I mean, following proper disclosure procedure requires giving the manufacturer adequate time to patch the vulnerabilities before they are released. Would you prefer they just publish the full how-to right up front, and create a zero-day for anyone (from script kiddies to state-sponsored hackers) to use as they see fit?
“We have one full-time staff member (that’s me) and 14 mostly part-time contractors,” Executive Director Elizabeth Barron of the PHP Foundation said via email in an interview on Friday. Board members are unpaid volunteers, and funding comes from donations and grants.
Sounds more like she is stretched thin, and pointing out how unsustainable the current model is. Only one person dedicated to maintaining a programming language that ~70% of the internet uses is the perfect example of that xkcd “all modern infrastructure, vs one obscure person who has thanklessly been maintaining a project since 1997” meme.
Yeah, as long as your firewall is properly hardened and up to date.
Human intervention only automatically kicks in when the car actually thinks it is trapped. Which can easily take a while for larger parking lots.
It’s really not that hard to believe. Waymo will use any available roads to navigate between fares, and it can become especially bad when they’re waiting for fares. Because they don’t just return to a home parking spot and wait between fares. The point is that they want to spread their potential reach (and ensure there are cars nearby regardless of where they’re requested) so they spread out until they need to return to charge/refuel.
They cruise around random backroads and residential areas until a fare requests them. They also avoid U-turns whenever possible, which means they’ll often path through some interesting places just to avoid turning around. They quickly clog residential roads when fares are slow, by all driving in a holding pattern around residential areas that aren’t designed for that kind of through traffic.
It doesn’t, as long as those images are good about keeping their dependencies up to date. Because containers are entirely self-sufficient and come packaged with all of their necessary dependencies, there is a potential security concern with images keeping dependencies pinned to old, insecure versions. And catching that can be tricky sometimes, especially when you’re running larger stacks with lots of containers.
But I’d argue that is actually the image being insecure, not containerization itself. It would be no different from a program being considered insecure. We would handle an insecure image the same way we would handle an insecure program: push the maintainer to update their shit to a secure version, or fork it (if it’s OSS) to update the dependencies directly.
Plus, containerization is much more straightforward compared to the olden days of directly managing dependencies. For larger setups, that was basically a full time job by itself. Especially when you have multiple different programs all trying to call different versions of the same dependency.
Just remember to put a MOCA filter on the line coming into your house from the street. Or disconnect it entirely, if you’re not using the cable connection. Otherwise you’ll be sending all your Internet traffic straight out the cable line to all of your neighbors.
And it’ll only take one NYC resident to challenge that before the requirement is outlawed. If it doesn’t require proof of address to sign up, it shouldn’t require proof of address to cancel.
Yeah, pretty much this. In regards to tech, the fediverse often suffers from the average familiarity problem:
The average user barely even knows how to check their email.
Do you think “anarchy” means you actively try to give zero-day methods to state-sponsored hackers?