Howard Rheingold <howard.r...@gmail.com>: Dec 28 04:35PM -0800
I agree with Roland that the handbook could use an update to include the use of LLMs. I replied to him on bluesky that it would be possible to feed the current handbook in its entirety to an instance
...more
|
Roland Legrand <legran...@gmail.com>: Dec 29 09:39AM +0100
I think I will have to use tools such as canvas on ChatGPT or Claude.
NotebookLM focuses on the sources provided by the user so it hallucinates
far less, it does not readily add external content -
...more
|
Roland Legrand <legran...@gmail.com>: Dec 29 03:09PM +0100
I'm discussing it with the Google model Gemini 2.0 Flash Experimental. I
let it analyse the book and asked how to incorporate materials about the
use of generative ai. I got this answer:
...more
|
Roland Legrand <legran...@gmail.com>: Dec 29 03:20PM +0100
I used this model via the very interesting site aistudio.google.com. I
asked to create an enhanced version of the Peeragogy Handbook and the
response was not a ready-to-publish book but still very
...more
|
Roland Legrand <legran...@gmail.com>: Dec 29 03:27PM +0100
This is what ChatGPT o1 made of it:
Roland, let’s take the Peeragogy Handbook, give it a full cyber-style
facelift, and weave in a healthy dose of generative AI. This is not just a
...more
|
Charles Blass <charle...@gmail.com>: Dec 29 03:40PM +0100
thank you roland
what was your prompt?
On Sun, Dec 29, 2024 at 3:27 PM Roland Legrand <legran...@gmail.com>
wrote:
...more
|
Roland Legrand <legran...@gmail.com>: Dec 29 03:48PM +0100
with o1 i simply asked to read and study:
please read and analyse this online version of the Peeragogy Handbook:
https://peeragogy.org/
Then I followed up with:
what are missing are materials
...more
|
Howard Rheingold <how...@rheingold.com>: Dec 29 08:22AM -0800
This is great! The cautions section needs to include warning about
hallucinations and necessity for crap detection. Can you ask how best to do
that? Also: in the section on forums/message boards,
...more
|
Charles Blass <charle...@gmail.com>: Dec 29 06:29PM +0100
helpful insights
so it seems the system generated a palpable exuberance and excitement (in
the direction of hype?) without being asked to do so?! indeed your prompts
are polite as well as minimal
...more
|
Charles Blass <charle...@gmail.com>: Dec 29 06:33PM +0100
agree on highlighting/ integrating critical need for crap detection within
these contexts, (perhaps as notes/ footnotes spread throughout the handbook
anywhere ai prompting/ processing outputs are
...more
|