lukevp

Same as well! Game is fun once you get it

This is mostly an engineering problem I think. The new Dewalt DCBL777Y1 leaf blowers are so quiet that even on loud I can use it in my garage without feeling like it’s loud or my watch triggering a sound alert. With my gas leaf blower, even outside it’s deafening and my watch immediately triggers. And the electric one is 2x more powerful than the gas one (in CFM, no bs about a smaller nozzle and “high mph”, the thing BLOWS). It also works upside down and doesn’t dump exhaust so I can use it inside as well as upside down when drying my car. No shill, I paid full price for mine and have no affiliation with Dewalt, just surprised and happy. I expected to return it because I didn’t believe it would be better than my gas blower. That gas blower has been on a shelf for the last 3 months. I’ll probably just sell it, there’s no use case where it’s better than the Dewalt except for long runtimes, which I don’t need.

Can you plant bushes or shrubs between you and the road? Also if you add a water feature between you and the road it would add more constant background noise which would raise the baseline that noises would have to exceed before you could hear them.

Same. The suggestion that it’s like audiophiles totally missed the mark, because lots of audiophile claims do not stand up to double blind tests. I can guarantee that 60 vs 120 hz blind tests would be insanely easy to pass if there was window movement or scrolling or basically anything but static frames.

How is that comparable? On-package RAM is lower latency and higher bandwidth and also much more expensive than external DDR5 sticks.

My Mac only sleeps on closing if it’s not plugged in and connected to external monitors. That’s how I would want it to work. How are you wanting it to work? Closing it keeps it on no matter whether it’s plugged in or not?

Not OP, but RTO forces geographic centralization and reduces mobility of their employees. If you can work remotely then you have a much larger pool. And I think that opening the door to remote work made employees realize that there was some power and some negotiation to be had on working conditions (basically our generation’s version of the 40 hr work week in response to the Industrial Revolution)

That’s amazing! I’d love to see it. Looking at a map, the road doesn’t look too long so we could hunt for it, but if you see this could you describe how to find the specific tree?

Part of the component model is codegen to build object structures in each language so that you can pass by reference objects that have an agreed upon shape.

Yes they each have their own linear memory, that’s one of the advantages of component model. It provides isolation at the library level and you don’t have to implicitly agree that each library gets the level of access your application does. It provides security against supply chain side attacks.

Having said that, component model isn’t supported by all runtimes and since its binding and code gen are static at compile time, it’s not useful for every situation. Think of it like a C FFI more than a web API receiving JSON, for example. Upgrading the library version would mean upgrading your bindings and rebuilding your app binary too, the two must move in lock-step.

I’d be the first to speak out against climate change, but I’m not sure I understand how co2 is relevant? The current ambient co2 in the world ranges around 400-500 ppm. It can easily get into the thousands indoors and that doesn’t impact birds. It seems like we’re a long way off from ambient co2 causing an issue with wild birds?

Here in Oregon we have a lot of them, but it never gets old seeing them! We usually see at least a couple on our way to the coast from Portland. Earlier this week, my wife and I saw a mating pair fly high over this field by our house and spiral down towards the ground, holding talons. It was amazing, they are huge incredible birds. I’m so glad to live somewhere that has lots of public land and habitat preservation so my family can enjoy the clean air, water, skies, and trees, as can all of the wildlife we have.

I’m working on a cross-platform fast multithreaded HTTP / FTP downloader that will download much more quickly than other clients like FileZilla, hash check files, perform follow up operations (like extracting RARs or deleting files on the remote,) and have a nice graphical UI that runs in the browser and allows local/remote/cloud control. It’s early (started last week) so there’s not much done yet, but if you’re interested, would love a star or watch: https://github.com/lukevp/Speedful

Why can’t the stolen coins be sold? If they can’t then what’s the point in stealing them?

As I understand it, the software surrounding and managing the wallets is always what’s compromised, right? So it’s really the exchanges are the only place that the money is being lost, because the protocol itself is sound.

I’m glad you’ve gotten relief, and really sorry you tried all those other things first. Did you happen to try a latex mattress? They seem “foamy” but they are much more like spring mattresses than they are like memory foam. Basically the more compression it is under, the more resistance there is, exponentially, whereas memory foam compresses linearly. My wife and I switched to Latex and it made a massive difference in our sleep quality and how we feel afterward. The memory foam mattress made us hot and made my back hurt really bad. Now I wake up and I feel refreshed, and I sleep an hour less but get the same quality of deep sleep as before. You may not want to tamper with anything since you’ve found a solution, but I wanted to mention it because it does give more pressure relief than an inner spring but with more support than memory foam. They’re expensive and heavy though.

Or you buy an Apple TV, that’s priced appropriately with its capabilities and doesn’t thieve everything from your network and your house.

Inference isn’t that expensive. A single junior dev costs orders of magnitude more than the amount of inference I use. Companies in growth mode don’t have to make money, it’s a land grab right now. But the expense is largely in the R and D. You can build a rig to run full models for 10-20k right? That’s only a month or two of a junior dev’s time, and after that it’s just electricity. And you could have dozens of devs using the same rig as long as they could timeshare. I don’t see where the economics wouldn’t work, it’s just there’s no use in investing in the hardware until we know where AI is going.

I’m a little shocked at how much negativity there is around LLMs among developers. It’s a new tool that requires some learning, and it’s sometimes not so great, but if you’ve used an IDE with real coding assistance built in (eg. VS Code in Edit with Copilot mode - NOT Chat mode, using Claude 3.5), it’s honestly not much worse than a junior dev and 100x faster. And if the code is bad you throw it away and try again 10 seconds later. The amount of speed up I see as a very experienced dev is astronomical. And just like 6 months ago it was awful. How great is it gonna be in a year or two? It doesn’t even have access to running unit tests or reading console errors or IDE hints, and it still generates mostly correct code. Once it gets more deeply embedded it’s just going to improve more and more.

Why in the world would you want a car?? They are horrible to maintain, difficult to operate, expensive, smell bad, and are slower than horses! -somebody salty about automobiles, circa 1900.

The law of thermodynamics you’re referring to is in a closed system. In this case, energy is being introduced by the burning gasoline pushing the pistons, and as a byproduct of that, exhaust gases are vented at a certain speed and volume. That is wasted energy as it just goes out the tailpipe more forcefully which adds minimal forward momentum to the car.

The reason turbos work is not because they are forcing more air, it’s because they are forcing more air which ALSO raises the compression ratio and the amount of fuel detonating (the MAF will sense the increased air density and add additional fuel to compensate). I.E. a turbo burns more fuel and therefore it produces more power.

Hot air is less dense than cold air (why heat rises). Probably keeping the hot side of the turbo hot helps with gas expansion, perhaps that increases the force on the turbine? I’m not sure about that. But the intercooler is to cool the air on the compression side because cold air is denser. Which means more air into the cylinder, which means more fuel. Probably the colder air helps with auto detonation too which lets you push the compression ratio further if the pistons and cylinders can handle it.

Google’s entire business is your personal data. I’d much rather my browsing info be sent to some small website for local analysis for their own purposes vs. hoovered up by the largest data broker in the world and aggregated with everything else about me.

Did you read the recent article about how reCAPTCHA is used for tracking you across the internet? I have no reason to believe GA isn’t similar. And just think about how much info they have on you from reading all of your emails.

I’m talking about the UI of the media browser part, not the video playback. I’m direct playing and the video is perfect.

I think it's great that it's free and open source so it's available to as many people as possible.

I don't really care about the free part though. If I could pay for Jellyfin and it would be better, I would happily do it. I pay for Infuse so I have something usable on Apple TV even though Jellyfin itself is free. I don't care if other people don't pay for it.

I would rather pool various other people's money who have the same perspective as me to allow one of the Jellyfin contributors to prioritize and perform changes in the project that I and others value, which raises the boat for all. There's no world in which I am going to learn how to contribute to Jellyfin to fix something fundamental about the Jellyfin architecture, for example. It would take years to even get buy-in from the core contributors to have that access. Maybe I could spend a few months and clean up minor UI things or something, but what's the good of that?

I see this take a lot, where a project being free / OSS means that no one should have an opinion on it or that if they don't like something, they should fix it themselves. That's literally what money is for, so I can transfer wealth for some benefit to someone who is better at something than I am, so I can spend my time doing things I'm good at. Nothing about that prevents Jellyfin from remaining FOSS but at the same time, making improvements and receiving and prioritizing community feedback.

Having placeholders is part of what I’m talking about when I say lag, and jank is about dropping frames so that things do not appear smoothly animated. Not everyone is as sensitive to jank or they’re ok with it, but if you use something that lacks jank in the animations it feels better even if you don’t know what jank is.

The media detection thing might be related to Infuse and not to Jellyfin itself, since it sounds like everyone else isn’t having this issue, which is good to know.

Thanks for the detailed reply, I’ll look through my configuration and see if I can find where the issues are with the file watching. You’re right, many of them are likely because I have to use Infuse, due to the native Apple TV client being so bad.

When I say there is UI jank and lag, what I mean is that content re-renders and shifts around when you search, results flash in and out with loading animations. the placeholder tiles are clearly visible and I can see the tiles loading in when I scroll. This happens in both the web client and Apple TV. I believe the issue is the API design that’s dictating how clients fetch and load data, some virtualization in the frontend implementation, and a lack of prefetching/caching.

I have a 10 gig network to a very fast NAS setup and a very small media library. the image assets should be able to be cached/streamed/prefetched from the server so that I do not ever see a placeholder tile (maybe if you jump ahead like 1/2 the library, then that makes sense).

Perhaps this is one of those things where people haven’t seen what’s really possible performance-wise from a local server and they’re OK with something that feels like a webpage. But nothing feels like a native app that just has all my content there all the time. It feels like a remote service even though it’s < 1 ms away on a hardwired multigig network. Does that make sense? Do you agree with that?

Jellyfin is pretty good, and I've been using it in place of Plex for the last few years. However, I feel like there are some underlying limitations that I just can't get past:

1. the UI jank. The thumbnail tiles are slow to load, even on a local network. Searches and filters flicker as you type and take a while to return. Scrolling fast in the web UI gets choppy/laggy.

2. The native app (at least in the case of Apple TV) is either nonexistent or terrible. I've been using Swiftfin since it was one of the first alpha versions, and it constantly lost pairing with my Jellyfin instance. When it did work, which was very cryptic and usually required re-enrolling the client every time, it would randomly fail to load things, and the UI was very choppy as well. I haven't used the native apps on other platforms, but I imagine they are equally or more janky, because the Apple TV is comparatively very beefy hardware-wise vs. most other platforms.

3. The polling for new media is slow. I upped it to 10 minutes (the quickest possible setting) but I shudder to think what a full scan of a media library every 10 minutes is doing to my disks. Why doesn't it use file watchers and webhooks for new content notification?

4. The homepage has very little actionable info and doesn't work for browsing. It's not like Netflix or any of the other services where you can boot it up and see a bunch of different categories, as well as your "list". It has playlists, but you have to drill down to see them. You can go to "Movies -> Suggestions" and it has a little bit, but nothing like Netflix does. No real recommendation engine.

5. You have to maintain your own trailers or use an app like Infuse that can download its own trailers.

6. You have to separately configure tiles to be rendered if you want a nice seeking experience where it shows a live preview as you scrub through the timeline.

7. Movies and TV Shows are separated even though pretty much every other platform doesn't separate them, which requires you to click into one of 2 options before you can do almost anything.

That said, it's still far better and less janky than Plex was before I switched, and Infuse actually plays back HDR / Dolby Vision content correctly.

Does anyone else have qualms with Jellyfin? And how does Plex compare to any of these gripes?

This is a really simplistic view. It’s not just policy. In Texas, the vast majority of land is privately owned (only 2% is public land), so zoning is the only thing preventing buildout. Most of the geography is flat, there are huge expanses of empty fields all around Austin, and the weather is buildable nearly year round. There is also a tremendous amount of cheap labor due to the proximity to Mexico and the lack of minimum wages / border policy enforcement / etc. there are so many factors that make Texas cheap to build in that aren’t true universally in America.

Another key point, look up net migration in Austin. During the pandemic and remote work, everyone and their brother wanted to move to Austin, it was super trendy and a lot of tech was moving there. Look it up now. In Q4 2023 not only did grow to slow but it reversed to net outflow. Thats a massive difference in demand for housing and a massive difference in supply of housing. [1]

Yes, more homes got built, but there were many other factors at play than something simple like “let them build houses”.

1. https://www.redfin.com/news/housing-migration-trends-q3-2023...

Are you familiar with how disk brakes on a car operate? They’re essentially a hydraulic piston clamping the brake pads against the rotor. When you let off the brake pedal, the pressure is removed, but there’s not a “retraction” of any kind. So there’s no pressure but nothing is actually making that gap. That article was articulating basically that the piston, when not compressing the brake pads, should not be too far out, because that’s just unnecessary travel of the pedal, but it does not mean that the brake pads are somehow gapped away from the rotors.

Drum brakes by contrast have springs that retract the pads from the braking surface, so the OP has a very valid point.

Yeah but there’s 2 people in there, and 180 people in the rest of the plane. Isn’t the volume of air relative to the amount of people breathing it relevant as well?

I think we’re a couple years out max from Peak Developer and a lot of the growth / entry level roles will start getting taken over by the experienced devs who can adapt to coordinating a set of AI agents to do the actual coding and testing and DevOps more efficiently, and it’s all downhill from there. So I don’t see an entry level, mess around directly writing code job, to exist in abundance in say 10 years.