If I read previous articles right, Google has stopped shipping Pixel specific hardware drivers down to the android open source project, so any hardware support graphene figures out going forward has to be reverse engineered anyway.
So that’s some additional motivation at least.
“Had”? Brother they still do. And so does anyone else who pays for access. Benn Jordan’s video on it doesn’t dwell on it, but it clearly demonstrates that anyone who has paid into their “network” has effectively full access to every other camera and recording in it.
I mean this softly, but I’m going to guess you haven’t used OneDrive recently, and haven’t used it where it’s been set up in a competent manner. The default settings absolutely are not conpetent, espiecally for how messy computers for personal use get.
My workplace uses OneDrive to sync a specific set of user profile folders so we approximate having profiles and files that follow us without everyone needing a personal folder on a network drive that mounts at login.
The only issues we’ve had are profiles auto-downloading too mant of peoples files and eating drives on shared machines (so you just have your meeting room computers wipe all profiles every reboot and schedule reboots nightly), and I’ve had some issues where OneNote hadn’t actually synced the notebook back to the cloud before I closed on one machine and opened on a different machine so I lost some notes.
Beyond that, it’s handled even situations where I have the same file open siniltaneously on multiple machines smoothly. Syncs between login on multiple machines take 3 minutes max, and I can force it faster if I really need by pausing and resuming the sync.
I’m sure there’s situations it’s still not suited for, like editing and syncing large monolithic files (think video files over 1GB a piece). It probably sucks big time on personal machines where you’re going to have a complete mess of every file type imaginable tossed in one big unorganized heap.
But configured correctly, for general business use, it can work very well.
I would be shocked if this hasn’t had some set of controls to disable it in Group Policy for months now.
This is just rent seeking against Home users.
People with One Drive through corporate Azure sjbscriptions (rather than the free “you have a microsoft login” tier) already have fairly robust controls available for handling and securing private data. There’s even special Azure tiers for government work that are even further secured.
This is only going to impact home users and conpanies without strong IT teams. Which is an egregious amount of people, don’t get me wrong. It’s also a horrible anti-consumer move. But this isn’t “Microsoft fucks over their golden calf: business users”.
With the upcoming restrictions on third-party apps that Google has announced maybe? It’ll be easier to get from Play, and may not be available otherwise at all.
In short it’s hard to do right.
It would have to be in a way that doesn’t chain you to one central identification server that could ban everyone, but also still handles moderation actions properly.
People would probably also want their identity to be portable (movable to other source instances) which will mean different things to different fediverse services in terms of what would move, and things like handling username collisions. We can’t even lift and shift between instances of the same service yet, whete it should be a simple one to one.
Most of all, a huge part of the conceptual point of the fediverse is that you should be able to interact with the rest of it from any point inside it (with exceptions for intentional defederation between some servers). On a conceptual level, you shouldn’t need to create a login for mastodon when you can post to mastodon from lemmy, and vice versa. You create your login on the type of fediverse system you want the interface of, on a server instance you agree with the moderation policies of, and then you interact with whatever you want from there.
In practice it’s not so simple, but that’s largely seen by the various devs as a software interoperability thing, not a sort of single sign on identity thing.
He and his family were caught doing pretty serious charity fraud, iirc, amd he kept throwing out completely absurd excuses rather than owning up to it.
Entirely wrong youtuber. Sorry.
Sounds like it sucks at every level. From what I’ve dealt with on just software/drivers:
You want to use scan to email through anything that isn’t a fully open, no auth, anonymous SMTP relay? Go fuck yourself.
Wait… we changed our mind. We’ll totally support SMTP authentication, but with an arbitrary undocumented limit on the password length we can store, which is definitely shorter than the password length requirements for most SMTP relay suites. Certificates? Holy shit are you from the future?
Or you can scan to network share, but I hope you enjoy finding all the hidden catches and caveats that are completely undocumented!
You want an option so people have to log in at the printer itself to release their print job? Enjoy six different interfaces for five different underlying standards for how that works across two different manufacturers. And we reserve the right to just stop supporting that feature or change it entirely with any firmware or driver update. And if there’s a mismatch between the driver and firmware then we’ll just make the print spooler/job queue shit itself and require manual intervention to continue printing.
You want our driver to properly communicate to end user software the paper sizes it supports? If it supports double sided printing or not? How it will collate multiple copies? Man, we can’t even care enough to indicate to software if we’re Black and White or Color. Best we can do is completely ignore the options you picked through your software and our driver and just do whatever we think is best. That’s a good compromise, right?
For the price of these god damn enterprise mfds, there’s no excuse.
What’s objectionable about it Apple? Hmm?
I’m going to guess that you don’t work in tech, because good luck accomplishing that.
The overwhelming majority of businesses that allow remote work do so via VPN. It’s possible for some companies to get away with just Office 365 and some basic collaboration tools, but not for most. Some businesses even set up secure VPN tunnels between sections of their network and a section belonging to a contracted partner, software vendor, data processing firm, auditors, etc.
So banning is absurdly unlikely. But let’s say they do, or they restrict “allowed” VPN through government approved VPN software or something.
There are countless ways to hide VPN traffic as other legitimate/normal web traffic. Plenty of people in war torn areas, reporting from countries with locked down internet, and other similar scenarios have effectively proven that you can’t completely lock things down.
So they could ban it, but that’s absurdly unlikely, and even if they do there will be plenty of relatively safe ways around it.
This is about all app makers, not just ones in the Google Play Store. This also applies to third party app devs and hobbiests making open source apps to put up on alternative app stores like the all open source F-Droid.
Have fun with that then. Sorry about your balls.
If you’re competent and technologically saavy enough to use a Linux distro as your daily driver, you can learn to make Windows work for you too.
Waste of effort if you don’t need to interact with any Windows environments for school or work, but definitely possible.
For me? I’m happy to get paid to automate shit using PowerShell that should have been basic built in functionality from the start. PowerShell is just the most convenient scripting language due to being packed-in with most Windows installs, and tons of built in functionality for interfacing with other Microsoft products. So as long as Microsoft keeps sucking, I’ve got a comfy paycheck.
And if the year of the Linux desktop ever finally happens? I’m ready, I’ll be cheering, and I’ll be ready to get paid helping companies to make the switch.
I’m usually a Windows “shill” or at least a casual defender of it, as I work in a Windows environment and it’s not as bad as people pretend it is. No shade against Linux, I love it and Windows is bad. Just not like “I’d rather self-castrate” bad.
Anyway…
But for a home server? Either be super lazy and set up samba shares from your Windows desktop for the drives (avoid having a server at all) or bite the bullet and use Linux. You’ll get so much more out of a Linux server that it’s not even funny.
I mean, there’s levels to this. If I’m looking for information, having a summary rather than a highly technical primary source can be very useful. Wikipedia cites its sources, and (ideally) has summaries made by groups of people familiar with the subject and following consistent and detailed publicly available style guides. Wikipedia isn’t running ads, and is not for profit.
When an AI summarizes these primary sources, or even summarizes Wikipedia, you get none of that. AI does not reliably cite sources (ones not made for it will just generate a convincing looking response, making up sources whole cloth. Ones made to cite sources will often not actually cite the ones they used, and still can make up sources more rarely). It can’t reliably summarize things accurately, as it doesn’t understand anything, especially not terms that have different meanings depending on the technical context. There’s no group of people reviewing and revising. There’s no incredibly detailed style guide. All these AI are explicitly for profit (the amount of self hosted out there is negligible and those are much less of a problem), and almost every one of the companies running them have openly spoken about future plans to try and seamlessly weave advertisements into them. Most importantly, there’s no guarantee that what it gives you will even be true.