My favorite chili oils take the chilis just beyond a perfect maillard "sear" and venture very slightly into burnt territory. Many peppers have a natural bitterness when dried, and that slight bitterness added from the barely burned flavor brings out some extra, magical, umami depth. Adding some acidic elements makes for some incredible dishes.
I have found that smoked chilis lose their smokiness with high heat though. I wonder if doing a lower heat extraction might retain those smoky flavors. Sounds like the multiple extraction/blend the author talked about would be worth a shot to get both.
I agree. Indian cooking uses a lot of flavored oil as “tadkas”. Spices (cumin, coriander, garlic, mustard etc.) including whole red chillies are brought up to temp in a small metal bowl with ghee over a flame and then added to the final dish.
I’ve made chili oil this way and it seems to draw out more flavor and there’s the slight burnt flavor that I love.
I've learned from cooking more Indian food recently that it's okay to burn spices a little. In most cuisines you're told to only bloom spices briefly but Indian food isn't afraid to really cook spices to their limit.
I think understanding this distinction is going to be one of the main challenges for mainstream WebAuthn adoption over passwords as the default authentication method.
> It sounds like you want to trivialize a problem that has existed for way longer than computer science and systematically relies at some point on human memory if you want a certain level of security and secrecy.
Assuming the "problem" you are referring to is authentication, this is not necessarily true. "Something you know" is just one auth factor of several. Using a thumbprint (something you are) to grant access to a private key stored on a device to sign a challenge does not rely on human memory, and is probably more secure than using a password in most cases.
An argument can also be made that using a password manager implies that you are specifically doing the opposite of relying on human memory for authentication.
That said, I also disagree with the premise of the original comment that "Passwords should be dead soon and this article should be irrelevant soon after." They will be around for a while, although I hope passwordless mfa can soon replace them as the default choice.
In response to this post, I just open sourced a starter project to a variation of this idea: https://github.com/secretlessai/audio-mnist. I've been interested in doing image classification techniques like CNN on audio data for a while.
A couple years ago for a weekend project I made a simple "audio-mnist" dataset from handwritten digit audio recordings. I never got past a few days worth of work, but open-sourcing it has been on my mind for a minute. This post kicked me into action. Getting some more data, basic CNN examples, etc. could provide a nice starting point for a lot of research and tools.
There is still separate code I'd have to find and make intelligible to create the recordings and split the audio.
Anyway, in case anyone finds part of this process interesting or useful.
Incredible work, this is the most exciting AI dev tool I've come across!
Do you have a strategy to supplement ChatGPT to handle post-2021 updates to languages and libraries? I tried it on a NextJS repo and it came up with something that looked like it would have been correct a few versions ago, but I had to make some manual changes. Certain fast-moving ecosystems might frequently have this issue.
Thank you! We're working on integrating external browsing using another agent. For now we do have link processing, so if you drop a publicly accessible link in the issue, Sweep will actually gather context from that link.
You can give Sweep docs about a framework and it should help a lot.
