The page says this thing will be opt-in. As it also says, they can't ignore the effect AI is having in the world. I'm not much of a fan of a lot of this effect, but see some benefits in places.
They could ignore it. Why not? I switched from Chrome to FF and FF hard-locks 3x a day for me now. Force quit, start over. Is an AI window going to stop me going back to Chrome? Who cares if FF has an AI window if FF is too janky to be a good web browser... It's just lipstick on a pig.
That's not normal, and certainly not a typical reason people avoid Firefox. Have you given it a try and investigated if this is solvable? Sounds like one of those issues that if you ignore, it'll just pop up elsewhere with some other program.
Well, they won't say what they see yet and they won't know what it is until they actually excavate. So, let's hope it's worth the expectation they've built up. At least it also appears that "Time Team" are involved to document the dig. A great team and a great program for those interested in archaeology. It's on YouTube now.
I've been using apt regularly on Debian for a long time and never seen it crash or segfault. Very strange that you do. All software has bugs of course, but apt is so heavily used that I expect it gets attention. It just works for me.
I agree to an extent but they are usefully kept somewhat separate. The introduction to the great "Encyclopedia of Fantasy" put this well. Re: Fantasy :
"Its roots go much deeper into history, and its concerns are more archetypal" [1]
There can be a lot of cross-over of course. Right now, "fantasy" (perhaps of the "romantic" variety) seems to be a juggernaut and is taking over.
Iain M Banks is science-fiction rather than fantasy, so I would not expect him in a "Fantasy Masterworks" series. The two genres have some over-lap but are distinct.
> "Security services" isn't a power center, they are a tool used by power centers.
No, they /are/ a power center. They have a lot of power: they advise governments, police and business, who listen to their advice and act on it very often. A lot of the work they do is (sadly) essential in the world as it exists today. Of course the "security services" want to open and read your mail and will advise accordingly.
I don't use "pi-hole", just an in-browser blocker (ublock origin) and am happy. But I would assume a "pi-hole" would be a useful addition to a household using a variety of potentially ad-infested devices e.g. a smart TV, various tablets etc.
The issue is that (some) smart devices are known to bypass local DNS servers entirely. They either use a public DNS server or hardcoded IPs. The best thing would be not to connect any "smart" TV to the Internet. These are closed firmware devices with cameras and microphones and they just can't be trusted.
> The best thing would be not to connect any "smart" TV to the Internet.
Agree! I regret letting my Vizio TV stay online for as long as I did.
At first it was fine, and I did get a UI refresh a couple years back that was OK.
But then some update caused it to start ripping control away from whatever my last HDMI input was so it could show me ads (which fails). Even though it's perma-offline now, it still messes with my inputs sometimes.
Spot on! My samsung TV menus are soooo much better in terms of snappiness by having Pi-Hole running and also setting up DNAT for those IoT devices who want to hardcode their DNS.
Yes! This is easy to do on OpenBSD as well, though it's called "redirect" instead of "DNAT":
pass in quick on $int_if inet proto udp to any port 53 rdr-to $dns_server port 53
pass in quick on $int_if inet proto tcp to any port 53 rdr-to $dns_server port 53
I also redirect port 53 traffic, and in addition filter traffic to "well known" public DNS servers like 1.1.1.1, 8.8.8.8, 9.9.9.9 and many others (lists can be found on GitHub), but this is ineffective against ads and telemetry served from hardcoded IPs.
Overall, it's just easier not to connect "smart" devices to the Internet at all. I prefer to use a Linux HTPC instead of a smart TV for example. It is completely under my control and I am not restricted to apps approved by Apple or Google, asked to log into anything or to accept ever-changing terms and conditions.
Given that Police Scotland recently found that MS cannot give any guarantees about where sensitive information ends up if it is ever uploaded to their cloud, I would expect any government or military to be evaluating their use of their software (or at least how it is used).
See the Computer Weekly article [2], discussed on HN here [2].
The page says this thing will be opt-in. As it also says, they can't ignore the effect AI is having in the world. I'm not much of a fan of a lot of this effect, but see some benefits in places.
reply