spinning_disk_engineer

joined 1 year ago
[–] spinning_disk_engineer@lemmy.ca 2 points 17 hours ago (1 children)

I think this is going to be the solution. It sort of feels like I'm making something that doesn't exist for a reason, but oh well.

[–] spinning_disk_engineer@lemmy.ca 6 points 1 day ago (1 children)

I'm not so sure, I'd rather have a prepopulated blocklist that I can further configure myself, as opposed to instance level blocks that I can't reverse.

[–] spinning_disk_engineer@lemmy.ca 2 points 1 day ago (2 children)

GOG doesn't have DRM. You can use a FOSS launcher if you like, this is just another option. I'll keep using Heroic myself, but I'm happy to have their effort on linux in general, even if I won't directly benefit from this application.

 

I've been working on a platformer game in rust. Game objects are stored in a Vec, and each is updated independently. However, a given object would like to interact with other objects. In C I would do something like fn update(&mut self, others: &mut [Self]). However, this would result in the parameters being aliased, which is not allowed.

What I'm looking for is some type that acts like a &mut [T], but remembers which element it is not allowed to access. I could use two &mut [T] values, built with split_at_mut, but this is unwieldy. I could make a struct that contains both. Is there a crate that does this? Ideally it would use unsafe internally so the compiler knows there is exactly one element in between. (that is to say, I would prefer something using only 24 bytes)

Chain does not do what I want because it can only be used as an iterator, not for indexing. To be clear, I would like to be able to index the result exactly like a &mut [T], except that it will find the element under consideration to be out of range.

[–] spinning_disk_engineer@lemmy.ca 3 points 1 month ago (1 children)

You make switching RF sound easier than it is, but this is quite possibly the solution.

That or taking the train, but those have their own problems (including not being available in all places).

[–] spinning_disk_engineer@lemmy.ca 15 points 1 month ago (4 children)

For as long as they let you. Watch, soon they'll be required to connect to the network at least every 30 days to function.

[–] spinning_disk_engineer@lemmy.ca 1 points 1 month ago (1 children)

That's possible. Everything after it is much faster, so I'd assumed it was reacting to some kind of failure event, but I'm not that knowledgeable about this. Edited for clarity.

 

EDIT: turns out I needed to set mountpoint=legacy, (sudo zfs set mountpoint=legacy <filesystem>) which is the standard way to use ZFS on nixos. Legacy in this context means that the mountpoint is decided by the system rather than ZFS; I guess ZFS was previously refusing to receive the mountpoint because it already had one, but I didn't notice because they happened to be the same.

I'm using ZFS on nixos, with an ext4 boot drive, and a ZFS pool whose key is loaded from the boot disk.

Despite mounting correctly, zfs also causes the system to fail and go into emergency mode. Even if zfs were to fail though, things like sshd and dbus shouldn't depend on it.

In particular, the system waits for about half a second after

         Starting Mount ZFS filesystems...

and the next line is

[FAILED] Failed to mount /ZFSmountpoint.

and then

[DEPEND] Dependency failed for Local File Systems.

A few lines later I get

[  OK  ] Finished Mount ZFS filesystems.

The next line is about emergency mode.

How can ZFS both fail and finish? And why does this affect the remainder of startup?

Plus, we can put up solar panels faster; we have the land. We don't (just) need more clean power in 2040, we need clean power now.

 

I'm not too knowledgeable about cellular equipment, but I was wondering how much of the phone based data actually gets to the SIM card.

Would it be possible, by sniffing the connection to the card, to listen to the content of calls and mobile data? Or would it be possible to get tower information in a similar format to rayhunter?

I understand there is some encryption, would getting the keys for that enable some of the previous ideas?

While this is certainly true, I'd like to highlight the difference between privacy and anonymity. Kagi is worse because they require payment information (though there is an option which attempts to separate your payment from your searches), but they are better because their privacy policy is better. If you don't trust the service itself, whether because of its location or any other reason, then you need to be paying in cash or monero, using free services. In both of those cases though, you also need to ensure you don't deanonymize yourself in other ways. In other words, pay if you don't want to think about too much, and you more or less trust the company.

I believe that even if it were GPL licensed you could do that, since you're only running it on your server, not publishing it. Of course, for this very reason, most copylefted server software uses the AGPL, which does require source code of a network service to be made available.

[–] spinning_disk_engineer@lemmy.ca 8 points 3 months ago (2 children)

While this is true, it's also worth noting that the original is permissively licensed. You can change it if you want, it's just that most people don't bother. Presumably changing the image would be as simple as replacing the three files in /web/static/img, but maybe that would look weird for some reason.

I generally agree, but I'd avoid gmail with very few exceptions. They are going to read your email, so whether your solution as another protonmail account, an alias, simplelogin, or another service entirely, it's almost certain to be an improvement. Keep in mind that you can tell gmail to forward your emails, for anyone who hasn't updated their address book yet.

[–] spinning_disk_engineer@lemmy.ca 10 points 4 months ago (1 children)

It is an option, but not a particularly good one seeing as it's run by amazon. I personally use Signal just because it's popular, but Wire seems equally good, and I'm glad they exist to provide competition.

 

I am trying to mount an nfs share from my local network using qubesos. I know that the same configuration works on a different debian machine, but it doesn't work under an app VM, creating the following error:

$ sudo mount -t nfs -o nfsvers=4.2 -o sec=krb5p <nfs share> /mnt/removable/ -vvv
mount.nfs: timeout set for Sat Mar 14 16:35:14 2026
mount.nfs: trying text-based options 'nfsvers=4.2,sec=krb5p,addr=10.206.0.206,clientaddr=10.137.0.9'
mount.nfs: mount(2): Invalid argument
mount.nfs: an incorrect mount option was specified for /mnt/removable

Is this a firewall issue somehow? I don't see how that would happen, since nfs should work over the internet. I have installed nfs-utils to the template, but I'm not completely certain that's actually what's being used.

 

I'm installing Qubes OS for the first time, and I was wondering if there is any disadvantage to installing librewolf (or something similar) as my primary browser instead of default firefox?

If I do want to do this, is there a recommended method beyond "install it as normal in the template"? I notice that when creating a new qube, there is a menu for additional applications; how exactly does that work?

For practical reasons I might also want to install some kind of chromium based browser (probably helium), but I wouldn't expect to use it much.

 

I recently decided that I want my resume to look better than the result of libreoffice. I installed texlive from the default trixie repository, and it works for the very simplest cases. However, trying to render a template which depends on CurVe resulted in an error about missing sty files.

By installing texlive from source, and installing CurVe to the working directory, I was able to fix that problem. However, there is still an error, and it appears to be an error in apa.bbx, a downstream dependency that comes with texlive. The error is

Package keyval Error: usenarrator undefined.

I'm not sure what I can do about this. I'm not very experienced with latex, mostly just using the default style as a convenient way to format math. Would swapping it for html be a recommended solution? Is there a good way from the command line to export html to pdf?

9
submitted 10 months ago* (last edited 10 months ago) by spinning_disk_engineer@lemmy.ca to c/privacy@lemmy.ml
 

I use mailbox.org. Mailbox.org provides an "encrypted mailbox" feature, which PGP encrypts incoming unencrypted emails. The server can of course intercept incoming messages, but it can't look at the entire backlog unless it was compromised the entire time.

Alternatively, using POP3 instead of IMAP (at least with the default settings) deletes emails from the server after downloading, whenever my laptop is connected. Thus, the server can intercept incoming messages, but not the entire backlog.

Of course, both of these have downsides. The encrypted mailbox is PGP, so it misses important details like the subject lines and source addresses. Meanwhile, POP3 can leave my mail entirely unprotected for as long as I'm offline, and it also means that I can't access it from anything other than my laptop, and means that I have to do manual backups.

Which is more important in terms of security, or should I use both? I'm looking for the legal perspective of law enforcement (In Canada and Germany, home to myself and my email provider respectively), but also that of some hacker who's trying to get into my (and everyone else's) accounts.

Would there be a server software that I could use to download emails from mailbox.org over POP3 and then provide them to all my own devices over IMAP? That might, in some sense be the best of both worlds. Right now, I am using both POP3 and the encrypted mailbox, but convenience is definitely not optimal, so I'd like to change if it can be done safely.

 

Jmp.chat provides an adapter (a eUICC device?) that can be used to store an eSIM on a physical card. I have such an adapter that I have already used for a plan tied to my identity. Is it safe to purge the chip, and then to reuse it for anonymous communications?

 

I'm looking into getting some domains for email, so I don't need to use the same few addresses for everything. In doing this, the domain name itself becomes the identity, but it's also entirely arbitrary.

What is a good method to choose domain names so that they look more or less normal? Catch all addresses can of course be detected in SMTP, but the idea is just to not look suspicious. Would anyone be comfortable sharing the constructions they use? (though not the domains themselves, for obvious reasons) Should I use subdomains for the things that can safely be correlated, (as spam defense) or is it better to only use different mailboxes on one domain?

 

What part of an android device is responsible for carrier locking? I understand that custom ROMs can sometimes be used to bypass a carrier lock, but how might I go about making such a ROM from stock?

This is complicated somewhat by the fact that the actual device I have, an Orbic Speed RC400L, isn't technically an android device; it runs busybox/linux, but its bootloader, adbd, and cellular systems seem to be android-like. I'm happy to experiment if someone can provide resources for the general android case.

13
submitted 1 year ago* (last edited 1 year ago) by spinning_disk_engineer@lemmy.ca to c/privacy@lemmy.ml
 

I am trying to get rayhunter, but I am confused about a few things. What kind of hardware may be used? The Orbic RC400L advertises itself as being for Verizon network only, which I am not going to use. Does the rayhunter software allow it to work on arbitrary networks?

Furthermore, I don't see a good way to acquire such a device. They aren't sold locally as far as I can tell. On amazon they are available, though expensive. However, that would mean losing any form of privacy. I don't need it to be completely anonymous—someone looking at the location could probably see home and work easily enough—but better than amazon would be much preferred.

Are there any alternate devices that are known to work well in Canada? I will be keeping the device powered off most of the time anyways, (so as to hide my location) but it must be reliable when I need it.

EDIT: eSIM support is a plus, but not required

 

I store playlists as m3u files, but sometimes I want to play it on an external device or send it over the network. The playlist references files using absolute paths, and they have various formats (wav, flac, mp3, ogg).

I would like to convert it into a single cue/flac, as if it were an album that I just ripped from CD (though it may be longer than 80min.) It would be acceptable to create a directory full of uniform flacs with appropriately adjusted metadata, again as if ripped from CD.

I would like to avoid writing intermediates to disk unnecessarily, and I would like to avoid GUI if possible, but the only hard requirement is that it run quickly (that is to say, faster than realtime.)

view more: next ›