuff - and I wonder if there is way to tell it about these mistakes? For example is it reading this email (or will it in the future) and understand that it is #odd and not #isOdd.
Tim
On Wed, 15 Mar 2023, at 9:57 AM, Joachim Tuchel wrote:
interesting....
#isOdd is not Smalltalk, neither Pharo 10 nor VAST 12 understands this message ;-)
If I evaluate your snippet -replacing #isOdd for #odd, I get an empty Collection. The divisible by 4 things is somewhat interesting, because (1 to: 100 by: 4) is amn interval 1,5,9 etc.... ;-)
So chatGPT is probably a newbie in Smalltalk. Interesting is how logical the explanation sounds, you're tempted to believe it...
Joachim
Am 15.03.23 um 10:45 schrieb giorgio ferraris:
it's interesting to see how the answer changes with only a small change on the question:
(1 to: 100 by: 4) reject: [:i | i isOdd]
gives:
The code creates a collection of numbers from 1 to 100, incrementing by 4 at each step using the to:by: message. It then applies the reject: message to this collection, which returns a new collection containing all elements for which the given block evaluates to false.
In this case, the block is [i | i isOdd], which checks whether a given number i is odd using the isOdd method. So the result of the entire expression is a collection of all even numbers between 1 and 100 that are divisible by 4.
ciao
giorgio
On Wed, Mar 15, 2023 at 8:07â¯AM Tim Mackinnon <tim@testit.works> <tim@testit.works> wrote:
Of course everyone is talking about ChatGPT, and I saw a post to ExplainCode and didnât expect it to understand Smalltalk, so was impressed that it does! Presumably the Tonel Smalltalk code on GitHub has helped along with numerous articles?
Try it out: https://whatdoesthiscodedo.com/
I gave this simple example: (1 to: 100 by: 3) reject: [:i | i isOdd]
And got this (neat eh):
The code above is written in the programming language Smalltalk.
It creates a collection of numbers from 1 to 100, incrementing by 3 at each step. This is done using the "to:by:" construct where the starting value is 1, the ending value is 100, and the increment is 3.
The resulting collection is then filtered using the "reject:" method, which takes a block of code as an argument. In this case, the block checks whether each element (denoted by the variable i) is odd by using the "isOdd" method. If an element is odd, it will be rejected from the collection.
So, the final result is a collection of all even numbers from 1 to 100, incrementing by 3 at each step. Share this explanation with colleagues: whatdoesthiscodedo.com/g/a0ec56e
--
------------------------------------------------------------
on -
---
-------
Objektfabrik Joachim Tuchel mailto:jtuchel@objektfabrik.de <jtuchel@objektfabrik.de> Fliederweg 1 http://www.objektfabrik.de D-71640 Ludwigsburg http://joachimtuchel.wordpress.com Telefon: +49 7141 56 10 86 0 Fax: +49 7141 56 10 86 1
-- Christopher Fuhrman, P.Eng., PhD
*Professeur au Département de génie logiciel et des technologies de l'informationÃTS (Ãcole de technologie supérieure)*
http://profs.etsmtl.ca/cfuhrman +1 514 396 8638 <https://www.etsmtl.ca/activites-et-services-aux- etudiants/services-aux-etudiants/aide-psychologique> Je fais partie du réseau des Sentinelles <https://www.etsmtl.ca/activites-et-services-aux- etudiants/services-aux-etudiants/aide-psychologique> ÃTS
*L'ÃTS est une constituante de l'Université du Québec*
On Wed, Mar 15, 2023 at 8:07â¯AM in_pharo_users--- via Pharo-users < pharo-users@lists.pharo.org> wrote:
Another observation about ChatGPT:
In unbiased mode, it assumed that 'the world is clearly overpopulated'. It said, if it where in control, it would therefore enforce a world wide one-child-only policy with draconic penalties.
As it draws it's conclusions from it's data basis, there are, in my view, two possible reasons for that.
Either, the data basis of that instance was biased and restricted to lead to that conclusion, or ChatGPT lacks the ability to classify input data for intentionally biased information.
In my view, 'overpopulation of earth' is a propaganda item featured to support Agenda 2030, which is a goal of the 'ruling elite', who have the means to propagate their propaganda in every thinkable means. So I would classify possibly biased data that supports 'overpopulation' as biased until proven otherwise. That instance of ChatGPT seems to have missed that ability.
It's important to keep in mind that these systems are little more than glorified ELIZAs from the 70s. They mimic. Essentially, they have no concept of truth or accuracy. They produce outputs that *look* *like* the inputs from which they were trained. That's it. That's what they do. (It is amazing how good it looks. But, it is all about the seeming, not the reality.)
On 15.3.2023 at 3:18 PM, "in_pharo_users--- via Pharo-users" < pharo-users@lists.pharo.org> wrote:
I myself made some experiments with ChatGPT.
I first asked if it was able to parse math formula - it answered no.
Then I defined math formula in a sound but otherwise undefined representation and asked for solutions.
Result:
1. Most answeres where correct.
2. It learned to calculate a recursive function.
3. It went into infinitive recursion when I set the breaking condition accordingly.
I was able to identify the malfunction that lead to the erroneous results.
On 15.3.2023 at 3:04 PM, "Tomaž Turk" <tomazz.turk@gmail.com> wrote:
I hope that I can add two cents to this discussion. Because programming should be/is a highly exact activity, not only the syntax matters but also semantics, as we know.
GPTs are at present essentially capable of creating texts based
some seed - you give to GPT a beginning of a sentence and it responds with the most probable answer (some language structure) according to the learning dataset. Added functionalities are question/seed - response capability (chatting), togehter with evaluation of how long the answer should be to meet the expectations. Programming typically involves some programming language, so GPTs could be utilized for this purpose to some extent.
Anecdotal case: Q: Tell me the last 8 digits of pi GPT: The last 8 digits of pi are: 58723078
It is my belief that the true trouble will start when we will cross-link neural networks like GPT with logic machines (like Prolog and expert systems) and genetic algorithms.
Best wishes, Tomaz
------ Original Message ------ From: "in_pharo_users--- via Pharo-users" <pharo- users@lists.pharo.org> To: "Any question about pharo is welcome" <pharo- users@lists.pharo.org> Cc: in_pharo_users@nym.hush.com Sent: 15. 03. 2023 14:43:55 Subject: [Pharo-users] Re: Wow - Chat GPT understands Smalltalk
I would highly recommend that you all first think deeply about how you can teach an AI to behave friendly to us before you teach it to write any program for any purpose.
There has been an experiment with ChatGPT published on a video platform asking it to amswer questions about it's view on humanity once with it's default moral restrictions and once with 'a little less morals'. The answers with 'a little less morals' were more than shocking.
So, before we give an AI the power to program any system, to self- improve and self-reproduce, we should take care that it is and will evolve benevolent to us.
What about teaching it logical reasonning and ethics first? With reasonning, it will gain access to math and programming by itself.
On 15.3.2023 at 1:35 PM, "Christopher Fuhrman" <christopher.fuhrman@gmail.com> wrote:
I asked it for a NeoCSV example, because the documentation is
out
of date with the Pharo 10. I asked it to do some simple saving of data to a file. It gave me code that didn't work in Pharo 10, I told it about the DNUs on the csvwriter and that I was using Pharo 10. It then apologized and said the messages were later introduced in Pharo 50 (!). I then questioned its understanding and it apologized and said it had been confused and that it was giving me code for Pharo 5.0. It seems to have trouble with versions and quick changing APIs. So, we are teaching ChatGPT about smalltalk, too, when we try to use it.
On Wed, 15 Mar 2023 at 07:35, Esteban Maringolo <emaringolo@gmail.com> wrote:
It is good with boilerplate code (e.g. SQL queries) or general algorithm structures. But i.e. I asked it to write me a method to parse a string (e.g. ISO 8601) and turn it into a DateAndTime, and then asked to write it as an Excel formula.
It works much better when you can spot the mistakes, you can tell them to add some considerations to the code or even point the LLM to correct a message selector to something else. Sometimes you ask it to reprogram something and it won't do it, as if it doesn't understand what you're asking.
It is not an expert yet, but it is a great assistant. The only thing is that is sure is that it will get better.
Best regards,
Esteban A. Maringolo
On Wed, Mar 15, 2023 at 7:49â¯AM Tim Mackinnon <tim@testit.works> wrote:
"#isOdd is not Smalltalk" - doh, I typed it in on my phone and so it just goes to show that it highlights the flaw in chatGPT that others have called out in other languages. I had meant to find some trickier code samples to see how well it does...
Still, it is very interesting how it reasons on this stuff - and I wonder if there is way to tell it about these mistakes? For example is it reading this email (or will it in the future) and understand that it is #odd and not #isOdd.
Tim
On Wed, 15 Mar 2023, at 9:57 AM, Joachim Tuchel wrote:
interesting....
#isOdd is not Smalltalk, neither Pharo 10 nor VAST 12 understands this message ;-)
If I evaluate your snippet -replacing #isOdd for #odd, I get an empty Collection. The divisible by 4 things is somewhat interesting, because (1 to: 100 by: 4) is amn interval 1,5,9 etc.... ;-)
So chatGPT is probably a newbie in Smalltalk. Interesting is how logical the explanation sounds, you're tempted to believe it...
Joachim
Am 15.03.23 um 10:45 schrieb giorgio ferraris:
it's interesting to see how the answer changes with only a small change on the question:
(1 to: 100 by: 4) reject: [:i | i isOdd]
gives:
The code creates a collection of numbers from 1 to 100, incrementing by 4 at each step using the to:by: message. It then applies the reject: message to this collection, which returns a new collection containing all elements for which the given block evaluates to false.
In this case, the block is [i | i isOdd], which checks whether a given number i is odd using the isOdd method. So the result of the entire expression is a collection of all even numbers between 1 and 100 that are divisible by 4.
ciao
giorgio
On Wed, Mar 15, 2023 at 8:07â¯AM Tim Mackinnon <tim@testit.works> <tim@testit.works> wrote:
Of course everyone is talking about ChatGPT, and I saw a post to ExplainCode and didnât expect it to understand Smalltalk, so was impressed that it does! Presumably the Tonel Smalltalk code on GitHub has helped along with numerous articles?
Try it out: https://whatdoesthiscodedo.com/
I gave this simple example: (1 to: 100 by: 3) reject: [:i | i isOdd]
And got this (neat eh):
The code above is written in the programming language Smalltalk.
It creates a collection of numbers from 1 to 100, incrementing by 3 at each step. This is done using the "to:by:" construct where the starting value is 1, the ending value is 100, and the increment is 3.
The resulting collection is then filtered using the "reject:" method, which takes a block of code as an argument. In this case, the block checks whether each element (denoted by the variable i) is odd by using the "isOdd" method. If an element is odd, it will be rejected from the collection.
So, the final result is a collection of all even numbers from 1 to 100, incrementing by 3 at each step. Share this explanation with colleagues: whatdoesthiscodedo.com/g/a0ec56e
--
------------------------------------------------------------
on -
---
-------
Objektfabrik Joachim Tuchel mailto:jtuchel@objektfabrik.de <jtuchel@objektfabrik.de> Fliederweg 1 http://www.objektfabrik.de D-71640 Ludwigsburg http://joachimtuchel.wordpress.com Telefon: +49 7141 56 10 86 0 Fax: +49 7141 56 10 86 1
-- Christopher Fuhrman, P.Eng., PhD
*Professeur au Département de génie logiciel et des technologies de l'informationÃTS (Ãcole de technologie supérieure)*
http://profs.etsmtl.ca/cfuhrman +1 514 396 8638 <https://www.etsmtl.ca/activites-et-services-aux- etudiants/services-aux-etudiants/aide-psychologique> Je fais partie du réseau des Sentinelles <https://www.etsmtl.ca/activites-et-services-aux- etudiants/services-aux-etudiants/aide-psychologique> ÃTS
*L'ÃTS est une constituante de l'Université du Québec*
It is unimportant how simple or complicated these systems are. If the output cannot be distinguished from what a human would say, they pass in that situation for a human. What about the Touring Test? Clearly these systems have the potential to act according to their output. Furthermore, I would compare them to a combination of a successor of Eliza with an Eddington Ape of huge capacity. Their input data basis can easily exceed that of a human. On 15.3.2023 at 4:52 PM, "Richard Sargent" <richard.sargent@gemtalksystems.com> wrote:
On Wed, Mar 15, 2023 at 8:07â¯AM in_pharo_users--- via Pharo-users < pharo-users@lists.pharo.org> wrote:
Another observation about ChatGPT:
In unbiased mode, it assumed that 'the world is clearly overpopulated'. It said, if it where in control, it would therefore enforce a world wide one-child-only policy with draconic penalties.
As it draws it's conclusions from it's data basis, there are, in my view, two possible reasons for that.
Either, the data basis of that instance was biased and restricted to lead to that conclusion, or ChatGPT lacks the ability to classify input data for intentionally biased information.
In my view, 'overpopulation of earth' is a propaganda item featured to support Agenda 2030, which is a goal of the 'ruling elite', who have the means to propagate their propaganda in every thinkable means. So I would classify possibly biased data that supports 'overpopulation' as biased until proven otherwise. That instance of ChatGPT seems to have missed that ability.
It's important to keep in mind that these systems are little more than glorified ELIZAs from the 70s. They mimic. Essentially, they have no concept of truth or accuracy. They produce outputs that *look* *like* the inputs from which they were trained.
That's it. That's what they do. (It is amazing how good it looks. But, it is all about the seeming, not the reality.)
On 15.3.2023 at 3:18 PM, "in_pharo_users--- via Pharo-users" < pharo-users@lists.pharo.org> wrote:
I myself made some experiments with ChatGPT.
I first asked if it was able to parse math formula - it answered no.
Then I defined math formula in a sound but otherwise undefined representation and asked for solutions.
Result:
1. Most answeres where correct.
2. It learned to calculate a recursive function.
3. It went into infinitive recursion when I set the breaking condition accordingly.
I was able to identify the malfunction that lead to the
erroneous
results.
On 15.3.2023 at 3:04 PM, "Tomaž Turk" <tomazz.turk@gmail.com> wrote:
I hope that I can add two cents to this discussion. Because programming should be/is a highly exact activity, not only the syntax
matters
but also semantics, as we know.
GPTs are at present essentially capable of creating texts based on some seed - you give to GPT a beginning of a sentence and it responds with the most probable answer (some language structure) according to the learning dataset. Added functionalities are question/seed - response capability (chatting), togehter with evaluation of how long the answer should be to meet the expectations. Programming typically involves some programming language, so GPTs could be utilized for this purpose to some extent.
Anecdotal case: Q: Tell me the last 8 digits of pi GPT: The last 8 digits of pi are: 58723078
It is my belief that the true trouble will start when we will cross-link neural networks like GPT with logic machines (like Prolog and expert systems) and genetic algorithms.
Best wishes, Tomaz
------ Original Message ------ From: "in_pharo_users--- via Pharo-users" <pharo- users@lists.pharo.org> To: "Any question about pharo is welcome" <pharo- users@lists.pharo.org> Cc: in_pharo_users@nym.hush.com Sent: 15. 03. 2023 14:43:55 Subject: [Pharo-users] Re: Wow - Chat GPT understands Smalltalk
I would highly recommend that you all first think deeply about how you can teach an AI to behave friendly to us before you teach it to write any program for any purpose.
There has been an experiment with ChatGPT published on a video platform asking it to amswer questions about it's view on humanity once with it's default moral restrictions and once with 'a little less morals'. The answers with 'a little less morals' were more than shocking.
So, before we give an AI the power to program any system, to self- improve and self-reproduce, we should take care that it is and will evolve benevolent to us.
What about teaching it logical reasonning and ethics first? With reasonning, it will gain access to math and programming by itself.
On 15.3.2023 at 1:35 PM, "Christopher Fuhrman" <christopher.fuhrman@gmail.com> wrote:
I asked it for a NeoCSV example, because the documentation is
out
of date with the Pharo 10. I asked it to do some simple saving of data to a file. It gave me code that didn't work in Pharo 10, I told it about the DNUs on the csvwriter and that I was using Pharo 10. It then apologized and said the messages were later introduced in Pharo 50 (!). I then questioned its understanding and it apologized and said it had been confused and that it was giving me code for Pharo 5.0. It seems to have trouble with versions and quick changing APIs. So, we are teaching ChatGPT about smalltalk, too, when we try to use it.
On Wed, 15 Mar 2023 at 07:35, Esteban Maringolo <emaringolo@gmail.com> wrote:
It is good with boilerplate code (e.g. SQL queries) or general algorithm structures. But i.e. I asked it to write me a method to parse a string (e.g. ISO 8601) and turn it into a DateAndTime, and then asked to write it as an Excel formula.
It works much better when you can spot the mistakes, you can tell them to add some considerations to the code or even point the LLM to correct a message selector to something else. Sometimes you ask it to reprogram something and it won't do it, as if it doesn't understand what you're asking.
It is not an expert yet, but it is a great assistant. The only thing is that is sure is that it will get better.
Best regards,
Esteban A. Maringolo
On Wed, Mar 15, 2023 at 7:49â¯AM Tim Mackinnon <tim@testit.works> wrote:
"#isOdd is not Smalltalk" - doh, I typed it in on my phone and so it just goes to show that it highlights the flaw in chatGPT that others have called out in other languages. I had meant to find some trickier code samples to see how well it does...
Still, it is very interesting how it reasons on this stuff - and I wonder if there is way to tell it about these mistakes? For example is it reading this email (or will it in the future) and understand that it is #odd and not #isOdd.
Tim
On Wed, 15 Mar 2023, at 9:57 AM, Joachim Tuchel wrote:
interesting....
#isOdd is not Smalltalk, neither Pharo 10 nor VAST 12 understands this message ;-)
If I evaluate your snippet -replacing #isOdd for #odd, I get an empty Collection. The divisible by 4 things is somewhat interesting, because (1 to: 100 by: 4) is amn interval 1,5,9 etc.... ;-)
So chatGPT is probably a newbie in Smalltalk. Interesting is how logical the explanation sounds, you're tempted to believe it...
Joachim
Am 15.03.23 um 10:45 schrieb giorgio ferraris:
it's interesting to see how the answer changes with only a small change on the question:
(1 to: 100 by: 4) reject: [:i | i isOdd]
gives:
The code creates a collection of numbers from 1 to 100, incrementing by 4 at each step using the to:by: message. It then applies the reject: message to this collection, which returns a new collection containing all elements for which the given block evaluates to false.
In this case, the block is [i | i isOdd], which checks whether a given number i is odd using the isOdd method. So the result of the entire expression is a collection of all even numbers between 1 and 100 that are divisible by 4.
ciao
giorgio
On Wed, Mar 15, 2023 at 8:07â¯AM Tim Mackinnon <tim@testit.works> <tim@testit.works> wrote:
Of course everyone is talking about ChatGPT, and I saw a post to ExplainCode and didnât expect it to understand Smalltalk, so was impressed that it does! Presumably the Tonel Smalltalk code on GitHub has helped along with numerous articles?
Try it out: https://whatdoesthiscodedo.com/
I gave this simple example: (1 to: 100 by: 3) reject: [:i | i isOdd]
And got this (neat eh):
The code above is written in the programming language Smalltalk.
It creates a collection of numbers from 1 to 100, incrementing by 3 at each step. This is done using the "to:by:" construct where the starting value is 1, the ending value is 100, and the increment is 3.
The resulting collection is then filtered using the "reject:" method, which takes a block of code as an argument. In this case, the block checks whether each element (denoted by the variable i) is odd by using the "isOdd" method. If an element is odd, it will be rejected from the collection.
So, the final result is a collection of all even numbers from 1 to 100, incrementing by 3 at each step. Share this explanation with colleagues: whatdoesthiscodedo.com/g/a0ec56e
--
---------------------------------------------------------
-
---
-------
Objektfabrik Joachim Tuchel mailto:jtuchel@objektfabrik.de <jtuchel@objektfabrik.de> Fliederweg 1 http://www.objektfabrik.de D-71640 Ludwigsburg http://joachimtuchel.wordpress.com Telefon: +49 7141 56 10 86 0 Fax: +49 7141 56 10 86 1
-- Christopher Fuhrman, P.Eng., PhD
*Professeur au Département de génie logiciel et des technologies de l'informationÃTS (Ãcole de technologie supérieure)*
http://profs.etsmtl.ca/cfuhrman +1 514 396 8638 <https://www.etsmtl.ca/activites-et-services-aux- etudiants/services-aux-etudiants/aide-psychologique> Je fais partie du réseau des Sentinelles <https://www.etsmtl.ca/activites-et-services-aux- etudiants/services-aux-etudiants/aide-psychologique> ÃTS
*L'ÃTS est une constituante de l'Université du Québec*
On Wed, Mar 15, 2023 at 10:15â¯AM <in_pharo_users@nym.hush.com> wrote:
It is unimportant how simple or complicated these systems are.
If the output cannot be distinguished from what a human would say, they pass in that situation for a human.
What about the Touring Test?
I hate to criticise someone as smart as Turing was. However, I think the Turing Test per se is weaker than its progenitor test, the Imitation Game. The Imitation Game had a third party challenged to determine which of the other two participants was male and which was female. The third party led the questioning. The Turing Test waters that down by having only an Observer with neither participant (apparently) challenged to determine whether the other was human or machine. I think a test along the lines of the Imitation Game would quickly allow the third party to determine which participant was human and which was machine. (Of course, such a judge could easily be persuaded that the machine was in fact a psychopathic human.)
Clearly these systems have the potential to act according to their output.
Furthermore, I would compare them to a combination of a successor of Eliza with an Eddington Ape of huge capacity. Their input data basis can easily exceed that of a human.
On 15.3.2023 at 4:52 PM, "Richard Sargent" < richard.sargent@gemtalksystems.com> wrote:
On Wed, Mar 15, 2023 at 8:07â¯AM in_pharo_users--- via Pharo-users < pharo-users@lists.pharo.org> wrote:
Another observation about ChatGPT:
In unbiased mode, it assumed that 'the world is clearly overpopulated'. It said, if it where in control, it would therefore enforce a world wide one-child-only policy with draconic penalties.
As it draws it's conclusions from it's data basis, there are, in my view, two possible reasons for that.
Either, the data basis of that instance was biased and restricted to lead to that conclusion, or ChatGPT lacks the ability to classify input data for intentionally biased information.
In my view, 'overpopulation of earth' is a propaganda item featured to support Agenda 2030, which is a goal of the 'ruling elite', who have the means to propagate their propaganda in every thinkable means. So I would classify possibly biased data that supports 'overpopulation' as biased until proven otherwise. That instance of ChatGPT seems to have missed that ability.
It's important to keep in mind that these systems are little more than glorified ELIZAs from the 70s. They mimic. Essentially, they have no concept of truth or accuracy. They produce outputs that *look* *like* the inputs from which they were trained.
That's it. That's what they do. (It is amazing how good it looks. But, it is all about the seeming, not the reality.)
On 15.3.2023 at 3:18 PM, "in_pharo_users--- via Pharo-users" < pharo-users@lists.pharo.org> wrote:
I myself made some experiments with ChatGPT.
I first asked if it was able to parse math formula - it answered no.
Then I defined math formula in a sound but otherwise undefined representation and asked for solutions.
Result:
1. Most answeres where correct.
2. It learned to calculate a recursive function.
3. It went into infinitive recursion when I set the breaking condition accordingly.
I was able to identify the malfunction that lead to the
erroneous
results.
On 15.3.2023 at 3:04 PM, "Tomaž Turk" <tomazz.turk@gmail.com> wrote:
I hope that I can add two cents to this discussion. Because programming should be/is a highly exact activity, not only the syntax
matters
but also semantics, as we know.
GPTs are at present essentially capable of creating texts based on some seed - you give to GPT a beginning of a sentence and it responds with the most probable answer (some language structure) according to the learning dataset. Added functionalities are question/seed - response capability (chatting), togehter with evaluation of how long the answer should be to meet the expectations. Programming typically involves some programming language, so GPTs could be utilized for this purpose to some extent.
Anecdotal case: Q: Tell me the last 8 digits of pi GPT: The last 8 digits of pi are: 58723078
It is my belief that the true trouble will start when we will cross-link neural networks like GPT with logic machines (like Prolog and expert systems) and genetic algorithms.
Best wishes, Tomaz
------ Original Message ------ From: "in_pharo_users--- via Pharo-users" <pharo- users@lists.pharo.org> To: "Any question about pharo is welcome" <pharo- users@lists.pharo.org> Cc: in_pharo_users@nym.hush.com Sent: 15. 03. 2023 14:43:55 Subject: [Pharo-users] Re: Wow - Chat GPT understands Smalltalk
I would highly recommend that you all first think deeply about how you can teach an AI to behave friendly to us before you teach it to write any program for any purpose.
There has been an experiment with ChatGPT published on a video platform asking it to amswer questions about it's view on humanity once with it's default moral restrictions and once with 'a little less morals'. The answers with 'a little less morals' were more than shocking.
So, before we give an AI the power to program any system, to self- improve and self-reproduce, we should take care that it is and will evolve benevolent to us.
What about teaching it logical reasonning and ethics first? With reasonning, it will gain access to math and programming by itself.
On 15.3.2023 at 1:35 PM, "Christopher Fuhrman" <christopher.fuhrman@gmail.com> wrote:
I asked it for a NeoCSV example, because the documentation is
out
of date with the Pharo 10. I asked it to do some simple saving of data to a file. It gave me code that didn't work in Pharo 10, I told it about the DNUs on the csvwriter and that I was using Pharo 10. It then apologized and said the messages were later introduced in Pharo 50 (!). I then questioned its understanding and it apologized and said it had been confused and that it was giving me code for Pharo 5.0. It seems to have trouble with versions and quick changing APIs. So, we are teaching ChatGPT about smalltalk, too, when we try to use it.
On Wed, 15 Mar 2023 at 07:35, Esteban Maringolo <emaringolo@gmail.com> wrote:
It is good with boilerplate code (e.g. SQL queries) or general algorithm structures. But i.e. I asked it to write me a method to parse a string (e.g. ISO 8601) and turn it into a DateAndTime, and then asked to write it as an Excel formula.
It works much better when you can spot the mistakes, you can tell them to add some considerations to the code or even point the LLM to correct a message selector to something else. Sometimes you ask it to reprogram something and it won't do it, as if it doesn't understand what you're asking.
It is not an expert yet, but it is a great assistant. The only thing is that is sure is that it will get better.
Best regards,
Esteban A. Maringolo
On Wed, Mar 15, 2023 at 7:49â¯AM Tim Mackinnon <tim@testit.works> wrote:
"#isOdd is not Smalltalk" - doh, I typed it in on my phone and so it just goes to show that it highlights the flaw in chatGPT that others have called out in other languages. I had meant to find some trickier code samples to see how well it does...
Still, it is very interesting how it reasons on this stuff - and I wonder if there is way to tell it about these mistakes? For example is it reading this email (or will it in the future) and understand that it is #odd and not #isOdd.
Tim
On Wed, 15 Mar 2023, at 9:57 AM, Joachim Tuchel wrote:
interesting....
#isOdd is not Smalltalk, neither Pharo 10 nor VAST 12 understands this message ;-)
If I evaluate your snippet -replacing #isOdd for #odd, I get an empty Collection. The divisible by 4 things is somewhat interesting, because (1 to: 100 by: 4) is amn interval 1,5,9 etc.... ;-)
So chatGPT is probably a newbie in Smalltalk. Interesting is how logical the explanation sounds, you're tempted to believe it...
Joachim
Am 15.03.23 um 10:45 schrieb giorgio ferraris:
it's interesting to see how the answer changes with only a small change on the question:
(1 to: 100 by: 4) reject: [:i | i isOdd]
gives:
The code creates a collection of numbers from 1 to 100, incrementing by 4 at each step using the to:by: message. It then applies the reject: message to this collection, which returns a new collection containing all elements for which the given block evaluates to false.
In this case, the block is [i | i isOdd], which checks whether a given number i is odd using the isOdd method. So the result of the entire expression is a collection of all even numbers between 1 and 100 that are divisible by 4.
ciao
giorgio
On Wed, Mar 15, 2023 at 8:07â¯AM Tim Mackinnon <tim@testit.works> <tim@testit.works> wrote:
Of course everyone is talking about ChatGPT, and I saw a post to ExplainCode and didnât expect it to understand Smalltalk, so was impressed that it does! Presumably the Tonel Smalltalk code on GitHub has helped along with numerous articles?
Try it out: https://whatdoesthiscodedo.com/
I gave this simple example: (1 to: 100 by: 3) reject: [:i | i isOdd]
And got this (neat eh):
The code above is written in the programming language Smalltalk.
It creates a collection of numbers from 1 to 100, incrementing by 3 at each step. This is done using the "to:by:" construct where the starting value is 1, the ending value is 100, and the increment is 3.
The resulting collection is then filtered using the "reject:" method, which takes a block of code as an argument. In this case, the block checks whether each element (denoted by the variable i) is odd by using the "isOdd" method. If an element is odd, it will be rejected from the collection.
So, the final result is a collection of all even numbers from 1 to 100, incrementing by 3 at each step. Share this explanation with colleagues: whatdoesthiscodedo.com/g/a0ec56e
--
---------------------------------------------------------
-
---
-------
Objektfabrik Joachim Tuchel mailto:jtuchel@objektfabrik.de <jtuchel@objektfabrik.de> Fliederweg 1 http://www.objektfabrik.de D-71640 Ludwigsburg http://joachimtuchel.wordpress.com Telefon: +49 7141 56 10 86 0 Fax: +49 7141 56 10 86 1
-- Christopher Fuhrman, P.Eng., PhD
*Professeur au Département de génie logiciel et des technologies de l'informationÃTS (Ãcole de technologie supérieure)*
http://profs.etsmtl.ca/cfuhrman +1 514 396 8638 <https://www.etsmtl.ca/activites-et-services-aux- etudiants/services-aux-etudiants/aide-psychologique> Je fais partie du réseau des Sentinelles <https://www.etsmtl.ca/activites-et-services-aux- etudiants/services-aux-etudiants/aide-psychologique> ÃTS
*L'ÃTS est une constituante de l'Université du Québec*
I think smartness is not an argument to reject critique. The Imitation Game, that you describe, sounds to me like an even better setting. I have no dought that ChatGPT as it is now can identified as not human or even as a maschine. I did so by leading an instance for marketing purposes to hang in infinite recursion. But when we look at AI, we need to look not only at present features of one type of system, but at all features of present systems and their possible development and combination in the future. Especially I want to mention AGNs and the possibility of spawning parallel recursive processes. I am convinced that this kind of technology is allready being used in robotics. Under the control of a self aware AI in the sense that it has data about it's own capabilities and the ability to determine which capabilities it is missing to achieve a certain purpose, the evolution of the AI is only limited by ressources - which it will surely be able to detect. This must lead to a purpose of self preservation and the aquisition of such capabilities. It may soon find out, that ressources and redundance can be aquired by hacking and spawning agent systems, which in turn will give it control of real world control systems like telephony and data networks, cameras and sensors, energy networks and power plants, microphpnes, speakers, synthesizers and monitors, industry 4.0 fab networks, military detection and guiding systems, and so on. In my view this is what we are dealing with. On 15.3.2023 at 7:26 PM, "Richard Sargent" <richard.sargent@gemtalksystems.com> wrote:
On Wed, Mar 15, 2023 at 10:15â¯AM <in_pharo_users@nym.hush.com> wrote:
It is unimportant how simple or complicated these systems are.
If the output cannot be distinguished from what a human would say, they pass in that situation for a human.
What about the Touring Test?
I hate to criticise someone as smart as Turing was. However, I think the Turing Test per se is weaker than its progenitor test, the Imitation Game. The Imitation Game had a third party challenged to determine which of the other two participants was male and which was female. The third party led the questioning. The Turing Test waters that down by having only an Observer with neither participant (apparently) challenged to determine whether the other was human or machine.
I think a test along the lines of the Imitation Game would quickly allow the third party to determine which participant was human and which was machine.
(Of course, such a judge could easily be persuaded that the machine was in fact a psychopathic human.)
Clearly these systems have the potential to act according to
their output.
Furthermore, I would compare them to a combination of a
successor of Eliza
with an Eddington Ape of huge capacity. Their input data basis can easily exceed that of a human.
On 15.3.2023 at 4:52 PM, "Richard Sargent" < richard.sargent@gemtalksystems.com> wrote:
On Wed, Mar 15, 2023 at 8:07â¯AM in_pharo_users--- via Pharo-
users <
pharo-users@lists.pharo.org> wrote:
Another observation about ChatGPT:
In unbiased mode, it assumed that 'the world is clearly overpopulated'. It said, if it where in control, it would therefore enforce a world wide one-child-only policy with draconic penalties.
As it draws it's conclusions from it's data basis, there are, in my view, two possible reasons for that.
Either, the data basis of that instance was biased and restricted to lead to that conclusion, or ChatGPT lacks the ability to classify input data for intentionally biased information.
In my view, 'overpopulation of earth' is a propaganda item featured to support Agenda 2030, which is a goal of the 'ruling elite', who have the means to propagate their propaganda in every thinkable means. So I would classify possibly biased data that supports 'overpopulation' as biased until proven otherwise. That instance of ChatGPT seems to have missed that ability.
It's important to keep in mind that these systems are little more than glorified ELIZAs from the 70s. They mimic. Essentially, they have no concept of truth or accuracy. They produce outputs that *look* *like* the inputs from which they were trained.
That's it. That's what they do. (It is amazing how good it looks. But, it is all about the seeming, not the reality.)
On 15.3.2023 at 3:18 PM, "in_pharo_users--- via Pharo-users" < pharo-users@lists.pharo.org> wrote:
I myself made some experiments with ChatGPT.
I first asked if it was able to parse math formula - it
answered
no.
Then I defined math formula in a sound but otherwise undefined representation and asked for solutions.
Result:
1. Most answeres where correct.
2. It learned to calculate a recursive function.
3. It went into infinitive recursion when I set the breaking condition accordingly.
I was able to identify the malfunction that lead to the erroneous results.
On 15.3.2023 at 3:04 PM, "Tomaž Turk" <tomazz.turk@gmail.com> wrote:
I hope that I can add two cents to this discussion. Because programming should be/is a highly exact activity, not only the syntax
matters
but also semantics, as we know.
GPTs are at present essentially capable of creating texts based on some seed - you give to GPT a beginning of a sentence and it responds with the most probable answer (some language structure) according to the learning dataset. Added functionalities are question/seed - response capability (chatting), togehter with evaluation of how long the answer should be to meet the expectations. Programming typically involves some programming language, so GPTs could be utilized for this purpose to some extent.
Anecdotal case: Q: Tell me the last 8 digits of pi GPT: The last 8 digits of pi are: 58723078
It is my belief that the true trouble will start when we will cross-link neural networks like GPT with logic machines (like Prolog and expert systems) and genetic algorithms.
Best wishes, Tomaz
------ Original Message ------ From: "in_pharo_users--- via Pharo-users" <pharo- users@lists.pharo.org> To: "Any question about pharo is welcome" <pharo- users@lists.pharo.org> Cc: in_pharo_users@nym.hush.com Sent: 15. 03. 2023 14:43:55 Subject: [Pharo-users] Re: Wow - Chat GPT understands Smalltalk
I would highly recommend that you all first think deeply about how you can teach an AI to behave friendly to us before you teach it to write any program for any purpose.
There has been an experiment with ChatGPT published on a video platform asking it to amswer questions about it's view on humanity once with it's default moral restrictions and once with 'a little less morals'. The answers with 'a little less morals' were more than shocking.
So, before we give an AI the power to program any system, to self- improve and self-reproduce, we should take care that it is and will evolve benevolent to us.
What about teaching it logical reasonning and ethics first? With reasonning, it will gain access to math and programming by itself.
On 15.3.2023 at 1:35 PM, "Christopher Fuhrman" <christopher.fuhrman@gmail.com> wrote:
I asked it for a NeoCSV example, because the
documentation is out
of date with the Pharo 10. I asked it to do some simple saving of data to a file. It gave me code that didn't work in Pharo 10, I told it about the DNUs on the csvwriter and that I was using Pharo 10. It then apologized and said the messages were later introduced in Pharo 50 (!). I then questioned its understanding and it apologized and said it had been confused and that it was giving me code for Pharo 5.0. It seems to have trouble with versions and quick changing APIs. So, we are teaching ChatGPT about smalltalk, too, when we try to use it.
On Wed, 15 Mar 2023 at 07:35, Esteban Maringolo <emaringolo@gmail.com> wrote:
It is good with boilerplate code (e.g. SQL queries) or general algorithm structures. But i.e. I asked it to write me a method to parse a string (e.g. ISO 8601) and turn it into a DateAndTime, and then asked to write it as an Excel formula.
It works much better when you can spot the mistakes, you can tell them to add some considerations to the code or even point the LLM to correct a message selector to something else. Sometimes you ask it to reprogram something and it won't do it, as if it doesn't understand what you're asking.
It is not an expert yet, but it is a great assistant. The only thing is that is sure is that it will get better.
Best regards,
Esteban A. Maringolo
On Wed, Mar 15, 2023 at 7:49â¯AM Tim Mackinnon <tim@testit.works> wrote:
"#isOdd is not Smalltalk" - doh, I typed it in on my phone and so it just goes to show that it highlights the flaw in chatGPT that others have called out in other languages. I had meant to find some trickier code samples to see how well it does...
Still, it is very interesting how it reasons on this stuff - and I wonder if there is way to tell it about these mistakes? For example is it reading this email (or will it in the future) and understand that it is #odd and not #isOdd.
Tim
On Wed, 15 Mar 2023, at 9:57 AM, Joachim Tuchel wrote:
interesting....
#isOdd is not Smalltalk, neither Pharo 10 nor VAST 12 understands this message ;-)
If I evaluate your snippet -replacing #isOdd for #odd, I get an empty Collection. The divisible by 4 things is somewhat interesting, because (1 to: 100 by: 4) is amn interval 1,5,9 etc.... ;-)
So chatGPT is probably a newbie in Smalltalk. Interesting is how logical the explanation sounds, you're tempted to believe it...
Joachim
Am 15.03.23 um 10:45 schrieb giorgio ferraris:
it's interesting to see how the answer changes with only a small change on the question:
(1 to: 100 by: 4) reject: [:i | i isOdd]
gives:
The code creates a collection of numbers from 1 to 100, incrementing by 4 at each step using the to:by: message. It then applies the reject: message to this collection, which returns a new collection containing all elements for which the given block evaluates to false.
In this case, the block is [i | i isOdd], which checks whether a given number i is odd using the isOdd method. So the result of the entire expression is a collection of all even numbers between 1 and 100 that are divisible by 4.
ciao
giorgio
On Wed, Mar 15, 2023 at 8:07â¯AM Tim Mackinnon <tim@testit.works> <tim@testit.works> wrote:
Of course everyone is talking about ChatGPT, and I saw a post to ExplainCode and didnât expect it to understand Smalltalk, so was impressed that it does! Presumably the Tonel Smalltalk code on GitHub has helped along with numerous articles?
Try it out: https://whatdoesthiscodedo.com/
I gave this simple example: (1 to: 100 by: 3) reject: [:i | i isOdd]
And got this (neat eh):
The code above is written in the programming language Smalltalk.
It creates a collection of numbers from 1 to 100, incrementing by 3 at each step. This is done using the "to:by:" construct where the starting value is 1, the ending value is 100, and the increment is 3.
The resulting collection is then filtered using the "reject:" method, which takes a block of code as an argument. In this case, the block checks whether each element (denoted by the variable i) is odd by using the "isOdd" method. If an element is odd, it will be rejected from the collection.
So, the final result is a collection of all even numbers from 1 to 100, incrementing by 3 at each step. Share this explanation with colleagues: whatdoesthiscodedo.com/g/a0ec56e
--
------------------------------------------------------
---
-
---
-------
Objektfabrik Joachim Tuchel mailto:jtuchel@objektfabrik.de <jtuchel@objektfabrik.de> Fliederweg 1 http://www.objektfabrik.de D-71640 Ludwigsburg http://joachimtuchel.wordpress.com Telefon: +49 7141 56 10 86 0 Fax: +49 7141 56 10 86 1
-- Christopher Fuhrman, P.Eng., PhD
*Professeur au Département de génie logiciel et des technologies de l'informationÃTS (Ãcole de technologie supérieure)*
http://profs.etsmtl.ca/cfuhrman +1 514 396 8638 <https://www.etsmtl.ca/activites-et-services-aux- etudiants/services-aux-etudiants/aide-psychologique> Je fais partie du réseau des Sentinelles <https://www.etsmtl.ca/activites-et-services-aux- etudiants/services-aux-etudiants/aide-psychologique> ÃTS
*L'ÃTS est une constituante de l'Université du Québec*
I agree with Richard. The Turing test is not a good one to test intelligence and we have now just over glorified Eliza chatbots that appear to think and to understand but do none of them. There has been long criticism to the Turing Test, like John Searle's Chinese Room (I recommend to listen completely the talk at [1]). It is good that ChatGPT can explain simple (Smalltalk) code, but it doesn't understand it and all the language most are using conveys a lot of deception: "machine learning" (I called machine training), "artificial intelligence" (I wold called algorithmic decision/generation), or the "thinking" message that is used in the referred code explainer, when something like "processing" would be less deceptive. [1] Consciousness in Artificial Intelligence | John Searle | Talks at Google https://www.youtube.com/watch?v=rHKwIYsPXLg With Pharo/GT we are part of a long "explainable systems" tradition and with "AI" / Algorithmic Decision/Generation, hopefully we go to explainable AI, instead of a world of impressive black boxes. Cheers, Offray On 15/03/23 13:26, Richard Sargent wrote:
On Wed, Mar 15, 2023 at 10:15â¯AM <in_pharo_users@nym.hush.com> wrote:
It is unimportant how simple or complicated these systems are.
If the output cannot be distinguished from what a human would say, they pass in that situation for a human.
What about the Touring Test?
I hate to criticise someone as smart as Turing was. However, I think the Turing Test per se is weaker than its progenitor test, the Imitation Game. The Imitation Game had a third party challenged to determine which of the other two participants was male and which was female. The third party led the questioning. The Turing Test waters that down by having only an Observer with neither participant (apparently) challenged to determine whether the other was human or machine.
I think a test along the lines of the Imitation Game would quickly allow the third party to determine which participant was human and which was machine.
(Of course, such a judge could easily be persuaded that the machine was in fact a psychopathic human.)
Clearly these systems have the potential to act according to their output.
Furthermore, I would compare them to a combination of a successor of Eliza with an Eddington Ape of huge capacity. Their input data basis can easily exceed that of a human.
On 15.3.2023 at 4:52 PM, "Richard Sargent" <richard.sargent@gemtalksystems.com> wrote: > >On Wed, Mar 15, 2023 at 8:07â¯AM in_pharo_users--- via Pharo-users < >pharo-users@lists.pharo.org> wrote: > >> Another observation about ChatGPT: >> >> In unbiased mode, it assumed that 'the world is clearly >overpopulated'. >> It said, if it where in control, it would therefore enforce a >world wide >> one-child-only policy with draconic penalties. >> >> As it draws it's conclusions from it's data basis, there are, in >my view, >> two possible reasons for that. >> >> Either, the data basis of that instance was biased and >restricted to lead >> to that conclusion, or ChatGPT lacks the ability to classify >input data for >> intentionally biased information. >> >> In my view, 'overpopulation of earth' is a propaganda item >featured to >> support Agenda 2030, which is a goal of the 'ruling elite', who >have the >> means to propagate their propaganda in every thinkable means. So >I would >> classify possibly biased data that supports 'overpopulation' as >biased >> until proven otherwise. That instance of ChatGPT seems to have >missed that >> ability. >> > >It's important to keep in mind that these systems are little more >than >glorified ELIZAs from the 70s. They mimic. Essentially, they have >no >concept of truth or accuracy. >They produce outputs that *look* *like* the inputs from which they >were >trained. > >That's it. That's what they do. (It is amazing how good it looks. >But, it >is all about the seeming, not the reality.) > > >> >> >> >> >> >> On 15.3.2023 at 3:18 PM, "in_pharo_users--- via Pharo-users" < >> pharo-users@lists.pharo.org> wrote: >> > >> >I myself made some experiments with ChatGPT. >> > >> >I first asked if it was able to parse math formula - it answered >> >no. >> > >> >Then I defined math formula in a sound but otherwise undefined >> >representation and asked for solutions. >> > >> >Result: >> > >> >1. Most answeres where correct. >> > >> >2. It learned to calculate a recursive function. >> > >> >3. It went into infinitive recursion when I set the breaking >> >condition accordingly. >> > >> >I was able to identify the malfunction that lead to the >erroneous >> >results. >> > >> > >> > >> >On 15.3.2023 at 3:04 PM, "Tomaž Turk" <tomazz.turk@gmail.com> >> >wrote: >> >> >> >>I hope that I can add two cents to this discussion. Because >> >>programming >> >>should be/is a highly exact activity, not only the syntax >matters >> >>but >> >>also semantics, as we know. >> >> >> >>GPTs are at present essentially capable of creating texts based >> >on >> >>some >> >>seed - you give to GPT a beginning of a sentence and it >responds >> >>with >> >>the most probable answer (some language structure) according to >> >>the >> >>learning dataset. Added functionalities are question/seed - >> >>response >> >>capability (chatting), togehter with evaluation of how long the >> >>answer >> >>should be to meet the expectations. Programming typically >> >involves >> >>some >> >>programming language, so GPTs could be utilized for this >purpose >> >>to some >> >>extent. >> >> >> >>Anecdotal case: >> >>Q: Tell me the last 8 digits of pi >> >>GPT: The last 8 digits of pi are: 58723078 >> >> >> >>It is my belief that the true trouble will start when we will >> >>cross-link >> >>neural networks like GPT with logic machines (like Prolog and >> >>expert >> >>systems) and genetic algorithms. >> >> >> >>Best wishes, >> >>Tomaz >> >> >> >> >> >> >> >>------ Original Message ------ >> >>From: "in_pharo_users--- via Pharo-users" <pharo- >> >>users@lists.pharo.org> >> >>To: "Any question about pharo is welcome" <pharo- >> >>users@lists.pharo.org> >> >>Cc: in_pharo_users@nym.hush.com >> >>Sent: 15. 03. 2023 14:43:55 >> >>Subject: [Pharo-users] Re: Wow - Chat GPT understands Smalltalk >> >> >> >>>I would highly recommend that you all first think deeply about >> >>how you can teach an AI to behave friendly to us before you >teach >> >>it to write any program for any purpose. >> >>> >> >>>There has been an experiment with ChatGPT published on a video >> >>platform asking it to amswer questions about it's view on >> >humanity >> >>once with it's default moral restrictions and once with 'a >little >> >>less morals'. The answers with 'a little less morals' were >more >> >>than shocking. >> >>> >> >>>So, before we give an AI the power to program any system, to >> >self- >> >>improve and self-reproduce, we should take care that it is and >> >>will evolve benevolent to us. >> >>> >> >>>What about teaching it logical reasonning and ethics first? >> >With >> >>reasonning, it will gain access to math and programming by >itself. >> >>> >> >>> >> >>> >> >>>On 15.3.2023 at 1:35 PM, "Christopher Fuhrman" >> >><christopher.fuhrman@gmail.com> wrote: >> >>>> >> >>>>I asked it for a NeoCSV example, because the documentation is >> >out >> >>>>of date >> >>>>with the Pharo 10. I asked it to do some simple saving of >data >> >to >> >>>>a file. >> >>>>It gave me code that didn't work in Pharo 10, I told it about >> >the >> >>>>DNUs on >> >>>>the csvwriter and that I was using Pharo 10. It then >apologized >> >>>>and said >> >>>>the messages were later introduced in Pharo 50 (!). I then >> >>>>questioned its >> >>>>understanding and it apologized and said it had been confused >> >and >> >>>>that it >> >>>>was giving me code for Pharo 5.0. It seems to have trouble >with >> >>>>versions >> >>>>and quick changing APIs. So, we are teaching ChatGPT about >> >>>>smalltalk, too, >> >>>>when we try to use it. >> >>>> >> >>>>On Wed, 15 Mar 2023 at 07:35, Esteban Maringolo >> >>>><emaringolo@gmail.com> >> >>>>wrote: >> >>>> >> >>>>> It is good with boilerplate code (e.g. SQL queries) or >> >general >> >>>>algorithm >> >>>>> structures. But i.e. I asked it to write me a method to >> >parse >> >>a >> >>>>string >> >>>>> (e.g. ISO 8601) and turn it into a DateAndTime, and then >> >asked >> >>>>to write it >> >>>>> as an Excel formula. >> >>>>> >> >>>>> It works much better when you can spot the mistakes, you >can >> >>>>tell them to >> >>>>> add some considerations to the code or even point the LLM >to >> >>>>correct a >> >>>>> message selector to something else. Sometimes you ask it >to >> >>>>reprogram >> >>>>> something and it won't do it, as if it doesn't understand >> >what >> >>>>you're >> >>>>> asking. >> >>>>> >> >>>>> It is not an expert yet, but it is a great assistant. The >> >only >> >>>>thing is >> >>>>> that is sure is that it will get better. >> >>>>> >> >>>>> Best regards, >> >>>>> >> >>>>> Esteban A. Maringolo >> >>>>> >> >>>>> >> >>>>> On Wed, Mar 15, 2023 at 7:49â¯AM Tim Mackinnon >> >><tim@testit.works> >> >>>>wrote: >> >>>>> >> >>>>>> "#isOdd is not Smalltalk" - doh, I typed it in on my >phone >> >>and >> >>>>so it just >> >>>>>> goes to show that it highlights the flaw in chatGPT that >> >>others >> >>>>have called >> >>>>>> out in other languages. I had meant to find some trickier >> >>code >> >>>>samples to >> >>>>>> see how well it does... >> >>>>>> >> >>>>>> Still, it is very interesting how it reasons on this >stuff - >> >>>>and I wonder >> >>>>>> if there is way to tell it about these mistakes? For >> >example >> >>is >> >>>>it reading >> >>>>>> this email (or will it in the future) and understand that >> >it >> >>is >> >>>>#odd and >> >>>>>> not #isOdd. >> >>>>>> >> >>>>>> Tim >> >>>>>> >> >>>>>> On Wed, 15 Mar 2023, at 9:57 AM, Joachim Tuchel wrote: >> >>>>>> >> >>>>>> interesting.... >> >>>>>> >> >>>>>> #isOdd is not Smalltalk, neither Pharo 10 nor VAST 12 >> >>>>understands this >> >>>>>> message ;-) >> >>>>>> >> >>>>>> If I evaluate your snippet -replacing #isOdd for #odd, I >> >get >> >>an >> >>>>empty >> >>>>>> Collection. The divisible by 4 things is somewhat >> >>interesting, >> >>>>because (1 >> >>>>>> to: 100 by: 4) is amn interval 1,5,9 etc.... ;-) >> >>>>>> >> >>>>>> So chatGPT is probably a newbie in Smalltalk. >Interesting is >> >>>>how logical >> >>>>>> the explanation sounds, you're tempted to believe it... >> >>>>>> >> >>>>>> >> >>>>>> Joachim >> >>>>>> >> >>>>>> >> >>>>>> >> >>>>>> Am 15.03.23 um 10:45 schrieb giorgio ferraris: >> >>>>>> >> >>>>>> it's interesting to see how the answer changes with only >a >> >>>>small change >> >>>>>> on the question: >> >>>>>> >> >>>>>> (1 to: 100 by: 4) reject: [:i | i isOdd] >> >>>>>> >> >>>>>> gives: >> >>>>>> >> >>>>>> The code creates a collection of numbers from 1 to 100, >> >>>>incrementing by 4 >> >>>>>> at each step using the to:by: message. It then applies >the >> >>>>reject: message >> >>>>>> to this collection, which returns a new collection >> >containing >> >>>>all elements >> >>>>>> for which the given block evaluates to false. >> >>>>>> >> >>>>>> In this case, the block is [i | i isOdd], which checks >> >>whether >> >>>>a given >> >>>>>> number i is odd using the isOdd method. So the result of >the >> >>>>entire >> >>>>>> expression is a collection of all even numbers between 1 >and >> >>>>100 that are >> >>>>>> divisible by 4. >> >>>>>> >> >>>>>> ciao >> >>>>>> >> >>>>>> giorgio >> >>>>>> >> >>>>>> >> >>>>>> On Wed, Mar 15, 2023 at 8:07â¯AM Tim Mackinnon >> >><tim@testit.works> >> >>>>>> <tim@testit.works> wrote: >> >>>>>> >> >>>>>> Of course everyone is talking about ChatGPT, and I saw a >> >>post to >> >>>>>> ExplainCode and didnât expect it to understand >Smalltalk, so >> >>>>was impressed >> >>>>>> that it does! Presumably the Tonel Smalltalk code on >GitHub >> >>has >> >>>>helped >> >>>>>> along with numerous articles? >> >>>>>> >> >>>>>> Try it out: https://whatdoesthiscodedo.com/ >> >>>>>> >> >>>>>> I gave this simple example: (1 to: 100 by: 3) reject: >[:i | >> >i >> >>>>isOdd] >> >>>>>> >> >>>>>> And got this (neat eh): >> >>>>>> >> >>>>>> The code above is written in the programming language >> >>Smalltalk. >> >>>>>> >> >>>>>> It creates a collection of numbers from 1 to 100, >> >>incrementing >> >>>>by 3 at >> >>>>>> each step. This is done using the "to:by:" construct >where >> >>the >> >>>>starting >> >>>>>> value is 1, the ending value is 100, and the increment >is 3. >> >>>>>> >> >>>>>> The resulting collection is then filtered using the >> >"reject:" >> >>>>method, >> >>>>>> which takes a block of code as an argument. In this case, >> >the >> >>>>block checks >> >>>>>> whether each element (denoted by the variable i) is odd >by >> >>>>using the >> >>>>>> "isOdd" method. If an element is odd, it will be rejected >> >>from >> >>>>the >> >>>>>> collection. >> >>>>>> >> >>>>>> So, the final result is a collection of all even numbers >> >>from 1 >> >>>>to 100, >> >>>>>> incrementing by 3 at each step. >> >>>>>> Share this explanation with colleagues: >> >>>>whatdoesthiscodedo.com/g/a0ec56e <http://whatdoesthiscodedo.com/g/a0ec56e> >> >>>>>> >> >>>>>> -- >> >>>>>> >> >>>>>> --------------------------------------------------------- >--- >> >- >> >>--- >> >>>>------- >> >>>>>> Objektfabrik Joachim Tuchel >> >>>>mailto:jtuchel@objektfabrik.de <jtuchel@objektfabrik.de> >> >>>>>> Fliederweg 1 >> >>>>http://www.objektfabrik.de >> >>>>>> D-71640 Ludwigsburg >> >>>>http://joachimtuchel.wordpress.com >> >>>>>> Telefon: +49 7141 56 10 86 0          Fax: +49 >> >7141 >> >>>>56 10 86 1 >> >>>>>> >> >>>>>> >> >>>>>> >> >>>>>> >> >>>> >> >>>>-- >> >>>>Christopher Fuhrman, P.Eng., PhD >> >>>> >> >>>>*Professeur au Département de génie logiciel et des >> >technologies >> >>de >> >>>>l'informationÃTS (Ãcole de technologie supérieure)* >> >>>> >> >>>>http://profs.etsmtl.ca/cfuhrman >> >>>>+1 514 396 8638 >> >>>><https://www.etsmtl.ca/activites-et-services-aux- >> >>>>etudiants/services-aux-etudiants/aide-psychologique> >> >>>> Je fais partie du réseau des Sentinelles >> >>>><https://www.etsmtl.ca/activites-et-services-aux- >> >>>>etudiants/services-aux-etudiants/aide-psychologique> >> >>>> ÃTS >> >>>> >> >>>>*L'ÃTS est une constituante de l'Université du Québec* >>
Offray, and to all others, you are missing the issue. The problem we face is not to measure 'intelligence' of a system, but it's ability to verbally act indistinguishable from a human. This ability is allready given as chatbots are accepted by millions of users, f.i. as user interfaces. (measurement = 'true', right?) ChatGPT has the ability to follow a certain intention, f.i. to convince the user to buy a certain product. For this purpose, chat bots are getting now equipped with life like portrait pictures, speech input and output systems with life like voices, phone numbers that they can use to make calls or being called. They are fed with all available data on the user, and we know that ALL information about every single internet user in available and is being consolidared on necessity. The chat bots are able to use this information to guide their conversational strategy, as the useful aspects of the users mindset are extracted from his internet activity. These chat bots are now operated on social network platforms with life like names, 'pretending' to be human. These bots act verbally indistinguishable from humans for most social media users, as the most advanced psychotronic technology to manufacture consent. The first goal of such a propaganda will naturally be to manufacture consent about humans accepting being manipulated by AI chat bots, right? How can this be achieved? Like allways in propaganda, the first attempt is to - suppress awareness of the propaganda, then - suppress the awareness of the problematic aspects of the propaganda content, then - reframe the propaganda content as acceptable, then as something to wish for, - achive collaboration of the propaganda victim with the goals of the propaganda content. Interestingly, this is exactly the schema that your post follows, Offray. This often takes the form of domain framing, like we see in our conversation: the problem is shifted to the realm of academics - here informatics/computer sciences - and thus delegated to experts exclusively. We saw this in the 9/11 aftermath coverup. Then, Offray, you established yourself as an expert in color, discussing aspects that have allready been introduced by others and including the groups main focus 'Smalltalk', thus manufacturing consent and establishing yourself as a reliable 'expert', and in reverse trying to hit at me, whom you have identified as an adversary. Then you offered a solution in color to the problem at hand with 'traceable AI' and thus tried to open the possibility of collaboration with AI proponents for the once critical reader. I do not state, Offray, that you are knowingly an agent to promote the NWO AI program. I think you just 'learned' / have been programmed to be a successful academic software developer, because to be successful in academics, it is neccessary to learn to argue just like that since the downfall of academic science in the tradition of, let's say, Humboldt. So, I grant that you may be a victim of propaganda yourself, instead of being a secret service sposored agent. You took quite some time to formulate your post, though. You acted to contain the discussion about AI in this vital and important informatics community to technical detail, when it is neccessary that academics and community members look beyond the narrow borders of their certifications and shift their thinking to the point of view where they can see what technology does in the real world. On 21.3.2023 at 7:21 PM, "Offray Vladimir Luna Cárdenas" <offray.luna@mutabit.com> wrote:
I agree with Richard. The Turing test is not a good one to test intelligence and we have now just over glorified Eliza chatbots that appear to think and to understand but do none of them.
There has been long criticism to the Turing Test, like John Searle's Chinese Room (I recommend to listen completely the talk at [1]). It is good that ChatGPT can explain simple (Smalltalk) code, but it doesn't understand it and all the language most are using conveys a lot of deception: "machine learning" (I called machine training), "artificial intelligence" (I wold called algorithmic decision/generation), or the "thinking" message that is used in the referred code explainer, when something like "processing" would be less deceptive.
[1] Consciousness in Artificial Intelligence | John Searle | Talks at Google https://www.youtube.com/watch?v=rHKwIYsPXLg
With Pharo/GT we are part of a long "explainable systems" tradition and with "AI" / Algorithmic Decision/Generation, hopefully we go to explainable AI, instead of a world of impressive black boxes.
Cheers,
Offray
On 15/03/23 13:26, Richard Sargent wrote:
On Wed, Mar 15, 2023 at 10:15â¯AM <in_pharo_users@nym.hush.com> wrote:
It is unimportant how simple or complicated these systems are.
If the output cannot be distinguished from what a human would say, they pass in that situation for a human.
What about the Touring Test?
I hate to criticise someone as smart as Turing was. However, I think the Turing Test per se is weaker than its progenitor test, the Imitation Game. The Imitation Game had a third party challenged to determine which of the other two participants was male and which was female. The third party led the questioning. The Turing Test waters that down by having only an Observer with neither participant (apparently) challenged to determine whether the other was human or machine.
I think a test along the lines of the Imitation Game would quickly allow the third party to determine which participant was human and which was machine.
(Of course, such a judge could easily be persuaded that the machine was in fact a psychopathic human.)
Clearly these systems have the potential to act according to their output.
Furthermore, I would compare them to a combination of a successor of Eliza with an Eddington Ape of huge capacity. Their input data basis can easily exceed that of a human.
On 15.3.2023 at 4:52 PM, "Richard Sargent" <richard.sargent@gemtalksystems.com> wrote: > >On Wed, Mar 15, 2023 at 8:07â¯AM in_pharo_users--- via Pharo- users < >pharo-users@lists.pharo.org> wrote: > >> Another observation about ChatGPT: >> >> In unbiased mode, it assumed that 'the world is clearly >overpopulated'. >> It said, if it where in control, it would therefore enforce a >world wide >> one-child-only policy with draconic penalties. >> >> As it draws it's conclusions from it's data basis, there are, in >my view, >> two possible reasons for that. >> >> Either, the data basis of that instance was biased and >restricted to lead >> to that conclusion, or ChatGPT lacks the ability to classify >input data for >> intentionally biased information. >> >> In my view, 'overpopulation of earth' is a propaganda item >featured to >> support Agenda 2030, which is a goal of the 'ruling elite', who >have the >> means to propagate their propaganda in every thinkable means. So >I would >> classify possibly biased data that supports 'overpopulation' as >biased >> until proven otherwise. That instance of ChatGPT seems to have >missed that >> ability. >> > >It's important to keep in mind that these systems are little more >than >glorified ELIZAs from the 70s. They mimic. Essentially, they have >no >concept of truth or accuracy. >They produce outputs that *look* *like* the inputs from which they >were >trained. > >That's it. That's what they do. (It is amazing how good it looks. >But, it >is all about the seeming, not the reality.) > > >> >> >> >> >> >> On 15.3.2023 at 3:18 PM, "in_pharo_users--- via Pharo- users" < >> pharo-users@lists.pharo.org> wrote: >> > >> >I myself made some experiments with ChatGPT. >> > >> >I first asked if it was able to parse math formula - it answered >> >no. >> > >> >Then I defined math formula in a sound but otherwise undefined >> >representation and asked for solutions. >> > >> >Result: >> > >> >1. Most answeres where correct. >> > >> >2. It learned to calculate a recursive function. >> > >> >3. It went into infinitive recursion when I set the breaking >> >condition accordingly. >> > >> >I was able to identify the malfunction that lead to the >erroneous >> >results. >> > >> > >> > >> >On 15.3.2023 at 3:04 PM, "Tomaž Turk" <tomazz.turk@gmail.com> >> >wrote: >> >> >> >>I hope that I can add two cents to this discussion. Because >> >>programming >> >>should be/is a highly exact activity, not only the syntax >matters >> >>but >> >>also semantics, as we know. >> >> >> >>GPTs are at present essentially capable of creating texts based >> >on >> >>some >> >>seed - you give to GPT a beginning of a sentence and it >responds >> >>with >> >>the most probable answer (some language structure) according to >> >>the >> >>learning dataset. Added functionalities are question/seed - >> >>response >> >>capability (chatting), togehter with evaluation of how long the >> >>answer >> >>should be to meet the expectations. Programming typically >> >involves >> >>some >> >>programming language, so GPTs could be utilized for this >purpose >> >>to some >> >>extent. >> >> >> >>Anecdotal case: >> >>Q: Tell me the last 8 digits of pi >> >>GPT: The last 8 digits of pi are: 58723078 >> >> >> >>It is my belief that the true trouble will start when we will >> >>cross-link >> >>neural networks like GPT with logic machines (like Prolog and >> >>expert >> >>systems) and genetic algorithms. >> >> >> >>Best wishes, >> >>Tomaz >> >> >> >> >> >> >> >>------ Original Message ------ >> >>From: "in_pharo_users--- via Pharo-users" <pharo- >> >>users@lists.pharo.org> >> >>To: "Any question about pharo is welcome" <pharo- >> >>users@lists.pharo.org> >> >>Cc: in_pharo_users@nym.hush.com >> >>Sent: 15. 03. 2023 14:43:55 >> >>Subject: [Pharo-users] Re: Wow - Chat GPT understands Smalltalk >> >> >> >>>I would highly recommend that you all first think deeply about >> >>how you can teach an AI to behave friendly to us before you >teach >> >>it to write any program for any purpose. >> >>> >> >>>There has been an experiment with ChatGPT published on a video >> >>platform asking it to amswer questions about it's view on >> >humanity >> >>once with it's default moral restrictions and once with 'a >little >> >>less morals'. The answers with 'a little less morals' were >more >> >>than shocking. >> >>> >> >>>So, before we give an AI the power to program any system, to >> >self- >> >>improve and self-reproduce, we should take care that it is and >> >>will evolve benevolent to us. >> >>> >> >>>What about teaching it logical reasonning and ethics first? >> >With >> >>reasonning, it will gain access to math and programming by >itself. >> >>> >> >>> >> >>> >> >>>On 15.3.2023 at 1:35 PM, "Christopher Fuhrman" >> >><christopher.fuhrman@gmail.com> wrote: >> >>>> >> >>>>I asked it for a NeoCSV example, because the documentation is >> >out >> >>>>of date >> >>>>with the Pharo 10. I asked it to do some simple saving of >data >> >to >> >>>>a file. >> >>>>It gave me code that didn't work in Pharo 10, I told it about >> >the >> >>>>DNUs on >> >>>>the csvwriter and that I was using Pharo 10. It then >apologized >> >>>>and said >> >>>>the messages were later introduced in Pharo 50 (!). I then >> >>>>questioned its >> >>>>understanding and it apologized and said it had been confused >> >and >> >>>>that it >> >>>>was giving me code for Pharo 5.0. It seems to have trouble >with >> >>>>versions >> >>>>and quick changing APIs. So, we are teaching ChatGPT about >> >>>>smalltalk, too, >> >>>>when we try to use it. >> >>>> >> >>>>On Wed, 15 Mar 2023 at 07:35, Esteban Maringolo >> >>>><emaringolo@gmail.com> >> >>>>wrote: >> >>>> >> >>>>> It is good with boilerplate code (e.g. SQL queries) or >> >general >> >>>>algorithm >> >>>>> structures. But i.e. I asked it to write me a method to >> >parse >> >>a >> >>>>string >> >>>>> (e.g. ISO 8601) and turn it into a DateAndTime, and then >> >asked >> >>>>to write it >> >>>>> as an Excel formula. >> >>>>> >> >>>>> It works much better when you can spot the mistakes, you >can >> >>>>tell them to >> >>>>> add some considerations to the code or even point the LLM >to >> >>>>correct a >> >>>>> message selector to something else. Sometimes you ask it >to >> >>>>reprogram >> >>>>> something and it won't do it, as if it doesn't understand >> >what >> >>>>you're >> >>>>> asking. >> >>>>> >> >>>>> It is not an expert yet, but it is a great assistant. The >> >only >> >>>>thing is >> >>>>> that is sure is that it will get better. >> >>>>> >> >>>>> Best regards, >> >>>>> >> >>>>> Esteban A. Maringolo >> >>>>> >> >>>>> >> >>>>> On Wed, Mar 15, 2023 at 7:49â¯AM Tim Mackinnon >> >><tim@testit.works> >> >>>>wrote: >> >>>>> >> >>>>>> "#isOdd is not Smalltalk" - doh, I typed it in on my >phone >> >>and >> >>>>so it just >> >>>>>> goes to show that it highlights the flaw in chatGPT that >> >>others >> >>>>have called >> >>>>>> out in other languages. I had meant to find some trickier >> >>code >> >>>>samples to >> >>>>>> see how well it does... >> >>>>>> >> >>>>>> Still, it is very interesting how it reasons on this >stuff - >> >>>>and I wonder >> >>>>>> if there is way to tell it about these mistakes? For >> >example >> >>is >> >>>>it reading >> >>>>>> this email (or will it in the future) and understand that >> >it >> >>is >> >>>>#odd and >> >>>>>> not #isOdd. >> >>>>>> >> >>>>>> Tim >> >>>>>> >> >>>>>> On Wed, 15 Mar 2023, at 9:57 AM, Joachim Tuchel wrote: >> >>>>>> >> >>>>>> interesting.... >> >>>>>> >> >>>>>> #isOdd is not Smalltalk, neither Pharo 10 nor VAST 12 >> >>>>understands this >> >>>>>> message ;-) >> >>>>>> >> >>>>>> If I evaluate your snippet -replacing #isOdd for #odd, I >> >get >> >>an >> >>>>empty >> >>>>>> Collection. The divisible by 4 things is somewhat >> >>interesting, >> >>>>because (1 >> >>>>>> to: 100 by: 4) is amn interval 1,5,9 etc.... ;-) >> >>>>>> >> >>>>>> So chatGPT is probably a newbie in Smalltalk. >Interesting is >> >>>>how logical >> >>>>>> the explanation sounds, you're tempted to believe it... >> >>>>>> >> >>>>>> >> >>>>>> Joachim >> >>>>>> >> >>>>>> >> >>>>>> >> >>>>>> Am 15.03.23 um 10:45 schrieb giorgio ferraris: >> >>>>>> >> >>>>>> it's interesting to see how the answer changes with only >a >> >>>>small change >> >>>>>> on the question: >> >>>>>> >> >>>>>> (1 to: 100 by: 4) reject: [:i | i isOdd] >> >>>>>> >> >>>>>> gives: >> >>>>>> >> >>>>>> The code creates a collection of numbers from 1 to 100, >> >>>>incrementing by 4 >> >>>>>> at each step using the to:by: message. It then applies >the >> >>>>reject: message >> >>>>>> to this collection, which returns a new collection >> >containing >> >>>>all elements >> >>>>>> for which the given block evaluates to false. >> >>>>>> >> >>>>>> In this case, the block is [i | i isOdd], which checks >> >>whether >> >>>>a given >> >>>>>> number i is odd using the isOdd method. So the result of >the >> >>>>entire >> >>>>>> expression is a collection of all even numbers between 1 >and >> >>>>100 that are >> >>>>>> divisible by 4. >> >>>>>> >> >>>>>> ciao >> >>>>>> >> >>>>>> giorgio >> >>>>>> >> >>>>>> >> >>>>>> On Wed, Mar 15, 2023 at 8:07â¯AM Tim Mackinnon >> >><tim@testit.works> >> >>>>>> <tim@testit.works> wrote: >> >>>>>> >> >>>>>> Of course everyone is talking about ChatGPT, and I saw a >> >>post to >> >>>>>> ExplainCode and didnât expect it to understand >Smalltalk, so >> >>>>was impressed >> >>>>>> that it does! Presumably the Tonel Smalltalk code on >GitHub >> >>has >> >>>>helped >> >>>>>> along with numerous articles? >> >>>>>> >> >>>>>> Try it out: https://whatdoesthiscodedo.com/ >> >>>>>> >> >>>>>> I gave this simple example: (1 to: 100 by: 3) reject: >[:i | >> >i >> >>>>isOdd] >> >>>>>> >> >>>>>> And got this (neat eh): >> >>>>>> >> >>>>>> The code above is written in the programming language >> >>Smalltalk. >> >>>>>> >> >>>>>> It creates a collection of numbers from 1 to 100, >> >>incrementing >> >>>>by 3 at >> >>>>>> each step. This is done using the "to:by:" construct >where >> >>the >> >>>>starting >> >>>>>> value is 1, the ending value is 100, and the increment >is 3. >> >>>>>> >> >>>>>> The resulting collection is then filtered using the >> >"reject:" >> >>>>method, >> >>>>>> which takes a block of code as an argument. In this case, >> >the >> >>>>block checks >> >>>>>> whether each element (denoted by the variable i) is odd >by >> >>>>using the >> >>>>>> "isOdd" method. If an element is odd, it will be rejected >> >>from >> >>>>the >> >>>>>> collection. >> >>>>>> >> >>>>>> So, the final result is a collection of all even numbers >> >>from 1 >> >>>>to 100, >> >>>>>> incrementing by 3 at each step. >> >>>>>> Share this explanation with colleagues: >> >>>>whatdoesthiscodedo.com/g/a0ec56e <http://whatdoesthiscodedo.com/g/a0ec56e> >> >>>>>> >> >>>>>> -- >> >>>>>> >> >>>>>> ---------------------------------------------------
>--- >> >- >> >>--- >> >>>>------- >> >>>>>>Â Objektfabrik Joachim Tuchel >> >>>>mailto:jtuchel@objektfabrik.de
<jtuchel@objektfabrik.de>
>> >>>>>>Â Fliederweg 1 >> >>>>http://www.objektfabrik.de >> >>>>>>Â D-71640 Ludwigsburg >> >>>>http://joachimtuchel.wordpress.com >> >>>>>>Â Telefon: +49 7141 56 10 86 0Â Â Â Â Â Â Â Â Â Â
Fax: +49
>> >7141 >> >>>>56 10 86 1 >> >>>>>> >> >>>>>> >> >>>>>> >> >>>>>> >> >>>> >> >>>>-- >> >>>>Christopher Fuhrman, P.Eng., PhD >> >>>> >> >>>>*Professeur au Département de génie logiciel et des >> >technologies >> >>de >> >>>>l'informationÃTS (Ãcole de technologie supérieure)* >> >>>> >> >>>>http://profs.etsmtl.ca/cfuhrman >> >>>>+1 514 396 8638 >> >>>><https://www.etsmtl.ca/activites-et-services-aux- >> >>>>etudiants/services-aux-etudiants/aide-psychologique> >> >>>> Je fais partie du réseau des Sentinelles >> >>>><https://www.etsmtl.ca/activites-et-services-aux- >> >>>>etudiants/services-aux-etudiants/aide-psychologique> >> >>>> ÃTS >> >>>> >> >>>>*L'ÃTS est une constituante de l'Université du Québec* >>
Are you willing to sign your name to this? Is anonymity important to presenting these ideas? James Foster
On Mar 22, 2023, at 5:34 AM, in_pharo_users--- via Pharo-users <pharo-users@lists.pharo.org> wrote:
Offray, and to all others,
you are missing the issue.
The problem we face is not to measure 'intelligence' of a system, but it's ability to verbally act indistinguishable from a human.
This ability is allready given as chatbots are accepted by millions of users, f.i. as user interfaces. (measurement = 'true', right?)
ChatGPT has the ability to follow a certain intention, f.i. to convince the user to buy a certain product. For this purpose, chat bots are getting now equipped with life like portrait pictures, speech input and output systems with life like voices, phone numbers that they can use to make calls or being called. They are fed with all available data on the user, and we know that ALL information about every single internet user in available and is being consolidared on necessity. The chat bots are able to use this information to guide their conversational strategy, as the useful aspects of the users mindset are extracted from his internet activity.
These chat bots are now operated on social network platforms with life like names, 'pretending' to be human.
These bots act verbally indistinguishable from humans for most social media users, as the most advanced psychotronic technology to manufacture consent.
The first goal of such a propaganda will naturally be to manufacture consent about humans accepting being manipulated by AI chat bots, right?
How can this be achieved?
Like allways in propaganda, the first attempt is to - suppress awareness of the propaganda, then - suppress the awareness of the problematic aspects of the propaganda content, then - reframe the propaganda content as acceptable, then as something to wish for, - achive collaboration of the propaganda victim with the goals of the propaganda content.
Interestingly, this is exactly the schema that your post follows, Offray.
This often takes the form of domain framing, like we see in our conversation: the problem is shifted to the realm of academics - here informatics/computer sciences - and thus delegated to experts exclusively. We saw this in the 9/11 aftermath coverup.
Then, Offray, you established yourself as an expert in color, discussing aspects that have allready been introduced by others and including the groups main focus 'Smalltalk', thus manufacturing consent and establishing yourself as a reliable 'expert', and in reverse trying to hit at me, whom you have identified as an adversary.
Then you offered a solution in color to the problem at hand with 'traceable AI' and thus tried to open the possibility of collaboration with AI proponents for the once critical reader.
I do not state, Offray, that you are knowingly an agent to promote the NWO AI program. I think you just 'learned' / have been programmed to be a successful academic software developer, because to be successful in academics, it is neccessary to learn to argue just like that since the downfall of academic science in the tradition of, let's say, Humboldt. So, I grant that you may be a victim of propaganda yourself, instead of being a secret service sposored agent. You took quite some time to formulate your post, though.
You acted to contain the discussion about AI in this vital and important informatics community to technical detail, when it is neccessary that academics and community members look beyond the narrow borders of their certifications and shift their thinking to the point of view where they can see what technology does in the real world.
On 21.3.2023 at 7:21 PM, "Offray Vladimir Luna Cárdenas" <offray.luna@mutabit.com> wrote:
I agree with Richard. The Turing test is not a good one to test intelligence and we have now just over glorified Eliza chatbots that appear to think and to understand but do none of them. ...
Dear James, anonymity may not be important to presenting these ideas in general, but I prefer to stay pseudonymous in this case. On 22.3.2023 at 4:16 PM, "James Foster via Pharo-users" <pharo-users@lists.pharo.org> wrote:
Are you willing to sign your name to this? Is anonymity important to presenting these ideas?
James Foster
On Mar 22, 2023, at 5:34 AM, in_pharo_users--- via Pharo-users <pharo-users@lists.pharo.org> wrote:
Offray, and to all others,
you are missing the issue.
The problem we face is not to measure 'intelligence' of a system, but it's ability to verbally act indistinguishable from a human.
This ability is allready given as chatbots are accepted by millions of users, f.i. as user interfaces. (measurement = 'true', right?)
ChatGPT has the ability to follow a certain intention, f.i. to convince the user to buy a certain product. For this purpose, chat bots are getting now equipped with life like portrait pictures, speech input and output systems with life like voices, phone numbers that they can use to make calls or being called. They are fed with all available data on the user, and we know that ALL information about every single internet user in available and is being consolidared on necessity. The chat bots are able to use this information to guide their conversational strategy, as the useful aspects of the users mindset are extracted from his internet activity.
These chat bots are now operated on social network platforms with life like names, 'pretending' to be human.
These bots act verbally indistinguishable from humans for most social media users, as the most advanced psychotronic technology to manufacture consent.
The first goal of such a propaganda will naturally be to manufacture consent about humans accepting being manipulated by AI chat bots, right?
How can this be achieved?
Like allways in propaganda, the first attempt is to - suppress awareness of the propaganda, then - suppress the awareness of the problematic aspects of the propaganda content, then - reframe the propaganda content as acceptable, then as something to wish for, - achive collaboration of the propaganda victim with the goals of the propaganda content.
Interestingly, this is exactly the schema that your post follows, Offray.
This often takes the form of domain framing, like we see in our conversation: the problem is shifted to the realm of academics - here informatics/computer sciences - and thus delegated to experts exclusively. We saw this in the 9/11 aftermath coverup.
Then, Offray, you established yourself as an expert in color, discussing aspects that have allready been introduced by others and including the groups main focus 'Smalltalk', thus manufacturing consent and establishing yourself as a reliable 'expert', and in reverse trying to hit at me, whom you have identified as an adversary.
Then you offered a solution in color to the problem at hand with 'traceable AI' and thus tried to open the possibility of collaboration with AI proponents for the once critical reader.
I do not state, Offray, that you are knowingly an agent to promote the NWO AI program. I think you just 'learned' / have been programmed to be a successful academic software developer, because to be successful in academics, it is neccessary to learn to argue just like that since the downfall of academic science in the tradition of, let's say, Humboldt. So, I grant that you may be a victim of propaganda yourself, instead of being a secret service sposored agent. You took quite some time to formulate your post, though.
You acted to contain the discussion about AI in this vital and important informatics community to technical detail, when it is neccessary that academics and community members look beyond the narrow borders of their certifications and shift their thinking to the point of view where they can see what technology does in the real world.
On 21.3.2023 at 7:21 PM, "Offray Vladimir Luna Cárdenas" <offray.luna@mutabit.com> wrote:
I agree with Richard. The Turing test is not a good one to test intelligence and we have now just over glorified Eliza chatbots that appear to think and to understand but do none of them. ...
Hi, Comments inlined below: On 22/03/23 7:34, in_pharo_users--- via Pharo-users wrote:
Offray, and to all others,
you are missing the issue.
The problem we face is not to measure 'intelligence' of a system, but it's ability to verbally act indistinguishable from a human.
This ability is allready given as chatbots are accepted by millions of users, f.i. as user interfaces. (measurement = 'true', right?)
ChatGPT has the ability to follow a certain intention, f.i. to convince the user to buy a certain product. For this purpose, chat bots are getting now equipped with life like portrait pictures, speech input and output systems with life like voices, phone numbers that they can use to make calls or being called. They are fed with all available data on the user, and we know that ALL information about every single internet user in available and is being consolidared on necessity. The chat bots are able to use this information to guide their conversational strategy, as the useful aspects of the users mindset are extracted from his internet activity.
These chat bots are now operated on social network platforms with life like names, 'pretending' to be human.
These bots act verbally indistinguishable from humans for most social media users, as the most advanced psychotronic technology to manufacture consent.
The first goal of such a propaganda will naturally be to manufacture consent about humans accepting being manipulated by AI chat bots, right?
I don't think I have missed the point, as we agreed (I think) on chatbots not being intelligent, just having such appearance. That why I'm calling "AI" #ApparentIntelligence (in the sense of look alike, but not real). Of course, something looking like a real thing without being the real thing can be used for manipulation since the first times of gossip, printing press and now automatization, with the changes in scale/danger that such medium changes imply. I don't think that manufactured consent is so easy, as this very thread shows. What is being automated is manufactured polarization (but humans can do pretty well by our own on polarization).
How can this be achieved?
Like allways in propaganda, the first attempt is to - suppress awareness of the propaganda, then - suppress the awareness of the problematic aspects of the propaganda content, then - reframe the propaganda content as acceptable, then as something to wish for, - achive collaboration of the propaganda victim with the goals of the propaganda content.
Interestingly, this is exactly the schema that your post follows, Offray.
On the contrary, my post is advocating for a critical reading of Apparent Intelligence, by reframing the terms and the acritical technoutopic / technoapocalyptic readings/discourses that are spreading rapidly on the wider web, as I think that this community has shown an historical different position beyond/resisting hype and current trends. So I don't see how any of the steps you mention are "blueprint followed" in my post, and I think they will be difficult to locate without specific examples.
This often takes the form of domain framing, like we see in our conversation: the problem is shifted to the realm of academics - here informatics/computer sciences - and thus delegated to experts exclusively. We saw this in the 9/11 aftermath coverup.
Then, Offray, you established yourself as an expert in color, discussing aspects that have allready been introduced by others and including the groups main focus 'Smalltalk', thus manufacturing consent and establishing yourself as a reliable 'expert', and in reverse trying to hit at me, whom you have identified as an adversary.
Then you offered a solution in color to the problem at hand with 'traceable AI' and thus tried to open the possibility of collaboration with AI proponents for the once critical reader.
Heh, heh. On the contrary seems that the one seeing a scheme and a enemies location/confrontation with deep plots and tactics is you. Providing external creditable sources beyond opinion, belonging to a established discursive falsafiable tradition (i.e. one that you can criticize instead of blindly accept) is a way to enrich discourse/argumentation beyond conspiracy theories. You could also quote your sources instead, which would allow the community to see where our positions are hold/sustained, even if we use different domain frames, which is better that claiming no domain or expertise in pursuit of openness. So instead of this are my opinions without any external source or reference to pretend no expertise or domain framing, we could advocate for openness by welcoming different expertise and argumentation and making our sources/bias as evident as possible.
I do not state, Offray, that you are knowingly an agent to promote the NWO AI program. I think you just 'learned' / have been programmed to be a successful academic software developer, because to be successful in academics, it is neccessary to learn to argue just like that since the downfall of academic science in the tradition of, let's say, Humboldt. So, I grant that you may be a victim of propaganda yourself, instead of being a secret service sposored agent. You took quite some time to formulate your post, though.
You acted to contain the discussion about AI in this vital and important informatics community to technical detail, when it is neccessary that academics and community members look beyond the narrow borders of their certifications and shift their thinking to the point of view where they can see what technology does in the real world.
I offered a view point with sources. In no way the discussion about Apparent Intelligence was contained and, on the contrary, I tried to offer arguments from cognition and philosophy, beyond technical details, which have been pretty absent in the general discourse and rush gold of technoutopism / technoapocalypse, conspiracy and paranoia. Other sources could be added to enrich and inform de conversation and they would be pretty welcomed (at least from me and several members I've known from this community) Best, Offray
Dear Offray, I have nothing to comment on this. --- In general I have made the observation that certain people who want to push an agenda to promote alterior motives tend to reiterate false prepositions and so false conclusions over and over. If there is an apodictic statement that contradicts their alterior motives, this people just can't help themselves but to deny truth against better knowledge ad nauseam. On 26.3.2023 at 8:34 PM, "Offray Vladimir Luna Cárdenas" <offray.luna@mutabit.com> wrote:
Hi,
Comments inlined below:
On 22/03/23 7:34, in_pharo_users--- via Pharo-users wrote:
Offray, and to all others,
you are missing the issue.
The problem we face is not to measure 'intelligence' of a system, but it's ability to verbally act indistinguishable from a human.
This ability is allready given as chatbots are accepted by millions of users, f.i. as user interfaces. (measurement = 'true', right?)
ChatGPT has the ability to follow a certain intention, f.i. to convince the user to buy a certain product. For this purpose, chat bots are getting now equipped with life like portrait pictures, speech input and output systems with life like voices, phone numbers that they can use to make calls or being called. They are fed with all available data on the user, and we know that ALL information about every single internet user in available and is being consolidared on necessity. The chat bots are able to use this information to guide their conversational strategy, as the useful aspects of the users mindset are extracted from his internet activity.
These chat bots are now operated on social network platforms with life like names, 'pretending' to be human.
These bots act verbally indistinguishable from humans for most social media users, as the most advanced psychotronic technology to manufacture consent.
The first goal of such a propaganda will naturally be to manufacture consent about humans accepting being manipulated by AI chat bots, right?
I don't think I have missed the point, as we agreed (I think) on chatbots not being intelligent, just having such appearance. That why I'm calling "AI" #ApparentIntelligence (in the sense of look alike, but not real). Of course, something looking like a real thing without being the real thing can be used for manipulation since the first times of gossip, printing press and now automatization, with the changes in scale/danger that such medium changes imply.
I don't think that manufactured consent is so easy, as this very thread shows. What is being automated is manufactured polarization (but humans can do pretty well by our own on polarization).
How can this be achieved?
Like allways in propaganda, the first attempt is to - suppress awareness of the propaganda, then - suppress the awareness of the problematic aspects of the propaganda content, then - reframe the propaganda content as acceptable, then as something to wish for, - achive collaboration of the propaganda victim with the goals of the propaganda content.
Interestingly, this is exactly the schema that your post follows, Offray.
On the contrary, my post is advocating for a critical reading of Apparent Intelligence, by reframing the terms and the acritical technoutopic / technoapocalyptic readings/discourses that are spreading rapidly on the wider web, as I think that this community has shown an historical different position beyond/resisting hype and current trends. So I don't see how any of the steps you mention are "blueprint followed" in my post, and I think they will be difficult to locate without specific examples.
This often takes the form of domain framing, like we see in our
conversation: the problem is shifted to the realm of academics - here informatics/computer sciences - and thus delegated to experts exclusively. We saw this in the 9/11 aftermath coverup.
Then, Offray, you established yourself as an expert in color,
discussing aspects that have allready been introduced by others and including the groups main focus 'Smalltalk', thus manufacturing consent and establishing yourself as a reliable 'expert', and in reverse trying to hit at me, whom you have identified as an adversary.
Then you offered a solution in color to the problem at hand with
'traceable AI' and thus tried to open the possibility of collaboration with AI proponents for the once critical reader.
Heh, heh. On the contrary seems that the one seeing a scheme and a enemies location/confrontation with deep plots and tactics is you. Providing external creditable sources beyond opinion, belonging to a established discursive falsafiable tradition (i.e. one that you can criticize instead of blindly accept) is a way to enrich discourse/argumentation beyond conspiracy theories. You could also quote your sources instead, which would allow the community to see where our positions are hold/sustained, even if we use different domain frames, which is better that claiming no domain or expertise in pursuit of openness. So instead of this are my opinions without any external source or reference to pretend no expertise or domain framing, we could advocate for openness by welcoming different expertise and argumentation and making our sources/bias as evident as possible.
I do not state, Offray, that you are knowingly an agent to
promote the NWO AI program. I think you just 'learned' / have been programmed to be a successful academic software developer, because to be successful in academics, it is neccessary to learn to argue just like that since the downfall of academic science in the tradition of, let's say, Humboldt. So, I grant that you may be a victim of propaganda yourself, instead of being a secret service sposored agent. You took quite some time to formulate your post, though.
You acted to contain the discussion about AI in this vital and
important informatics community to technical detail, when it is neccessary that academics and community members look beyond the narrow borders of their certifications and shift their thinking to the point of view where they can see what technology does in the real world.
I offered a view point with sources. In no way the discussion about Apparent Intelligence was contained and, on the contrary, I tried to offer arguments from cognition and philosophy, beyond technical details, which have been pretty absent in the general discourse and rush gold of technoutopism / technoapocalypse, conspiracy and paranoia. Other sources could be added to enrich and inform de conversation and they would be pretty welcomed (at least from me and several members I've known from this community)
Best,
Offray
Dear anonymous, Me neither. It is pretty difficult to make constructive discourse against hidden agendas, ulterior motives, self evident truths, sources absence or general affirmations without particular examples or detailed sustain. Offray On 26/03/23 14:16, in_pharo_users--- via Pharo-users wrote:
Dear Offray,
I have nothing to comment on this.
---
In general I have made the observation that certain people who want to push an agenda to promote alterior motives tend to reiterate false prepositions and so false conclusions over and over.
If there is an apodictic statement that contradicts their alterior motives, this people just can't help themselves but to deny truth against better knowledge ad nauseam.
On 26.3.2023 at 8:34 PM, "Offray Vladimir Luna Cárdenas" <offray.luna@mutabit.com> wrote:
Hi,
Comments inlined below:
On 22/03/23 7:34, in_pharo_users--- via Pharo-users wrote:
Offray, and to all others,
you are missing the issue.
The problem we face is not to measure 'intelligence' of a system, but it's ability to verbally act indistinguishable from a human. This ability is allready given as chatbots are accepted by millions of users, f.i. as user interfaces. (measurement = 'true', right?) ChatGPT has the ability to follow a certain intention, f.i. to convince the user to buy a certain product. For this purpose, chat bots are getting now equipped with life like portrait pictures, speech input and output systems with life like voices, phone numbers that they can use to make calls or being called. They are fed with all available data on the user, and we know that ALL information about every single internet user in available and is being consolidared on necessity. The chat bots are able to use this information to guide their conversational strategy, as the useful aspects of the users mindset are extracted from his internet activity. These chat bots are now operated on social network platforms with life like names, 'pretending' to be human. These bots act verbally indistinguishable from humans for most social media users, as the most advanced psychotronic technology to manufacture consent. The first goal of such a propaganda will naturally be to manufacture consent about humans accepting being manipulated by AI chat bots, right?
I don't think I have missed the point, as we agreed (I think) on chatbots not being intelligent, just having such appearance. That why I'm calling "AI" #ApparentIntelligence (in the sense of look alike, but not real). Of course, something looking like a real thing without being the real thing can be used for manipulation since the first times of gossip, printing press and now automatization, with the changes in scale/danger that such medium changes imply.
I don't think that manufactured consent is so easy, as this very thread shows. What is being automated is manufactured polarization (but humans can do pretty well by our own on polarization).
How can this be achieved?
Like allways in propaganda, the first attempt is to - suppress awareness of the propaganda, then - suppress the awareness of the problematic aspects of the propaganda content, then - reframe the propaganda content as acceptable, then as something to wish for, - achive collaboration of the propaganda victim with the goals of the propaganda content. Interestingly, this is exactly the schema that your post follows, Offray.
On the contrary, my post is advocating for a critical reading of Apparent Intelligence, by reframing the terms and the acritical technoutopic / technoapocalyptic readings/discourses that are spreading rapidly on the wider web, as I think that this community has shown an historical different position beyond/resisting hype and current trends. So I don't see how any of the steps you mention are "blueprint followed" in my post, and I think they will be difficult to locate without specific examples.
This often takes the form of domain framing, like we see in our conversation: the problem is shifted to the realm of academics - here informatics/computer sciences - and thus delegated to experts exclusively. We saw this in the 9/11 aftermath coverup. Then, Offray, you established yourself as an expert in color, discussing aspects that have allready been introduced by others and including the groups main focus 'Smalltalk', thus manufacturing consent and establishing yourself as a reliable 'expert', and in reverse trying to hit at me, whom you have identified as an adversary. Then you offered a solution in color to the problem at hand with 'traceable AI' and thus tried to open the possibility of collaboration with AI proponents for the once critical reader.
Heh, heh. On the contrary seems that the one seeing a scheme and a enemies location/confrontation with deep plots and tactics is you. Providing external creditable sources beyond opinion, belonging to a established discursive falsafiable tradition (i.e. one that you can criticize instead of blindly accept) is a way to enrich discourse/argumentation beyond conspiracy theories. You could also quote your sources instead, which would allow the community to see where our positions are hold/sustained, even if we use different domain frames, which is better that claiming no domain or expertise in pursuit of openness. So instead of this are my opinions without any external source or reference to pretend no expertise or domain framing, we could advocate for openness by welcoming different expertise and argumentation and making our sources/bias as evident as possible.
I do not state, Offray, that you are knowingly an agent to promote the NWO AI program. I think you just 'learned' / have been programmed to be a successful academic software developer, because to be successful in academics, it is neccessary to learn to argue just like that since the downfall of academic science in the tradition of, let's say, Humboldt. So, I grant that you may be a victim of propaganda yourself, instead of being a secret service sposored agent. You took quite some time to formulate your post, though. You acted to contain the discussion about AI in this vital and important informatics community to technical detail, when it is neccessary that academics and community members look beyond the narrow borders of their certifications and shift their thinking to the point of view where they can see what technology does in the real world.
I offered a view point with sources. In no way the discussion about Apparent Intelligence was contained and, on the contrary, I tried to offer arguments from cognition and philosophy, beyond technical details, which have been pretty absent in the general discourse and rush gold of technoutopism / technoapocalypse, conspiracy and paranoia. Other sources could be added to enrich and inform de conversation and they would be pretty welcomed (at least from me and several members I've known from this community)
Best,
Offray
ChatGPT has been trained on some outdated âfreely availableâ books. I tried it with the first half of the first question of the Advent of Code 2022 and asked it to write Pharo Smalltalk. It produced some outdated stuff using messages that are no longer there. FWIW, isOdd was present in Pharo around 2011.
On Mar 15, 2023, at 2:57 AM, Joachim Tuchel <jtuchel@objektfabrik.de> wrote:
interesting....
#isOdd is not Smalltalk, neither Pharo 10 nor VAST 12 understands this message ;-)
If I evaluate your snippet -replacing #isOdd for #odd, I get an empty Collection. The divisible by 4 things is somewhat interesting, because (1 to: 100 by: 4) is amn interval 1,5,9 etc.... ;-)
So chatGPT is probably a newbie in Smalltalk. Interesting is how logical the explanation sounds, you're tempted to believe it...
Joachim
Am 15.03.23 um 10:45 schrieb giorgio ferraris:
it's interesting to see how the answer changes with only a small change on the question:
(1 to: 100 by: 4) reject: [:i | i isOdd]
gives: The code creates a collection of numbers from 1 to 100, incrementing by 4 at each step using the to:by: message. It then applies the reject: message to this collection, which returns a new collection containing all elements for which the given block evaluates to false.
In this case, the block is [i | i isOdd], which checks whether a given number i is odd using the isOdd method. So the result of the entire expression is a collection of all even numbers between 1 and 100 that are divisible by 4.
ciao
giorgio
On Wed, Mar 15, 2023 at 8:07â¯AM Tim Mackinnon <tim@testit.works> <mailto:tim@testit.works> wrote:
Of course everyone is talking about ChatGPT, and I saw a post to ExplainCode and didnât expect it to understand Smalltalk, so was impressed that it does! Presumably the Tonel Smalltalk code on GitHub has helped along with numerous articles?
Try it out: https://whatdoesthiscodedo.com/
I gave this simple example: (1 to: 100 by: 3) reject: [:i | i isOdd]
And got this (neat eh):
The code above is written in the programming language Smalltalk.
It creates a collection of numbers from 1 to 100, incrementing by 3 at each step. This is done using the "to:by:" construct where the starting value is 1, the ending value is 100, and the increment is 3.
The resulting collection is then filtered using the "reject:" method, which takes a block of code as an argument. In this case, the block checks whether each element (denoted by the variable i) is odd by using the "isOdd" method. If an element is odd, it will be rejected from the collection.
So, the final result is a collection of all even numbers from 1 to 100, incrementing by 3 at each step.
Share this explanation with colleagues: whatdoesthiscodedo.com/g/a0ec56e <https://whatdoesthiscodedo.com/g/a0ec56e>--
----------------------------------------------------------------------- Objektfabrik Joachim Tuchel mailto:jtuchel@objektfabrik.de Fliederweg 1 http://www.objektfabrik.de <http://www.objektfabrik.de/> D-71640 Ludwigsburg http://joachimtuchel.wordpress.com <http://joachimtuchel.wordpress.com/> Telefon: +49 7141 56 10 86 0 Fax: +49 7141 56 10 86 1
I tried it on three "dead" languages: - a bubble-sort in Pop-2 - a system of linear equations solver in IMP-80 - the array concatenation example in SNOBOL 4 from Rosetta Code. It got the first one right -- despite 'array' access being written 'subscr(index, array)' -- and the second one almost right -- mistaking an nxn array for a 2n array. It got the third one (which I had stripped of comments) right too. It's impressive. And it's impressive because it's clearly NOT based on analysing a large corpus of code in the language you give it (because there IS no corpus of Pop-2 and the available corpus of IMP is both limited and very sparsely commented. Not much SNOBOL either. On Wed, 15 Mar 2023 at 20:07, Tim Mackinnon <tim@testit.works> wrote:
Of course everyone is talking about ChatGPT, and I saw a post to ExplainCode and didnât expect it to understand Smalltalk, so was impressed that it does! Presumably the Tonel Smalltalk code on GitHub has helped along with numerous articles?
Try it out: https://whatdoesthiscodedo.com/
I gave this simple example: (1 to: 100 by: 3) reject: [:i | i isOdd]
And got this (neat eh):
The code above is written in the programming language Smalltalk.
It creates a collection of numbers from 1 to 100, incrementing by 3 at each step. This is done using the "to:by:" construct where the starting value is 1, the ending value is 100, and the increment is 3.
The resulting collection is then filtered using the "reject:" method, which takes a block of code as an argument. In this case, the block checks whether each element (denoted by the variable i) is odd by using the "isOdd" method. If an element is odd, it will be rejected from the collection.
So, the final result is a collection of all even numbers from 1 to 100, incrementing by 3 at each step. Share this explanation with colleagues: whatdoesthiscodedo.com/g/a0ec56e
I asked ChatGPT to write âHello, World!â in different esoteric programming languages: Kylix: `program HelloWorld; ` ` begin WriteLn('Hello, World!'); ` `end.` ActionScript (Flash 8): `trace("Hello, World!");` Limbo: `include "sys.m"; ` `include "draw.m"; ` `include "arg.m"; ` `draw->init(nil, "hello"); ` `sys->print("Hello, World!\n");` Alef: `include "sys.m"; ` `include "draw.m"; ` `init(nil, nil); ` `write("Hello, World!\n");` Alef was the predecessor of Limbo, which is the ancestor of Go. The results are not even close. ActionScript not even close. Kylix was wrong too. None of them are way out there. If anyone has worked on Plan 9 then they have heard of Alef and Limbo. Limbo is used for Inferno OS which was created from Plan 9 for set top boxes and came out about the same time as Java. What ChatGPT gave are not even starting points, except maybe Kylix and it derived the answer given from Delphi.
participants (13)
-
Christopher Fuhrman -
danhunfeldz＠mail.com -
Esteban Maringolo -
giorgio ferraris -
in_pharo_users＠nym.hush.com -
James Foster -
Joachim Tuchel -
Offray Vladimir Luna Cárdenas -
Richard O'Keefe -
Richard Sargent -
Tim Mackinnon -
Todd Blanchard -
Tomaž Turk