I read the bill and I feel like it's missing any technical details. It's almost like they read my suggestion [1] but then left some parts out. The technical parts. As I read it one can just enter whatever name, age and other details in the setup of a computer they desire. It's missing any checks for a header on the server to detect adult content labels. What am I missing? What forces me to enter my real information? Are operating system developers going to be granted access to the DMV databases? Or forced to use some third party that scratched the back of some politicians? If I block connectivity to this will I not be able to log in? If someone performs a successful DDoS to the site will I not be able to log in? It feels like several pages of the bill are missing. How does the OS know it is visiting an adult site?
The site "reclaimthenet" calls it age "verification", but it's not a "verification" at all. There's your mystery.
All the bill wants is that you can set up an iPhone for kids, an children account on Ubuntu (YOU decide whether it's a children's account) and then, presumably, the browser vendors implement an AgeAPI that allows website operators to query the user age.
Your device tells us you're 10 years old. Access to Instagram denied.
Your device tells us you're 16. You're not allowed to visit gambling-porn-and-industrial-accidents.org
It's, of course, exactly the opposite of the "identity-tied age verification government-control, ID-document-leak" dystopia that the scare crowds here are peddling. But you'll never hear a word of acknowledgement from them.
These people act as if those "I'm 13 or older, i can create an Instagram account and waste my life" or "I'm 18 or older, let me watch porn and strangle my girlfriend" buttons are the peak of civilization.
Ah, well that's at least half of what I suggested. Telling the site the age seems leaky to me, I would still prefer the apps check for the RTA header so all decisions stay on the device and not leak anything. Curious where it goes from here but based on your reply it does not seem quite as bad as I imagined. Thankyou for the clarification. I imagine eBPF or MAC rules could be used to block this.
Government should like the RTA header as they can fine sites daily that are missing it. Lobbyists could push companies that do the header checks.
They have dropped all the decision making for the details in the lap of the politically controlled FTC. Which also means that future FTCs could change the rules based on political goals.
Not the one down-voting you but I suspect IPv4 and IPv6 will run side-by-side dual-stack at least until 2050 and IPv4 would not be fully deprecated until probably 2323 star-date 0 as future quantum interstellar communications subspace will deprecate both standards. Only half kidding as nothing ever gets deprecated or decommissioned until it breaks something important.
During the dot-com crash I had to put a /16, some /17's and a /19 on one vlan and connected a 1U Linux box running Labrea Tarpit just so those ranges would respond to ping because InterNIC used to harass us for not utilizing all the space. They threatened a few times to take them back. AFAIK nobody nags like that any more. They probably should.
What areas/domains of software are least likely to be replaced by AI?
I am entirely guessing here but probably anything that is involved in extremely tight regulations and have legal oversight. Probably things related to the FAA and aircraft navigation, especially automated decisions such as TCAS Traffic Alert and Collision Avoidance Systems that automatically control real time aircraft navigation decisions. I would not doubt that AI may contribute some code at some point but everything would have to be strictly reviewed, tested and approved by multiple humans or at least that is what I would expect or hope.
On a similar note, probably land based traffic management systems that make real time decisions about traffic flows. These systems have insane levels of regulations around them and only specific people with specific degrees, certifications and licenses are permitted to even touch them much less engineer anything to do with them.
Probably something similar for water navigation especially navigating around structures like bridges, oil derriks and related platforms.
I am specifically talking about things that the HN crowd would consider to be AI LLM, big-data based platforms. There are proprietary systems that have their own concept of intelligence and automated navigation, mitigation and defense that the defense industry utilize that are unrelated to what people here call AI.
As someone that has managed very large outbound transactional email environments, email campaign platforms and some corporate email I just wanted to wish Cloudflare the best of luck on this endeavor. This is an entirely different animal from anything related to a CDN. Stay vigilant and don't let the cute and fuzzy bunnies ruin it for everyone else. They are evil and mischievous and will do whatever they technically can do.
I don't get particularly excited by birds normally
Same. Sometimes one of my deer will get thwacked by a car just hard enough it stumbles up my driveway and falls over. There will be 3 golden eagles and 2 bald eagles fighting over it. The first time I saw them I had a double-take ... I swore at first I saw men sitting on my driveway fence. Golden eagles are massive and quite awe inspiring to watch. When they fight over road kill they stretch their wings out entirely.
Each time I have to make sure I still have an outdoor cat and I have to keep an eye on him until they are done. They seem to only eat the soft bits and leave the muscle meat for the ravens. Then the deer turns into a fly factory which I have to spray.
Unpopular answer but ask your favorite AI to show the history of how taxes increased in the USA since 1913 and what those taxes were spent on. Then ask how often such programs are ever removed and the taxes are reduced and surplus given back to the people.
Related recent discussion of taxes in California [1]
Why is there a flock camera indoors at a school in the first place? Are the schools supposed to be putting video and audio footage of children on 3rd party storage platforms? Are the parents aware of this? Perhaps PTA meetings should discuss. That seems like something that should be using close circuit PoE cameras to local NVR's with on-prem encrypted storage with a retention policy if there must be cameras. Encrypted CEPH perhaps? [3]
Just as one example Zoneminder [1][2] can be clustered and distributed assuming a large campus. I'm sure there must be other open source NVR's that can do the same. School IT staff should try out a small deployment first and then extend it year over year. Local AI should detect and alert on fights, abuse from teachers, anyone with a weapon, someone injured, etc...
Bob can be granted access to specific cameras that relate to his role to avoid Repetitive Strain Injury RSI among other issues.
The main reason that organizations choose commercially managed solutions is because they don't have local expertise or staff to do things themselves. I do agree that on-prem solutions are better, but Zoneminder is probably not a great option. Besides being old and clunky, it also isn't anywhere near a complete solution, and the IP cameras people often choose to connect to them are often security nightmares. There are many good and complete commercial offerings that are secure and keep video locally.
I totally get what you are saying and there are certainly some schools that lack IT staff, budget and experience but there are some schools that have big budgets and plenty of IT people sitting on their hands that could slowly build this out, document it in a way that schools could budget around YoY and set examples for other schools. Maybe even use it as a project to get students some college credits.
If there are better options than Zoneminder please do share the tutorial videos with others here so they have greater options. I am old and clunky so ZM works for me. Some may even say old and clunky can mean reliable and low maintenance. There are probably some school IT admins reading this. ZM has great documentation and tutorial videos in my opinion. It is also used by a large number of corporations.
Just my own philosophy but I am leery of expensive turn-key commercial solutions as they lead to proprietary solutions that school IT won't understand and will just lead to dead cameras and empty NVR's when law enforcement need them the most. It will be one of the first maintenance contracts that get cut from budgets.
Just because someone has an IT staff doesn't necessarily mean that staff really has the expertise to set up a bespoke surveillance system properly. Nor does it really make it a good idea to do so. Nor is it even a good use of time when packaged systems can fulfill most requirements with less integration risk.
The software running on an NVR is only one small part of a surveillance system. I'd be much more worried about the choice IP cameras themselves, which are notoriously problematic. And if you look at the cameras which are well regarded and high quality -- typically those vendors have their own NVR solutions which are also well regarded and already tested to work well with their cameras.
> I am leery of expensive turn-key commercial solutions as they lead to proprietary solutions that school IT won't understand
If IT can't adequately evaluate and choose a turn-key solution, I doubt their ability to piece together their own system.
> If there are better options than Zoneminder please do share the tutorial videos with others here so they have greater options. I am old and clunky so ZM works for me. Some may even say old and clunky can mean reliable and low maintenance.
The last time I tried Zoneminder, the problem I had was that the detection algorithms were so bad that I found them useless. The cameras I had were all outdoors and their algorithm struggled to strike a balance between detecting legitimate motion and not falsely triggering when lighting conditions changed. I tried some other projects that had better algorithms for filtering out changes in exposure and lighting (I forget which ones), but there's also some now that have AI object detection. But ultimately I've migrated away because commercial options got better, cheaper, and more feature filled.
If I picked a new system today I'd probably try something like: https://www.ui.com/us/en/camera-security I don't have any personal experience with it but the value looks incredible.
The last time I tried Zoneminder, the problem I had was that the detection algorithms were so bad that I found them useless.
Fair enough. I've had them set off by deer no matter how hard I try to avoid it. I think they know they are getting my attention.
For what it's worth in a school setting there can be monitors in multiple admin offices, the admin waiting area, school police office and other offices to group source monitoring of strange activity. Otherwise if nothing else it is useful to be able to go back an hour, a few hours or days to verify the "he said, she said" accusations often uttered in school.
That and paying to offload legal liability to a vendor.
Lots of great, free, widely adopted open source technology solutions aren't adopted by public sector because their legal staff won't accept the liability of not having a paid contract that makes guarantees. Great use of tax dollars.
> Are the parents aware of this? Perhaps PTA meetings should discuss.
Not everyone grows up in such an idyllic environment where there is an active and engaged PTA or concerned parents who feel like they have a voice. Moreover the perceived need for security cameras is probably inversely proportional to places with active PTA groups (though maybe not). Either way, suggesting tech solutions is rearranging deck chairs on the Titanic.
Either way, suggesting tech solutions is rearranging deck chairs on the Titanic.
My gripe will be the music they are playing whilst I am moving the deck chairs on the Titanic. Enough ragtime already. I will take some Moonlight Bay please [1].
Oh and to your point of course there will be places that can't do this. They should be focusing on the proper disassembly cleaning lubricating and reassembly of their Hi-Point's. Such schools should have mandatory handgun safety courses like the old days. Or current times for the Swiss [2].
The board might not even be the right place. They’re watching children in minimal dress athletics. I think the DA or the AG’s office might be the right place.
Arguing over on-prem vs cloud misses the entire point.The architecture doesn't matter when the core requirement itself is just insane surveillance.We should be angry that our engineering is being weaponized to fulfill such a sick requirement in the first place.
[1] - https://news.ycombinator.com/item?id=46152074
reply