Looks like MLH + DEV have taken over Hacktoberfest, and are trying something new this year: pushing AI development skills and local gatherings, rather than counting how many PRs folks open.
Just ran into a bunch of my E2Es failing on innocuous POST requests. Really interested to know how Cloudflare managed to outright break POST requests going through them.
I can unfortunately report that these dots have not helped me in cars or trains; anything more than a few seconds looking at a screen during a journey will ensure I feel awful until I have an opportunity to sit or lie still for quite a while after. To be fair, even facing backwards on a train usually makes me sick rather rapidly.
He’s a streamer who talks about tech. Previously had a sponsorship relationship with Vercel so is theoretically more well connected than average on the topic. He’s also very divisive because he does a lot of ragebait, grievance reporting, and contrarian takes but famously has blind spots for a few companies and technologies that he’s favored in past videos or been sponsored by. I have friends who watch a lot of his videos but I’ve never been able to get into it.
> Ones NOT marked as sensitive should be rolled out of precaution
if it's not marked as sensitive (because it is not sensitive) there is no reason to roll them. if you must roll a insensitive env var it should've been sensitive in the first place, no?
There's a difference between sensitive, private and public. If public (i.e. NEXT_PUBLIC_) then yeah likely not a reason to roll. Private keys that aren't explicitly sensitive probably are still sensitive. It doesn't seem to be the default to have things "sensitive" and I can't tell if that's a new classification or has always been there.
I can imagine the reason why an env variable would be sensitive, but need to be re-read at some point. But overwhelmingly it makes sense for the default to be set, and never access again (i.e. Fly env values, GCP secret manager etc)
