I want threads to succeed so badly. It's the only app I've ever actually liked the UI of. I've tried numerous Mastodon apps (Ivory included), BlueSky, Twitter, and all of their web variants. I find myself returning to Threads even if the functionality is limited, because it's just that enjoyable. I'm rooting for 'em!

I chose Rust here because that was what the command line tool that I referenced were written in. I'm not going to argue on HN about memory management (because I too, prefer Zig over Rust) because it's all about the program you're trying to write.

In terms of generating the object file with a single toolset, It's just not something I care about. This was like, a two day project. I don't need to make it more complicated than it needs to be

I haven't used 'Next Content', but one big difference is that Astro is framework agnostic. I'm not a huge web dev so I wouldn't know how different these frameworks are, but not being locked in/being able to try another framework is cool

It looks like it's another framework to me. The basic example looks exactly like Nuxt to me as well! (component, pages, and layout folder!)

To answer my previous question, they have a page talking about the similarities and differences here: https://docs.astro.build/en/guides/migrate-to-astro/from-nux...

They look really really similar to me (Nuxt content and Astro)

Astro is akin to other “meta-frameworks” with SSR/SSG like Next/Nuxt/etc, but it’s client-framework agnostic so its meta-ness is too. What primarily sets it apart is that it’s designed to ship just HTML/CSS by default, with any client side JS being opt-in where a client component is invoked (termed “islands”), and those components being authored in any client component library of your choosing (if an adapter for it exists, and many do). Or even vanilla JS, but then Astro is more akin to Jekyll or other similar SSGs.

Were you thinking of Level1techs?

https://store.level1techs.com/products/14-kvm-switch-triple-...

I'm 90% sure it was, but it looks like they did a major UI update to the site so it's not triggering the "that was it!" lightbulb in my brain.

I use one, so I can confirm they work extremely well, subject to some caveats:

* The total cable length is important, both between the host / KVM and the KVM / monitor, as well as any daisy chained displays you have. I had to use certified cables to get everything working reliably with my setup.

* There's a weird interaction with BIOS power on. The boot display drivers I have freak out if they aren't the active display and fail. I solve this by switching the KVM before I turn the computer on. After everything is booted into an OS, it works fine to switch.

* Power supply quality is important. I had some issues before I made sure the power supply was reliable.

KVM switches are just inherently difficult little devices. I haven't had issues since I got it working though.

>There's a weird interaction with BIOS power on. The boot display drivers I have freak out if they aren't the active display and fail. I solve this by switching the KVM before I turn the computer on. After everything is booted into an OS, it works fine to switch.

Do you have an AMD GPU by any chance? I have the level1tech 2-head DP 1.4 KVM, with an AMD RX 560 on a Linux host, and after updating to kernel 6.4 recently my computer now boots fine without a monitor attached.

I had a similar issue where a display had to be _on_ and _connected_ (i.e: active on the KVM) at boot time, or the GPU wouldn't work at all. I could get in via SSH, so I tried various amdgpu recovery options, poking the device to reset it, reloading the kernel modules, etc., and never had any luck. I just lived with the quirk. It was problematic because if you left home with the KVM selected on the Windows guest, and needed to reboot the Linux host remotely, you'd come home to a non-functional Linux desktop.

I have a similar issue with a Nvidia 1080Ti on my old desktop. It's related to the UEFI deciding if the iGPU or the Nvidia GPU should be primary and to disable the iGPU or if the iGPU should stay enabled.

Curious how you went about determining that was the source of the issue.

I'd swap the cables around and could see video from the iGPU output.

If this is the problem, you can usually force the GPU choice one way or another in BIOS.

Nvidia on both, one consumer and one workstation. Neither CPU has built-in graphics iirc, nor do the motherboards expose the ports.

Alternatively, you may have been thinking about ConnectPro. I ordered a kvm from them around the same timeframe and it was delayed quite a bit from backorder. (Though, they also did a major UI change, so might not be able to tell either).

Two thumbs down for connect pro. I ordered their top of the line 4 computer, 2 monitor DisplayPort KVM and it took months to arrive. I could not cycle between inputs using the buttons. They were more like a suggestion to use that signal path; I would constantly need to power cycle the kvm, monitors, or both.

I ended up ditching it on eBay at a significant loss for a $30 usb switch and just switch monitor inputs manually. Far cheaper solution and way less fussy.

I had the same issue with this device. I ended up writing some code that you could run on a machine to operate the switching via the RS232 port: https://github.com/timgws/kvm-switch/

Bonus for adding 'glide and switch' functionality, so you can move the mouse to the edge of the screen and it would jump the input to the next display in your layout. It's like a hardware version of Synergy.

Very finicky device, but if you don't touch it - and you don't use any of the shortcuts - it works.

Neat! I used to use ShareMouse (pay-ware, if you want more than two machines tied together) for this, because setting it up and keeping it working are so easy compared to whatever version of Synergy existed at the time.

Synergy made me manually configure my monitors by dragging little boxes around in a window, would frequently refuse to connect, would spontaneously disconnect, and repeatedly mangled my config such that I had to keep manually configuring it over again.

With ShareMouse, it was "open a copy of it on each machine, slide mouse in direction of next machine, then optionally enable encryption (to prevent other users' instances of ShareMouse from being able to attach)".

I should also add that their customer support was totally worthless. They promised me a firmware update, and stopped responding after I confirmed my firmware version - the process to get that value was already quite arcane, so ultimately I felt like they never really had any intention of helping and were simply stalling me out.

Would you mind elaborating on the complications you ran into? I ask because I think you used std.HashMap instead of std.AutoHashMap, the latter of which automatically chooses a hash function based on the types provided.

It was a little over a week ago now so my memory is a bit hazy, but I'll try to the best of my knowledge.

Without getting into the weeds of why (happy to do so, just want to keep this readable), basically I needed to define and populate a hashmap in a new script and then import it into my main script, which to my mind left me with two options:

* Define and initialise it at the same time (my preferred method) as a constant. I don't have much to say on this as iirc, I had no luck with it at all, never even got close.

* Define it in a (public) function, add each field in with "put" and then return the hashmap. I tried with std.AutoHashMap and various other things, but to what I could work out there was no type of hashmap, so it wouldn't accept my return type.

I think for the first point you wanted a block that evaluated to the map, unsure if that’s what you wanted though

I think so, if I write it out in pseudo code it might make more sense. What I was trying to do was pretty much:

    const mymap = {1: "hello", 2: "world"};

But the only thing I could find any answers for was something more like:

    const mymap; mymap.put(1, "hello"); mymap.put(2, "world");

Here's how you return a hashmap from a function:

  fn buildMap(allocator: std.mem.Allocator) !std.AutoHashMap(u64, u64) {
      var result = std.AutoHashMap(u64, u64).init(allocator);
      errdefer result.deinit();
      try result.put(10, 100);
      try result.put(20, 200);
      return result;
  }

Your #1 option should be possible once Zig has comptime allocators -- it's on the roadmap, but not possible yet iirc.

If I remember correctly I tried almost exactly that, I think it was just the bang operator I missed out (obviously essential in this case).

Cheers for letting me know though!

I’ve seen https://tildes.net/ pop up but it’s invite only atm I think, would totally be down for an invite from anyone here though :)

I liked the discussion there. Some people claimed they were subject to capricious bans though.

Invite, slightly obfuscated: uggcf://gvyqrf.arg/ertvfgre?pbqr=LD8C3-NL8XQ-Q28BV

Still unused, in case anyone is wondering whether it's worth copying into a decoder. (Please comment when you use it btw.)

I snagged it, Thanks!

If anyone here wants to send me an invite, my email is my first name at gmail.com.

My first name is 'lelanthran'.

a few first come first serve:

https://tildes.net/register?code=UBZX0-DBQOR-H2W6A

https://tildes.net/register?code=IYUPP-2VZHD-EPZWM

https://tildes.net/register?code=TW3P3-YK0LZ-L7FLY

https://tildes.net/register?code=MAUNA-6IE43-OY86X

https://tildes.net/register?code=8RC2T-40G03-AFSYF

https://tildes.net/register?code=INH4U-7HC5H-AESTL

https://tildes.net/register?code=92S7J-WQR3B-I7OQJ

They are all used now, unfortunately.

Those who got invites should be able to invite more folks

> "Most of the existing users have the ability to invite others, so if you know someone that has an account, you can ask them for an invite."

https://tildes.net/login

Takes somewhere between 48 hours and a week to generate the first invite codes, though. (Those are the two closest moments at which I checked before and after I had invites available)

Do you know if a single account can invite people all the time or there's some limited number of generated invite codes?

Thanks!

The discussion there looks very good

Tildes is great but it's a lot more analagous to HN than reddit.

The moderators are limited in number and absolute in power, you can't make new subs (I asked for a few and was declined), and the community is more tight and discussion-oriented (which is a good thing IMO, but I never quite felt like I fit in so I don't comment often).

It has some nice things going for it, but it has a few issues that can be off-putting.

1. People are discouraged from being too negative about posts, but posting negative articles is fine. This led to people occasionally linking articles containing invectives about some subject they didn't like, but you're not supposed to respond to the article with invectives. It produced a natural asymmetry in the conversation and felt really passive aggressive.

2. The long post format is a noble goal, but personally I couldn't see the incentive in putting in a ton of work to craft a thoughtful, researched, and thorough post only to have it read by maybe 20 people then buried after a day. If I'm going to put that much work into a piece of writing, I'll create a blog where it will at least have some more permamence. I think in the end the long post format attracts people who enjoy the process of writing itself--which is fine--but it simply doesn't equate to higher content. A lot of the opinions in those long posts are still underbaked, and the length doesn't improve them. Tildes threads can be a grind to read while producing little information.

I can...but where?

I'm really enjoying tildes. The community/culture is correct. Just needs more content (which will come with time).

I as well would love an invite.

Yeah. Seems like it could really benefit from a larger user base, but obviously they understand that and are taking it slowly.

I'm also open to an invite if anyone is feeling charitable.

You and GP can send me an email for an invite (see profile).

Thank you so much! I sent you an email.

As an aside, do you get a bunch of spam from having your email semi-public?

My email has been public for ~10 years and running on Google Workspace for just as long. I get roughly 2 spam emails / day. Maybe 1-2 false negatives and false positives per month (both these rates have been getting worse with gmail).

Oh wow. I get the same amount of spam with a completely private email.

I just glanced at it, but Tildes feels like reddit did 14 years ago.

Tildes is amazing!

I tried the HoloLens ages ago and honestly the only upgrade would be the display & how often the inputs worked. Microsoft was demoing the HoloLens @ some hackathon and had a program where you could flip pages in a virtual book. On top of the book looking super low quality / illegible, I could never get the controls to work. The only thing that will make the vision pro seem like a worthwhile purchase would be if the controls worked flawlessly. I already have faith in the displays & picture.

As much as I like private trackers, very few use the ratio-less model to protect against serial leechers.

Typically on a tracker you’re given a currency (although not as sound as some e-coins) and can use that to influence your upload or download statistics, which in turn affect your ratio. Some trackers might employ rules where your user class has to have a certain ratio, or else you’ll lose privileges like certain forums or even the ability to download at all. (The trackers are private and can control which peers you can see)

[spoilers ahead]

Thankfully, this is not like the "No Russian" mission in the original Modern Warfare 2. Instead, you're tasked with intercepting a known terrorist, and have to protect yourself against the cartel. You will be punished for harming anyone that isn't part of the cartel.

Thank you, that's a relief indeed :)