antonyh

Love/hate systemd as I might, it's been rock solid everywhere I've used it, and I've used it heavily. It has it's quirks, as does the init-scripts that came before, and launchd on OSX (not sure what the modern equivalent is for MacOS).

However, the systemd journal raw format is binary data and would much rather a plain text log. All things being equal I'd rather deal with human readable files.

I got the same email on the same date. Unsubscribe told me it was from the 'important announcements' list - I fail to see how this could possibly fall into that category.

I guess I can't have important announcements from Proton in the future if it's polluted with these low value messages.

Epiphany (aka Gnome Web) on Linux shows this correctly, as expected for a Webkit-based browser.

Thanks for confirming, I look forward to one day getting Wayland to run properly for me, but that's a whole different story. I'm just glad that middle click paste works and that it's not really an 'X11-ism'.

Wonderful, another default I'll have to change to have the correct behaviour for me.

However... "This is an X11ism, originally an xsetting 1 which frequently results is in unexpected behavior when people pressing the middle mouse button."

Does this even work on Wayland? I'm still using X11 for many reasons, including highlight/middle-click-paste.

Nothing on the scale of most the things mentioned here, but I'm trying to assemble the first version of my digital garden to publish online a bunch of notes I've collected over many years. I'm also trying to put together a workable system to catalogue and index hundreds of thousands of digital images scattered across multiple devices so I can deduplicate them and collate them effective. Digital Librarian is not a hat I ever thought I'd end up wearing, but I refuse to buy more 18Tb HDDs and still not have any means to locate pictures in a meaningful way.

I can run 99% on Linux + AMD + X11, on Ubuntu 24. The only problematic one is Witcher 3 which insists on Wayland which breaks everything else. Even the UT5 titles work well enough, but then again our gaming is not your gaming and you might be more demanding that we are.

I'm trying to recall how NVidia behaved for games, but my daily driver is an old 1050 Ti that's been rock solid for years now, also X11.

Maybe the problem is Wayland not NVidia?

Having failed to get Wayland working on Debian Trixie with a 1050 Ti as an upgrade from X11, I've given up for now and will try again when I switch to AMD. This is a workstation not used for games so it'd be good to have Wayland working right but I'm not wasting time fighting it, and it'll get the GPU from the gaming rig when it becomes due nullifying the problem.

What I don't get is if these are proscribed steps (and they do read as such) why are they not automated with the module install? Why are we still fighting these issues if the 'workaround' is linear and well described? Is it as flimsy a reason as "write-an-article, collect-advertising-revenue" rather than contribute code to the installer?

Implying that Linux gamers have to play with cheaters. It may as well not be playable at all.

Maybe not, but they are approaching it. I wouldn't use it for anything funded with my own cash, I no longer recommend it as a first choice, but I'm not suggesting it gets replaced yet. It's somewhat in the 'legacy tech' category now in terms of how I perceive it and deal with it.

How much is BigAI paying Moz to integrate these things? I has to be about the dollar surely?

OS/Xamp, pronounced Oh-sex-amp :o)

If it doesn't run on Windows, how can it call itself Winamp? Macosamp makes more sense.

It's a clone with a stolen name, and given that it's not been abandoned it's likely they'll come down on this like a bag of hammers.

Apple Music would seem the obvious choice.

Absolutely is fragmented. Even though I own a Wii I've never played Zelda or any Mario games, and I don't think I know anyone who owns a modern Nintendo. We all live in bubbles. And we change bubbles occasionally; I no longer play Fifa or CoD mostly because of the kernel anti-cheat. I got bored of CSGO. I play less gory games now because of family. We play less Lego games because we grow up.

The problem for me was more a software issue than hardware or cost. It (Quest 2) just felt awkward, the software selection was meh, IPD bad for me, resolution only so-so, but most of all it felt isolated from the rest of my gear. I've been planning to give it another go recently but don't have much enthusiasm for it, but a Steam headset with my collection is something I'm very keen to try.

Probably, I'm not sure it's not https://gcscc.ox.ac.uk/

I don't think it's https://www.infosec.ox.ac.uk/

There's also this AI security research lab, https://lasr.plexal.com/

It looks like Oxford are quite busy in this space.

I have a Quest 2 that I don't care for, so it's caught me somewhat offguard how much I want this. The whole set please. I thought I'd just want the controller, but everything here looks exciting to me.

Rumour has it that after the Crowdstrike fiasco future versions of Windows won't allow kernel level modules. I can only hope this is true if it kills off the main reason titles don't work on Linux as a side effect. I'd have bought BF6, some version of EAFC, and more.

I don't think they meant OXCIS, that seems to be a centre for Islamic Studies https://en.wikipedia.org/wiki/Oxford_Centre_for_Islamic_Stud...

I can't quite work out who they donated to - it seems there are a number of Oxford Uni cybersec/infosec units. Any idea which one?

I see what you mean, that one really is ordered chaos.

I had a load of stickers on my old machine. I kept the laptop but felt the need to remove the stickers. I think part of me had moved on from what they represented - mostly past interests. Another part of me never liked how they overlapped, I tried but it didn't really work for me. It raises the question: are stickers randomly placed or is there some deliberate pattern to them that I just can't see? Do you just slap them on, or precision place them to look random?

I'm using Joplin sync'd via webdav. It's barely adequate but gets the job done.

I 'cancelled' down to a non-AI family sub, but ended up full cancelling before the kick-in date for the annual renewal. I'm done with the nonsense and dark patterns. The shenanigans has worn thin.

I'm tired of the price hikes for AI - if it was good enough without AI, it's should be optional and a separate sub. It's an enhancement not a core feature.

The questions I have mostly centre around how much precision of power delivery it has - it is an all or nothing proposition, can it deliver 0.1% smoothly for real world use, and what is the MTBF / duty cycle / failure mode? I would imagine the last thing anyone would want is a locked wheel, or only one wheel delivering that much power. I know this is unlikely, but as someone with a 22-year-old ICE vehicle I do tend to take the long view on these things and want to know how they will fail as much as how they work. Same applies to the Tesla motors - is there much information on failure modes publicly available?

I loved the Affinity apps, but not on a subscription. I don't do subscriptions or 'freemium' for apps, only for content. On desktop I'll move to Krita / Inkscape / Scribus. On iPad, not sure yet. The key here is watching out for proprietary formats that won't open without the app or subscription; this is something we are going through with Microsoft Publisher offboarding right now to ensure we can archive and read the designs in the future. The same applies to Affinity - if you value past creations, export them to a common future-proof format.

Because it's opinionated? So maybe there are scripts that use sudo, and perhaps he needs more than 3 tries to fat-finger his password?

Personally, my opinion, I use sudo, and if I take more than 3 goes then I deserve a timeout to get my act together. Anyway, 10 attempts isn't enough to brute-force a decent password, and if bruteforcing is a concern then add 2FA codes or hardware.

There's more serious concerns in the article though - the part about the screensaver / hyprlock? That's just security theatre.

I don't see the problem here in theory - if I want to trust something fully I'll build it myself in my own pipeline, often with additional hardening as needed. It only needs scripting out the build process to fit alongside my other code. I even do this for Linux apps like Signal because I want a clean binary that matches the Git tag, packaged exactly right for my system, built with the libraries already in place locally.

What's not cool is not pushing a fresh Docker image to secure the CVE, leaving anyone using Docker hanging. Regardless of the new policy, they should have followed through and made the fix public on all distribution channels. Leaving a known unsafe version as the last release is irresponsible.

They do this, only using the Latin alphabet for brand names etc. For example, pravda.ru spells Finland as Финляндия

(Edit: I misread your comment, fixed mine, I agree with you)