Tag: opinion
Public whispers tagged opinion
Text Highlight2026-08-12 10:14:25
Original Highlight Excerpt
"next generation of highly compute-efficient multimodal models"
Whisper Note
Compute-efficient but what about the energy cost for training these things?
Text Highlight2026-08-11 03:09:18
Original Highlight Excerpt
"A Foolish Consistency is the Hobgoblin of Little Minds"
Whisper Note
Classic quote, still makes me think twice about over-obsessing with style.
Text Highlight2026-08-09 21:09:57
Original Highlight Excerpt
"error handling is intuitive"
Whisper Note
Error handling being intuitive is a huge win, honestly.
Text Highlight2026-08-09 21:09:18
Original Highlight Excerpt
"Flat is better than nested."
Whisper Note
The flat is better than nested one really hits home with every refactor I do.
Text Highlight2026-08-08 15:09:18
Original Highlight Excerpt
"this is sounding awfully familiar."
Whisper Note
baseline choice is like picking what 'missing' means for your model, tricky stuff.
Text Highlight2026-08-04 21:09:18
Original Highlight Excerpt
"When appropriate"
Whisper Note
Good point, sync order makes debugging way easier.
Text Highlight2026-07-30 21:09:18
Original Highlight Excerpt
"not a modular toolbox of building blocks for neural nets."
Whisper Note
That's a refreshing take, keeping it simple for researchers to hack on.
Text Highlight2026-07-28 13:01:34
Original Highlight Excerpt
"Content-based filtering methods are based on a description of the item and a profile of the user's preferences"
Whisper Note
But what if the user's taste changes over time? This seems too static for that.
Text Highlight2026-07-28 09:49:34
Original Highlight Excerpt
"Collaborative filtering is based on the assumption that people who agreed in the past will agree in the future"
Whisper Note
But what if I suddenly change my taste? That assumption feels shaky.
Text Highlight2026-07-27 12:40:15
Original Highlight Excerpt
"since it often causes poorly readable code."
Whisper Note
I only use * in the REPL, it's too handy to give up there.
Text Highlight2026-07-27 12:31:15
Original Highlight Excerpt
"since it often causes poorly readable code."
Whisper Note
Yeah, wildcard imports make it impossible to know where names come from.
Text Highlight2026-07-27 10:21:52
Original Highlight Excerpt
"Graph neural networks (GNNs) are artificial neural networks designed for tasks whose inputs are graphs."
Whisper Note
But aren't graphs just a special case of tensors? Seems overhyped to me.
Text Highlight2026-07-27 09:49:28
Original Highlight Excerpt
"Language models are useful for a variety of tasks"
Whisper Note
They forgot spell check, which is the only one I actually use daily.
Text Highlight2026-07-27 09:41:50
Original Highlight Excerpt
"Deeper neural networks are more difficult to train."
Whisper Note
Yeah but adding layers never made anything easier in my experience.
Text Highlight2026-07-27 09:40:28
Original Highlight Excerpt
"Language models are useful for a variety of tasks"
Whisper Note
Sure, but route optimization feels like a stretch—that's more graph theory than language.
Text Highlight2026-07-26 16:09:06
Original Highlight Excerpt
"Use spread syntax with restraint."
Whisper Note
Yeah but when you have like 10 props it's just so tempting...
Text Highlight2026-07-26 16:04:21
Original Highlight Excerpt
"each generation ends up corresponding to a new step in a random walk of model parameters"
Whisper Note
So basically AI models are just wandering off randomly with each new generation, wild.
Text Highlight2026-07-26 12:59:31
Original Highlight Excerpt
"There is no information about the world which is hidden from the state."
Whisper Note
So if the state is complete, we never need to infer anything? Seems too ideal for real life.
Text Highlight2026-07-26 12:38:41
Original Highlight Excerpt
"unpredictable phenomenon that we refer to as emergent abilities"
Whisper Note
But is it really unpredictable or are we just bad at measuring small gains?
Text Highlight2026-07-26 10:20:43
Original Highlight Excerpt
"A model is transparent "if the processes that extract model parameters from training data and generate labels from testing data can be described and motivated by the approach designer.""
Whisper Note
I think this is spot on — transparency is about the designer's ability to explain, not just the code.