The source for this article is another article, which tbh is a better article:
https://cybernews.com/security/global-data-leak-exposes-billion-records/
That article does not itself have a source link, but it does show some redacted sample data and a breakdown of what countries are affected.
Is this actually targeted at the UK?
It’s a super American ad. Everyone sounds American, we don’t really have drive thru’s like this etc.
In terms of message it’s spot on for both the UK and US though. Especially prescient given the Flock stuff going on in the US right now.
EDIT: From the article, it sounds like this has been used in a few places, but the UK is the only one that has banned it. The reasoning behind the ban is laughable.
It looks like there’s a longer version available on the site, but it’s broken for me.
Nextcloud isn’t really a direct equivalent to Google Photos. For that you want Immich.
Of course. It’s not like I want this. I was out there messaging my MP when the UK law had just been pushed through and filling out the petitions. It didn’t help though.
You can send a message to Discord by leaving, but they won’t revert it for places where it’s legally required.
I will say that this has been in the works for a little bit now and it’s more related to the UK’s new laws. We’ve had this age verification stuff on Discord for some months now.
As for why it’s being rolled out globally, not sure. The UK is not the only country / state starting to ask for ages verification on “social media” so maybe they thought it would be easier to use the same approach everywhere? Seems like a bad idea though as everywhere it’s implemented it’s unpopular.
My experience with it so far has been that… well… It hasn’t affected me at all. Mostly because I only use friends servers and activism servers. Those aren’t age restricted, so you never hit this check.
As far as age verification law compliance / implementation goes, this is not the worst one. That data leak early on was a clown show though.
+1
I discovered it when I found it was shipped with CalyxOS and honestly it’s way better than anything else I’ve found.
When a country does war crimes, the entirety of its populace is not to blame, the systems, power structures and people in charge are.
Most people in a country are not for wars or other attacking other countries. This is true for America and Russia alike, to the best of my knowledge (which granted, is anecdotal).
More people should be engaged in activism and should push back against the inhumane actions of their state, and people should be less gullible to the promises of their leaders.
Both America and Russia are overdue for major reform. Their systems are broken and many of their people brainwashed. Wishing harm on them does not help and imo is not a constructive approach. If anything, hate just begets more hate, making things worse.
If you are going to have hate, it should be more focused on the people in power.
Splitting hairs. If you think a group of people are intrinsically worse than others, you are effectively racist.
We can come up with new names, but ultimately it’s all the same. If you think racism means black vs white then your understanding is kindergarten level.
If you prefer, we can use the term bigoted. That term also covers other intrinsic groups like trans, gay and disabled people among others.
But WHERE’S THE PROFIT INCENTIVE.
“Don’t protect yourself or others, even peacefully. You’re obstructing the oppression.”
When you log into Windows with a Microsoft account, your recovery key is often automatically uploaded to Microsoft’s servers as a backup in case you forget your password. Legally, this means Microsoft owns the key and must surrender it under the U.S. CLOUD Act.
I find that really quite shocking, but I guess I shouldn’t be surprised.
Given the legal and technical risks, the advice for business travelers is clear: do not carry data.
The US really is a hostile surveillance state.
Agreed, but recommended specs are >4GB RAM which should be achievable even with a really old secondhand computer or SBC like a Raspberry Pi.
Been using it for about 7 months.
After about 2 months I decided to completely move away from Google Photos and now I only use Immich.
This was a big move for me. I have over 90 000 photos and 1000 videos totalling over 200GB.
The performance is great even on spinning rust, though I am running it on a Ryzen 2700X and 32GB RAM. That said it’s only when the machine learning background tasks take off that that CPU horsepower is used. You really don’t need that much. The recommended specs are fairly small.
Since I first installed it, they’ve added an auto-OCR feature which is a godsend. I can search my entire library for text on a screenshot and it works really well.
Weirdly, the missing feature that really pissed me off on Google Photos and got me to move over was the lack of the ability to search for images not currently in any album. The search functionality is much better on Immich.
I think the only feature I’m missing from Google Photos is the non-destructive editor, and that’s coming real soon (the PR is already merged as of last week).
For second hand, user-to-user sales via EBay and Vinted are the cheapest, but it takes more effort and you could get scammed.
I’ve bought a few devices second hand at a second hand electronics shop. In the UK, this could be CeX, for example. With many of these you’ll get a limited warranty, and you can walk into a shop and ask to look at the phone and ask questions. You can ask if it is locked and return it if you can’t use it.
Why not buy the phone from the manufacturer directly so as to not give money to Big Bezos, if you’re going to buy new?
Or was this talking about used Amazon listings?
That’s a fair point. I think some of our interviewers have said that they don’t mind the candidate using a LLM, as long as they are up-front that they are doing so.
I’d say the kind of use is important. If they are using it as a form of advanced auto-complete, that’s fine. If they are using it uncritically, or to avoid thinking about the problem, I doubt I’d hire them.
We need engineers who can solve problems, not a salaried middle-man to an LLM.
Many of our candidates are from abroad, and we pay their VISAs and help them move here if they are hired.
You can offer in-person as an option, but I’m not sure most of our applicants would want to travel hours for an interview. Especially if there is more than one stage with deliberation needed in between.
Most of our applicants seem to be people currently in employment but who don’t like their job. They are likely doing interviews on the sly during work hours and likely don’t want to take a full day off or signal to their employer they are looking for a job.
All this to say I doubt forcing employees to do in-person interviews is a good option for most people, but I do agree it should be an option the interviewee can ask for.
Playing devil’s advocate: The reason companies feel the need to put these systems in place is most likely because many candidates cheat using chatbots.
In my company, until very recently, engineers were running the first and second stages of interviews (right after CV vetting) and I’ve heard many times in the last couple of years that my colleagues suspected candidates of using LLMs. There would be unnatural pauses, typing after every asked question etc.
Granted, I don’t think any have slipped through to being hired, as it’s still pretty obvious, but I can understand why companies may want to put safeguards in place.
Are they going too far here? Absolutely.
For us, we actually sit with the candidate in a pair-programming kind of setup to gauge their vibes, way of thinking and confidence as they solve coding problems that closely match what they would do on the job. That usually eliminates “seniors” that haven’t coded for 5 years or that got there by nepotism or sheer passage of time.
I went through the exact same thing with Dyson back in ~2018 worst interview process I’ve ever experienced.
The one I hear is good is NVDA by NVAccess, but it doesn’t have AI.
Website: https://www.nvaccess.org/about-nvda/
Source code: https://github.com/nvaccess/nvda
I would be fairly surprised if you found an open source screen reader that had AI built-in.
It would likely have to run locally if it was open source.
Typically companies don’t make their stuff open source, and non-profits are unlikely to host an AI for you because of the cost. It’s unlikely to run locally as the cost for that model to run and download size may make it unusable or impractical for some hardware. Typically screen readers need to be accessible to everyone, and therefore need to run on very old and / or cheap hardware.