16 karma · joined January 11, 2017
> most AI companies will slightly change the way their AIs respond, so that they say slightly different things to the same prompt. This helps their AIs seem less robotic and more natural.
To my understanding this is managed by the temperature of the next token prediction which is picked more or less randomly based on this value. This temperature plays a role in the variability of the output.
I wasn't under the impression that it was to give the user a feeling of "realism", but rather that it produced better results with a slightly random prediction.
https://github.com/leogout/rasper-ducky
Duckyscript is a language for the USB rubber ducky that costs approximately 100$. A usb rubber ducky is an usb key that gets recognized as a keyboard and that starts typing text and shortcuts automatically once you plug it to anything. To specify to the key what to type, you can use duckyscript.
I'm using circuitpython. The last thing I did was to de-recursify the interpreter with a stack.
The more I'm implementing of duckyscript, the more i think that i should create my own language. Duckyscript sucks as a language...
One thing I found annoying was that I cannot copy paste part of the prompt while it is being generated because I cannot select it, the selection gets canceled.
Pros :
- A note can fall under as many tags as you want
- The map view is pleasing to watch because tags can be shown but folders cannot, it also helps to categorize orhpan notes visually
Cons :
- You have to maintain the habit to tag each notes you write
Finally, Tag Wrangler allows you to rename / edit tags in bulk so that it is close to be as fluid as moving notes between folders.
I use a plugin called Tag Wrangler to show my tags as a tree like folders (it interprets the "/" as a parent directory).
This allows me to keep a note below multiple tags which I think is useful, at least for my own usecase.
I see it like when I learned about programming, I was frustrated to learn about language theory, complexity, graphs, etc. I wanted to learn langages, frameworks, specifics for being ready to work right at the end of my degree but it would have made me more fragile and less versatile to future changes. Although law and economy are less likely to change as fast as the latest cool tech stack so this example is not the best.
Exactly my point when I was working on a new project architecture and we had to choose between two authentication methods. Because once all your applications rely on an authentication system you can't just switch to the other one like that... Sadly we did not take the time six month ago and today we are working on a migration which could have been avoided.