Permanently Deleted
Chobbes @ Chobbes @lemmy.world Posts 0Comments 465Joined 2 yr. ago
Chobbes @ Chobbes @lemmy.world
Posts
0
Comments
465
Joined
2 yr. ago
Deleted
Permanently Deleted
Jump
I like Haskell, but the syntax is probably the worst part about it. The ability to define your own infix operators with arbitrary precedence / associativity is really cool and useful, but can make it a complete mess to read because then you have no idea how any of the operators combine. I vaguely like the syntax, there's something kind of clean about it, but frankly if it was just a lisp it would be so much easier for people to pick up (aside from the fact that nobody would because it would look like lisp).
EXACTLY. Academic publishing is actually kind of insane. It’s not like buying the paper even funds the research, it just goes to the publisher… and sure, you could argue that they do a service by basically being a muster point for peer review… but it just feels SO WRONG to put research (especially public funded research) behind a paywall.
The case for movies and games is absolutely a more contentious issue, but there are some similar arguments for the current system being a little dated (especially with respect to copyright law and how long it takes before works enter the public domain). All things being equal (e.g., assuming the creators can still make a living) the more people who have access the better! Of course it’s far less important that you can watch the latest movies or whatever (vs somebody accessing a research paper about a disease they have or something), but it’s still a bit of a shame to restrict access, and it feels like your ability to own copies of media is being eroded over time… And maybe that is worth caring about a little bit?
Absolutely! It’s also frustrating that you can’t just buy a digital copy of a movie or something for a relatively cheap price? I just want the digital equivalent of bargain bin DVDs and libraries. I don’t want to pay $20/mo to not own anything?
It definitely seems more idealistic when it’s academic papers instead of Marvel movies, haha. Still, I guess it feels like the only way to get the upper hand for some people, so I can kind of understand the catharsis. I think for most people it’s just “get shit free”, some people feel like it’s a small act of rebellion, some people want an excuse to feel better about it, and some people probably are actually more idealistic about it.
I think there are legitimate questions for society in relation to piracy, though. Copying and transmitting data is essentially free in the modern era, and I feel like we still haven’t really figured out what that should mean. The issue of ownership is getting really weird in the digital age too… Like you used to be able to lend and resell things, and libraries with physical media were a simpler concept… but these things are tricky and cause problems when copying and sharing is so effective. I think there’s arguments to be had that this a pretty fundamental shift and that we should potentially reconsider how media and other things work economically.
It does feel like a lot of people trying to justify their piracy are just looking for excuses for something that they personally feel is wrong, but they want to do anyway… but I do think that free access to data and information is something that people can legitimately care about. Libraries are a good socially acceptable example of this (and they definitely fall under the “get shit for free” movement!), and it’s not entirely crazy or cringe for people to want more libraries in life and fewer walled gardens.
I've never heard of anybody relaying just the Microsoft e-mails, but that's a really funny spiteful solution.
Lately I've been able to send to outlook just fine (maybe it's just dumb luck, who knows). I think I had troubles initially because they're really picky about rDNS matching the MX exactly. I also signed up for SNDS just in case, but I don't know if they factor that in...
It would take some time to turn around of course but the same was probably the case for https, 2fa, ipv6, and tpm’s.
Oh yeah... Definitely good IPv6 support everywhere. That really turned around, and we're not dragging our feet on implementing IPv6 at all 🥲.
Oh, they are. I keep telling people to WRITE DOWN YOUR PASSWORDS, and NEVER use same password on two sites. They dont listen. Its a lot easier to just remember 1-4 variations of a password and use that than carry around a password notebook. And they think themselves safe.
Honestly, the best solution for this is a password manager and not a notebook. The average person is not going to come up with strong passwords on their own for every website. A password manager once setup can be more convenient than whatever they were doing before, so if you can get people to use one they'll be in much better shape.
I’m thinking most people shouldnt use passwords at all anymore. They are a huge point of failure because people are people. We need something else to be the norm. How can we make hardware keys or something the norm for logging in? Have everyone carry around a bankcard-like thing that fit into every computer where people need credentials. Would’nt that be safer while still being accessible and convenient?
My understanding is that this is basically what the whole passkeys initiative is. I have sort of mixed feelings on it. Hardware tokens for logging in is great, but I worry about people stealing the hardware tokens from others. Mostly people are going to use their phones, though, which should have some other mechanism of authentication.
I had a similar situation with my health insurance company, except I think they added the character limit a while after I had set my password T_T. So, it worked for months, then they changed the mobile app so I couldn't enter a long password... And then eventually they changed the website too and then I couldn't log in at all. Thaaaaanks.
Jesus, I thought you were joking, but I looked it up and almost did a spit-take...
- https://www.yodeck.com/news/raspberry-pi-5/
- https://www.raspberrypi.com/news/introducing-raspberry-pi-5/#comment-1594037
Wat. Isn't that a fairly big deal?
You can absolutely use nix and home-manager without NixOS, you can even use it on a Mac!
Most of the tutorials I find are tailored to NixOS
I think that's less true than you think. There's definitely a lot of tutorials from people who do use NixOS... But this mostly doesn't matter, using nix is more or less the same regardless of operating system. There is some stuff specific to NixOS, but that's mostly just using nix to configure system services.
Is there an easy way to learn this for just the package manager?
That said... There's not a super easy gateway to learning nix, unfortunately. The easiest way to learn is probably to try to package one of your personal projects, or by learning home-manager and trying to fully declare your home environment. Packaging programs in certain languages can involve having to learn more about the nix ecosystem than others (e.g., python is a little more complicated than what you'd need to package a simple C program, because python has it's own ecosystem that you have to integrate with).
Probably the best introduction is the nix pills if you want to learn how nix works (not necessarily how to use it, but it will give you a really solid foundation):
https://nixos.org/guides/nix-pills/
Though, it's really common for people to bounce off of the nix pills because they start kind of slow. If you can follow through them, though, you'll have a good understanding of the nix language and how it's used to construct derivations for building packages.
The nix language is much maligned, but it's actually not as bad as you think going in (error messages can be ATROCIOUS, however). It looks a bit weird, but I think if you go in with the understanding that it's "Turing Complete JSON" you'll have a better time. Almost everything you're doing in nix is building up attribute sets (which are basically just JSON dictionaries) which specify the dependencies and build stages of a project... But it's not a static representation like JSON, you can call functions to do arbitrary computations to build up fields (e.g., I have some functions to automatically set up wireguard configurations from a description of the network layout of my machines, so it's useful to be able to do this stuff).
There's a version control system called darcs that stores its configuration files in a _darcs/ directory in your project and it wounds my soul.
A filesystem is just writing to random unused blocks with extra steps.
May depend what you want and where (location can matter a little bit for latency critical stuff, but streaming video won't care), and what operating system you would run on it. The Hetzner ARM servers are pretty cheap for what you get (and it looks like they include 20TB of bandwidth). I've been pretty happy with Lunanode. I think people often look here for deals: https://lowendbox.com/ they often recommend Racknerd boxes... I think there's some affiliation with Racknerd and lowendbox.com, but I threw something on a Racknerd machine recently and have had a good experience so far. You may want to do some research if you want to send mail directly from these machines. Not everybody allows it (sometimes you just have to ask), and I hear tell that sometimes you can end up with an IP somebody spammed with before with a bad reputation.
I was considering a VPS! That said, if I’m say, accessing my jellyfin library externally through a VPS, wouldn’t that just end up costing ludicrous amounts of money?
Depends on your usage, but probably not? If you can transcode on your jellyfin server you'll be able to serve lower quality versions remotely if you want to save bandwidth... But most VPS's provide around a terabyte of bandwidth per month by default. If you use more it will cost more. I think it's usually fairly cheap to get more, but if you're the only one accessing it you're probably not going to use that much. Like if you rip a blu-ray you might end up streaming a 50gb or so file for a movie, but that's only a twentieth of the bandwidth allotted to you (roughly)... Plus if you reencode it to something smaller before putting it on your jellyfin server, or if your jellyfin server can transcode fast enough you can send a smaller video stream to your mobile devices or whatever.
I don’t use Arch btw ;)
I don't either, that article was just what I found that mentioned setting up Tunnelbroker with a dynamic IP.
Glad it was helpful! I was worried I'd be a little off-topic talking about self-hosting e-mail instead of this Anon Addy thing. Hope you find a solution that works for you soon :).
And yeah... Unfortunately if you you're behind CGNAT and don't have a static IP I think doing this for free on your existing internet connection might be challenging. One thing that people in a similar position might be interested in is Hurricane Electric's free Tunnelbroker service, but I think you might still be out of luck behind CGNAT.
You'll be able to get public IPv6 addresses for free and can allocate them to your home network. You can set it up to dynamically update the IPv4 address on your end... But I think if you're behind CGNAT you can't do that, unfortunately. Another problem with this approach for something like a mail server is that not everything speaks IPv6... If a sender only supports IPv4 they won't be able to send mail to you.
I think behind CGNAT pretty much your only option is to pay somebody for a real IP somewhere. Either a VPS somewhere where you set up wireguard (there are cheap options for this, and then you can run other things on the machine), or a VPN with a dedicated IP.
Oh god, yeah. I personally would not try to self host e-mail or any service that you need other people to be able to reliably connect to without a static IP. As to losing power... In theory mail servers are supposed to queue mail and resend later, and you can also set up a backup MX that will queue mail for you (senders will automatically switch to the backup mail server if they cannot connect to your primary one). There are even free services for backup MX http://www.junkemailfilter.com/spam/free_mx_backup_service.html (though they use this to train spam filters, so if you have privacy concerns you may want to avoid it). In the past I have had some prolonged downtime on my mail server and I have noticed that some senders will give up entirely and never send to that address anymore (which I think is poor form on their part, especially since somebody could register that email account later). I've since setup my own backup MX to avoid these issues, and it's worked great when my primary has had network issues (needed a spare box for backup nameserver and stuff anyway, haha).
You absolutely can use an external mail service as a catchall with your own domain. For instance protonmail has support for this:
https://proton.me/support/catch-all
You'd have to look into the pricing and read the fine print, though. A lot of mail providers charge per inbox and I'm not sure if they'd charge extra for catchall services or not.
Relays do cost money, though I think some have a free tier for small volumes of mail. You might also see if your registrar or host provides anything for email.
The easiest way to do this is to start with just receiving email and not worrying about forwarding, though. You can host your own imap server and just have a catchall account that’s separate from your main email to start, and if you really want to forward you can worry about send later. Receiving email is easy, the thing that people struggle with for email is sending because there are a few requirements like dkim / spf / DMARC and reverse dns that you might not know about and may configure incorrectly and feedback is hard. Also if you have a residential ip I’ve heard it can be harder to send too. If you’re just forwarding to yourself, though, that’s probably a little easier because you can test more easily / mark yourself not spam. If this is your use case I wouldn’t worry about setting up a paid relay service. You don’t need it unless you really want to forward and have troubles making send work in your own.
With all that said maybe anon addy is easy to set up on your own and gives you what you want. I wouldn’t know! I’ve never used it before.
I have a catchall inbox so I can just make up any email I want and everything gets forwarded to the catchall inbox. It’s pretty easy to set up if you do host your own mail server (which is relatively easy for receiving mail). Obviously this doesn’t integrate with bitwarden or anything, though. If you want to forward emails to your main email account on a big provider you’re going to have to make sure your server can send emails you can potentially use a relay service for this, or just set it up yourself (you’ll mostly just need some DNS records for SPF / DMARC / DKIM).
... Maybe I'm too Haskell brained to understand why this is a problem, haha. That looks fine? I don't know Unison at all, but you can probably elide the type annotation too?