Hello Roger,
It is indeed a lot of fun, I couldn't agree more! What's truly amazing
is how often I can transform that impressive "looking" LC script into
something actually functional. Using LLMs in combination with the Script
Profiler was all the proof I needed to ensure LLMs are now a dedicated
part of my workflow.
I do wish there was an offline model that didn't produce just
javascript-laden gibberish, but I have a feeling that won't be the case
for much longer.
Thanks for sharing,
Derek Bump
On 9/29/23 13:30, Eller, Roger via use-livecode wrote:
Then I asked ChatGPT to create code in LiveCode to solve the same problem. It generated
some surprisingly great "looking" LC script which on the surface seemed like it
would work. I allowed ChatGPT 4 tries to generate functional and complete code to solve
this problem. It was never able to. Try it. It's interesting to see the variations
that it comes up with.
_______________________________________________
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription
preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode
