It greeted me with a message: "Oh, I see you disabled JavaScript. Keep up the good work, my fellow cleanweb person!" which is an interesting departure from the usual "this app won't work without javascript". But I couldn't select the text from the message to paste it here... while looking at the header above it "Just let me select text" I thought: yeah!
I used an Indy with X11 forwarding for a while, I could run a modern Firefox from my Linux machine, which worked nicely. As much as it looked like it was running natively, it really wasn't, so downloads would end up on the other computer, and sound would also play remotely.
Because I never throw anything away I still have a small bunch of these machines in the attic (apart from the Indigo they are all teal though) waiting for this kind of treatment, which have been super high on my to-do list.
Well, if you’d ever be interested in lightening your collection by one for a fellow enthusiast, I’ve always wanted to get my hands on an SGI to restore/rebuild. I’m currently working on an IBM 5150+5153, but the workshop always has space for more.
I don’t know what it is, but there’s something just so mystical about SGI’s computers. I don’t know animation, and never got to work with IRIX, but at some point I want to get my hands on a Silicon Graphics computer (as well as a Sharp X86000).
I recently upgraded my gaming PC from Win10 to Bazzite with Wayland, and my use cases are primarily streaming to a laptop and to a quest 3. I don't find these very niche use cases tbh. The challenge was, unfortunately, getting around Wayland. It may be a security feature, but to me it feels a bit like making a car safer by removing the wheels so you can't crash it.
It could ask for permission or let me whitelist applications like Steam, Sunshine and ALVR, but it doesn't. I ended up plugging a dongle into the HDMI port, so there's always a monitor "on" and it now works perfectly fine, but I would have been happier if I didn't have to do that, like Windows before it.
Maybe there are good reasons for Wayland to be the way it is, but my experience is that it's unnecessarily limited, it's in the way and "the rest" doesn't have these problem (Windows, Xorg, ...) so I'm not particularly fond of Wayland, even if it does a lot of things fine and it is quite stable (in my experience).
I've had wireguard in a container for a few years, and it's never failed me. I will say it took me a long time to get the firewall part of the configuration right but the configuration is otherwise simple. When I'm on the road I can access all the things I self host, which I don't have to expose anything of to the outside world.
The amount of people here just exposing their network to Tailscale, and recommending others to do the same, is surprising, to say the least.
I've set up Wireguard on a VPS once six years ago, and nothing needed adjustment since. It is as easy as you make it out to be, and depending on the use case the firewall rules can also be simple.
If I need to add a new device, which is probably a rarity for the average user, and once a year for me, it takes two minutes to edit two files and restart a service.
I can see reasons why one would want to use Tailscale, especially in an organization. But just uncritically recommending it for home-lab like setups seems as harmful as pushing people to Cloudflare for everything.
Inter-node mesh with raw Wireguard is an exercise in patience to say the least; I have a few different colo sites, my house, my phone, LTE/5G hotspots, raspberry pi projects in the field, etc that I want to fully connect together.
Raw Wireguard is fine for a road warrior or site-to-site VPN setup as is common, but when you want multipoint peer-to-peer connections without routing through what might be a geographically distant point, magic DNS, etc, Tailscale really shines through.
Although at that point I'm sure you, and any similar user, would not actually rely on ad-hoc advice like in this thread, and instead just evaluate what is needed.
As an aside, personally speaking, headscale solves basically none of my concerns associated with introducing more software, complexity and third parties (the maintainers) into my network setup.
Less so because of paranoia towards the software/product itself, and more so because of the increased surface area to attack.
But I also think that anyone actually bothering to set headscale up probably falls into the aforementioned group of people that actually thinks about their requirements.
I've been using Netbird on my home network and on my daughter's laptop to provide remote support while she has been at college. This year she moved into an apartment, which has its own cable modem and router/network that I set up. I haven't figured out how I will configure a "zero-trust" architecture that will allow me to act as remote support for her remote network. I'm not the best at networking and I'm afraid of connecting the networks in a manner that I don't expect. I'd be interested to hear if anyone can suggest how to configure this arrangement. I've always had her leave the Netbird client on her laptop turned off unless she is specifically asking for help. I plan to do something similar, where I would have her remote network normally disconnected from whatever VPN bridge network I set up.
I have a VPS and have thought about using Wireguard on it for accessing my home network, but I worry that I don't understand the security well enough to use it. Wouldn't less experienced people like myself be safer with Tailscale or Netbird or something that doesn't require extensive knowledge of a publicly-hosted server?
whoa that's super useful. I've been trying to figure out what I'm going to do to let my family access my server. What client do you recommend on the phone end? Or does the phone support connecting to wireguard out of the box?
The best client is from WireGuard. It’s super efficient in my experience. It even supports on demand VPN where you can define network it should activate or deactivate the VPN.
It's not related to Gnome Boxes. It is an application that makes using Wine easier and more robust.
The statement means to say that it allows you to run Windows applications inside an isolated environment (a "bottle").
My favorite feature, especially considering the past of "heavily tied into Visual Studio" is the platform independent nature of (most of) "modern" dotnet. That is to say, I run Fedora on my laptop, and I do not want Windows, yet I don't feel like a second class citizen and dotnet runs well and is available from repositories that Microsoft provides for various Linux distributions (it just hit me how strange this is, for _Microsoft_ to offer official repositories for Linux operating systems...).
I also really like that its development is fairly open (it's on github).
From a more technical point of view, I think C# is a slightly better Java. A pretty nice language overall. It's not gui-or-nothing with dotnet core so it's not too difficult to create workflows for ci/cd pipelines, although the .net framework solutions we still have aren't too much harder to script: msbuild is pretty capable.
I work at a small web company (.net based, Netherlands) and we're just experimenting with it. We have a paid copilot subscription, but nothing about it is mandatory in any way.
But this place is conservative in the sense that self hosting is the norm and cloud services like Azure or even github (we self host Gitea) are not, other than MS 365 for Teams and e-mail.
In my first job, when .net didn't yet exist, xml + xslt was the templating engine we used for html and (html) e-mail and sometimes csv. I'd write queries in sql server using "for xml" and it would output all data needed for a page and feed it to an xsl template (all server side) which would output html. Microsoft had a caching xsl parser that would result in less than 10ms to load such a page. Up until we though "hey, let's start using xml namespaces, that sounds like a good idea!". Was a bit less fun after that!
Looking back it was a pretty good stack, and it would still work fine today imho. I never started disliking it, but after leaving that job I never wrote another stylesheet.
If you type "login" at the ha> prompt, you'll get a root shell.
This is something I also had to accept about HA. It runs in a VM in my case so it worked out-of-the-box, but you don't just ssh to it after installing it, and the ha> prompt is just a bit different. So far, it hasn't been in the way, although it occasionally takes time to find out how to do things.
It's very flexible though, and apart from devices in your house there are many outside sources of data to use, like weather data, sun elevation or trash pickup dates. The HA app on your phone gives you many sensors usable in a flow.
The time spent on it usually results in something worthy of that time, in my experience.
In case of Earth, Wikipedia describes [1] it as being "[..] generated by electric currents due to the motion of convection currents of a mixture of molten iron and nickel in Earth's outer core". This makes Earth a geodynamo [2]. (The aforementioned Wikipedia page is actually really long and detailed, a lot more than I would have thought)
