benaubin

our app is electron and you can open the chromium dev tools via the help menu. if you want to verify yourself.

in seriousness: we don't listen to your microphone unless you click the button that says "start transcribing" in big letters.

we dropped the ball with the support response quoted the top of the thread. we don't process your microphone data until you click the button to start transcribing, and remind the user to confirm they have two-party consent. we merely detect when a meeting app is using your microphone.

source: i work at notion, just checked the code. it's electron, you're welcome to check for yourself.

have any in mind? we're hiring :)

http://job-boards.greenhouse.io/notion/jobs/6345904003

To be clear: recording is opt-in. We also remind users to ask for two-party consent. (obligatory "i work at notion")

I work at Notion. We don't ship code that would let us listen in until users explicitly click start recording. We don't want to, we aren't in the business of selling data and this would be incredibly expensive - and destroy trust. We are cash flow positive so we can sustain our values for the long term. We build useful software and get paid by our users for it. It's pretty simple. Agree that operating systems should display prominent indicators when the microphone is in use.

it's electron so you can just open chromium dev tools and see almost all network activity - im pretty sure this is exposed to everyone in the debug menu. takes seconds. http proxy the rest. (i work at notion and do this all the time to debug)

now you've banned several different arbitrary strings!

Notion employee here. We don't put images themselves in Postgres- we use s3 to store them. The article is referring to image blocks, which are effectively pointers to the image.

> However, as the number of members in the set increases, the number of possible IDs increases exponentially. Case-sensitive: 53^8 = 62,259,690,411,361 Case-insensitive: 22^8 = 54,875,873,536

Nitpick, but isn't this polynomial to the members of the set?

I'd assume that there's a difference between picking the best _token_ across an assortment of randomly selected tokens, versus picking the best _string_ of randomly-selected tokens.

"'All You Need Considered Harmful.' Is All You Need"

About 3 years ago, Bird used to be the only scooter useable on my college campus. And they were pretty commonly used. Then, 2 years ago, the bird app completely stopped working on my phone. Every time I tried to unlock one, the app would just show an endless loading animation. I tried reinstalling. I tried using different payment methods. I couldn't find any way to get a scooter to unlock, and the moment I needed one, I was always in a hurry. By then, Lime began to comply with my university's rules and their scooters were allowed to park on campus. I tried the app many more times, even a few weeks ago, and just... never worked.

I don't think it was just me - I stopped seeing Bird scooters get used by others. I still saw the occasional Lime.

It's entirely possible I'm extrapolating too much from my experience and a rough observation. But still, I'm kind of surprised it took them so long to fail.

Needs async/await:

async: DEVELOPING STORY [DISCOVER HOW TO...]

await: STAY TUNED FOR [expr]

This is actually a pretty neat idea. Should be pretty easy by just swapping out the tokenization loop logic. https://github.com/thesephist/tabloid/blob/e0d16ea54eced809e...

10 [EASY|QUICK|WEIRD] STEPS = Range

[list expr] FOR [variable name] = map operator