Hacker Newsnew | past | comments | ask | show | jobs | submit | sedatk's commentslogin

I was there today. We happened to notice the smoke over Kilauea while driving to Hilo, then checked out USGS cams, and immediately drove there and spent the next 7 hours getting mesmerized.

As my first eruption encounter, I didn’t expect to experience several things like the heat even from a long distance, enough to keep me warm in my shorts at 60F, and the loud rumble, like a giant waterfall. The flow of lava was way faster than I expected too, almost like oil.

Mind blown.


Hawaii volcanism is what geologists seem to call "nice and friendly" - low viscosity lava, not prone to explosive eruptions (unlike the stratovolcanoes of the Andes or the Pacific rim in general) - this is because it's caused by hotspot volcanism in Hawaii.

It’s not nice and friendly because of the hotspot volcanism.

It’s because the hotspot has a chemical composition that is generally low in dissolved gasses and very liquid/flows well. (Lots of silicates I think?)

It tends to come out nicely and stay liquid as it leaves, so vents don’t plug up. Also, because of the relatively low dissolved gasses, pressure doesn’t spike as high when it’s flowing out of the chamber like many other magmas do, causing explosions.

Hawaii exists basically because it’s great for building up islands/mountains without blowing them up as part of the process.


Low viscosity is due to the LOW silica content. Tectonic setting is the primary factor determining the magma's final composition (basaltic) and therefore its physical properties (low silica, low viscosity). The gentle nature of Hawaiian volcanism is a direct result of its basaltic magma, which it has because the hotspot is located under thin, basaltic oceanic crust

I don't think hotspots can be said to have different composition. But if the hotspot is under a continent or an ocean plate makes a difference for the type of eruptions. Hawaii is in the middle of an ocean plate, no continental crust there. So we get a basaltic eruption (comparatively lower silica content, low viscosity lava).

Yellowstone is also caused by hotspot volcanism. The friendly eruption is a property of location, not hotspot origin.

Your initial post read the other way, which the parent post is addressing.


Same here. I'd add that viewing the lava fountain at night was a mesmerizing experience. You get to see the full extent of red, glowing lava lake and the fountain. There were thousands of people and yet they appeared so small in front of the volcano. We did experience some ash and Pele's hair on the way to the park, near the black sand beach. I do recommend carrying a torch though at night, since it is pitch black at night. The lava illuminated the park with the red glow, but there were some parts where you do need a torch esp. if you park far away and walk.

> viewing the lava fountain at night was a mesmerizing experience

Is there anything comparable to the overview effect [1] that attends seeing a lava fountain in person?

[1] https://en.wikipedia.org/wiki/Overview_effect


Perhaps a total solar eclipse? Especially so when seeing half the solar system lined up neatly in a row with the sun and moon (and a comet in case of the recent April 2024 eclipse in the US), and truly feeling that one is riding on a chunk of rock hurtling through space, tethered invisibly to that impossibly distant mass of gas and plasma like a ball at the end of a rope...

Still get frisson every time I think about it ;)


Yes we watched it in the night too. It was disappointing to see iPhone “normalizing” its brightness down. I had to take raw photos to capture it.

I was at the Fagradalsfjall eruption at 2023. Had been to Iceland for two weeks w wife and daughter, and on the last day (since the signs were there) I decided to postpone travel home for two days (w + child wanted to go home). On my last day, the hike opened up and I went at approx 2100 hours to the volcano. That was an approx 10 km hike one way.

Amazing experience. A bunch of us were stupid as can be, but got as close as approx 50 m. Sounds really dangerous, but the sputter were not that violent yet, and the ground sloped away from us. Still, really stupidly dangerous (the sputter wall could've broken down, wind direction change, etc). But it didn't. Lots of moss fires, and walked into a small slope and immediately felt a sting in my nose and lungs from trapped gases so took that as a nope and went back.

Started walking back at 0130 something, boarded flight at 0600, fainted (I had done Mt Esja in the morning too). Sorry other passengers, it was inconsiderate of me and I was an asshole for that. But... that experience...!


Why do you think you fainted? Because of the gases?

I was reading about shield volcanoes and it sounds like the low viscosity lava is the defining feature. So the flow is par for the course.

I am severely tempted to hop on a flight to go and see it, but wondering if it's such a "once in a lifetime" thing to go see? That, and if it'll peter out by the time I get there, and $500+ to just fly on a whim and stay overnight.

Yeah it went on about 12 hours which might be a record. But this was the 38th eruption this year. Others may not have been as big as this though.

That said, it feels like a once in a lifetime experience, yes. See my video: https://www.instagram.com/s/aGlnaGxpZ2h0OjE4NDM1Nzg1OTAyMTA0...


> but wondering if it's such a "once in a lifetime" thing to go see?

Depends how close you go to see.

If you wanna see it more than once, don't go too close.


It’s a very trimmed down kernel with minimal set of drivers and modules. It doesn’t even support SD cards out of the box. That’s probably the main reason as no hardware probing/initialization delays are incurred.

Those GUIDs aren't related to NT kernel but Windows Explorer and its COM-based component system. They were introduced with Windows 95, IIRC.

Weird, I’m from the same era, and a predominantly Windows user, but I have zero complaints about Linux gaming on my Steam Deck.

Well, that's apples and oranges. I was playing starcraft on wine earlier this year and there were constant freezes and crashes on new hardware.

Have you tried playing it via Proton or Proton-GE (via Steam or Lutris)?

Vanilla Wine is a bit of a rough experience, I'd only recommend it for advanced users who know what they're doing.


Ah sorry, it was lutris (the name escaped me). I've done all the combinations of wine and wrappers to wine. All of them have a sort of unstable relationship with the calling environment. alt-tabbing out of games never works, there are freezes and game crashes when you try to switch in and out of games a fluidly as you can on windows.

I believe you that steamos is better, but it's also a single task environment where you're not checking your email between rounds of halo.

I'll give proton a try, though i believe that was in play when i was laddering with lutris.


I can relate about OP’s comment about OOP. On a similar note, I had written about re-discovering visitor pattern:

https://ssg.dev/are-interfaces-code-smell-bd19abc266d3/


That blog post is a poor example. It replaced a simple straightforward if-else loop with a hard to understand abstraction spaghetti. A few unit tests and you'd be well on your way to shipping rather than messing around with beautiful patterns.

Obviously, because it was a fictional, simplified example for the post. My real use case was more complicated and involved multiple developers working on different parts of that flow.

The problem with if-else chains is it's easy for a programmer to forget to handle a case that another developer added in the called component. Unit tests can't help a spec miscommunication. But, visitor pattern can as it forces the handling logic to be complete.

Hence my example at the end using discriminated unions and exhaustive pattern matching in F#. Much, much simpler with the same benefits.


As a "half way" point, modern (21+) java brings pattern matching switch statements to the language, but you'd probably construct the F# version in Java using a sealed "marker" interface. Something like

    sealed interface Result permits ValidationError, SearchQuery, UserProfile {}
Along with the specific implementations of those ValidationError, SearchQuery and UserProfile classes and then a switch statement like:

    Result res = db.query(input);
    return switch(res) {
      case ValidationError ve -> context.renderError(ve);
      case SearchQuery sq -> context.redirect("Search", Map.of("q", sq));
      case UserProfile up -> context.redirect("Profile", Map.of("user", up));
    };
The sealed interface gives you compile time checks that your switch statement is exhaustive wherever you use it.

Before that pattern matching, I might have used a Function<Context, R> instead in the Result interface. This is off the top of my head without the benefit of an IDE telling me if I've done a stupid with the generics and type erasure but something like:

    interface Result<R> {
      public R handleWithContext(Context c);
    }

    class ValidationError<RenderedError> {
      public RenderedError handleWithContext(Context c) {
        return c.renderError(this);
      }
    }

    class SearchQuery<Redirect> {
      public Redirect handleWithContext(Context c) {
        return c.redirect("Search", Map.of("q", this);
      }
    }
etc. In either case though I think you're right that an empty interface is something that should be examined closer.

Sealed interface doesn’t look bad for some use cases. I wish C# had it too.

m.youtube.com and m.facebook.com redirect you to main "m-less" domain when on desktop. That was the greatest problem with Wikipedia. You had to experience that mobile layout on desktop unless you edited the address line and reloaded the page.

m.wikipedia.org was a feature, not a bug. The interface is good on desktop. For some time, before Wikipedia did a desktop site rework, this was my go-to frontend.

Late for fixing design and UX bifurcation.

That's a welcome development albeit late, but more importantly, they should address the "can't link to a highlight" problem on mobile. When all sections are collapsed by default, browser won't scroll to the relevant section.

A random "link to highlight" example: https://en.wikipedia.org/wiki/Henry_I_of_Cyprus#:~:text=On%2...

Such a link doesn't work on mobile if it points inside a collapsed section.

That makes directing people to relevant content on mobile really hard, and I end up sending screenshots instead.

EDIT: "Link to fragment"s had the same problem, but apparently, they fixed it. Thanks for that too!


You also can't search for text in collapsed sections.

The link in your comment works perfectly fine for me in Chrome Android, and highlights the part

Ah okay, it's an iOS-specific problem then.

For some reason I've got the android app and it won't show the selection. Does the fragment is part of the deep link anyway?

Oh, I thought GPU passthrough was enabled on vanilla Windows 11, but I didn't delve into that feature enough. It's still extremely impressive of course. Perhaps I should write another article about graphical features.

No, PCI-E passthrough is not enabled on non-Server, and you arguably wouldn't do this on a Windows hypervisor. anyways, you'd do it with a Linux+KVM hypervisor for either Linux or Windows guests.

Using GPU passthrough, however, is allowed. WSL2 does this by using the existing Mesa/DRI/DRM open source stack, but instead of a GPU-specific DRM driver, it is one that speaks WDDM (the DRM equivalent in the Windows driver stack), and only requires a GPU-specific ennoblement package (provided by the vendor, and matches the Windows driver it is talking to; AMD, Nvidia, and Intel all ship one inside of WSL2).


> No, PCI-E passthrough is not enabled on non-Server, and you arguably wouldn't do this on a Windows hypervisor. anyways, you'd do it with a Linux+KVM hypervisor for either Linux or Windows guests.

If it was enabled on Pro, I would use PCI passthrough. I use hyper-v for a Linux dev environment on a windows workstation. My NIC supports virtual functions, so if I could passthrough one to the dev VM, I wouldn't need software bridging and that might be nice. (OTOH, I don't know if my motherboard has reasonable passthrough groups and all the other stuff that makes passthrough never work for me)


And why you not run Linux dev env on bare metal and avoid the pain of using W11 ?

I don't want two computers for work, and I'm not getting paid enough to fight with Linux GUI. I can do all the work without Linux GUI and have a working desktop (still W10 for now). I just need a VM to get a close enough match to prod VM.

I think part of the answer is that if you're going to use both it's nicer to use Windows with Linux as the guest than the reverse. MS clearly put a lot of effort to make the integration nice and it shows. Like how Parallels on macOS makes Windows a very nice guest.

If there was software that made Windows as seamless on Linux I bet it would get a lot of use.



Thanks!

https://learn.microsoft.com/en-us/windows-server/virtualizat...

It relies upon SR-IOV and only several server-specific Nvidia GPUs are listed as supported.

Intel's Flex dGPUs and Arc iGPUs have supported SR-IOV for years now, but they aren't listed there. It would be super awesome if Microsoft could add it for Intel Arc iGPUs, desktop versions of Windows, and WSL2! Intel's GPU SR-IOV already works with KVM on Linux!


Regular Hyper-V VMs can use it too but it's not well documented.

The keyword to search for: "GPU-P"

https://learn.microsoft.com/en-us/windows-server/virtualizat...


Reportedly, current desktop OS marketshare[1] is:

- 66% Windows

- 18% MacOS

- 3% Linux

Right now, I can't try out Orion as one of the 66%.

[1] https://gs.statcounter.com/os-market-share/desktop/worldwide...


A large % of desktop Windows installations are on employer-owned/managed hardware. And the % of users who are interested in trying newer alternate browsers and/or care about privacy also skews heavily away from Windows. So it might be more rational than it first seems.

> A large % of desktop Windows installations are on employer-owned/managed hardware

What's a "large %"?

> And the % of users who are interested in trying newer alternate browsers and/or care about privacy also skews heavily away from Windows

I mean, enterprise users wouldn't try anything new of course, but I don't see the relevance to Windows. Would enterprise Mac users be any more inclined to try new things?


> "What's a large %?"

Perplexity says the share of US desktops managed by employers is estimated in the range of approximately 50-60%...

... and Enterprise Macs in the US represent close to 10% of commercial desktops and laptops.

My point stands.


I mean they say in the article that the have started on the windows version and are aiming for late 2026

I know, the prioritization just looks counter-intuitive to me, especially if you make money primarily from subscribers.

Well, paying money for consumer apps/services (besides games) has always seemed a bit more normalized in the iOS ecosystem, and I imagine it must be a _somewhat_ simpler interface than the lovecruftian monolith of windows-N

saying this as a Linux user, I've never owned an apple device in my life (nor do I want to)


AFAIK, you can't pay for Kagi in Apple ecosystem. Users of both platforms have to visit their web page.

Mac users pay for quality software. Linux users would never pay for anything. Windows users only pay for video games and corporate software if they're a corporation.

So that's why boutique software makers always focus on Mac first.


So? Companies are free to choose their markets. For a long time there were a lot of shops that only focused on the Mac.

Sure, but Kagi isn't a Mac shop; their only source of revenue is subscriptions. Majority of their desktop customers must be Windows users too. So, their prioritization just looks strange to me.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: