olex

Some home routers also offer built-in VPN using Wireguard. Works amazingly well with my Fritzbox, near zero setup on the router itself and very simple config on the end devices.

When I built my DIY NAS, the most important factor was: I wanted it _silent_. No constantly spinning drives or fans. It sits in a sideboard in my living room, and I've been using a M1 Mac for years now that is completely silent in normal daily operation, so I've become a bit sensitive to any "computer noises" and wanted my continuously running hardware to be as quiet as possible.

In practice this meant: a passively cooled Intel N100 SoC, a Corsair PSU that shuts down its fan under a certain power threshold (iirc 35W-ish), and SSD-only main storage. I did include a system fan (low-RPM 120mm Noctua) that is actively controlled based on various system temps (stays off 99% of the time), and two HDDs that sit in standby spindown and only spin up for snapshot backups once every three days deep at night.

Very happy with this system so far. It houses my data dump, backups for all my systems (replicated as snapshots to the HDDs), hosts HomeAssistant/Z2M, and hosts a local-only Gitea that keeps up-to-date clones of all my Github and Gitlab repos.

Anything I host that's available on the public Internet, I don't do from home - that's all on various VPS' or AWS. To access my local stuff remotely, I can always VPN in to my home network.

Which one do you have? I had the same issues with my '19 M3P, all the brakes were badly corroded a year into the lease. Apparently the software to keep this under control came a bit later into the lifecycle, and is not very effective if the problem is already apparent. Anecdotally, what was _very_ effective at cleaning up the brakes was blasting it for a couple laps around the Nürburgring Nordschleife - the grooves / corrosion went away almost completely :D

Now I have a RWD Model 3 with an LFP pack (SR+/"base model"), and have to use the brakes much more than with the old car (that had an NMC pack and almost never limited regen). Over two years in, brakes seem good as new, hopefully they stay this way.

Some EVs (Teslas are ones I know of) have software that occasionally lightly applies the brakes to keep them clean.

Also, on EVs with LFP batteries the brakes do get used a bit more, because a full and/or cold pack usually can only take very limited power from regenerative braking. Depending on the implementation in the vehicle, brakes are either applied automatically to still allow for one-pedal-driving (aka. "brake blending", making the difference unnoticeable to the driver), or throttle-pedal regen is simply capped and the driver has to use the brake pedal when they notice that regen power is not sufficient.

> Once Copilot is done, it’ll tag you for review. You can ask Copilot to make changes by leaving comments in the pull request.

To me, this reads like it'll be a good junior and open up a PR with its changes, letting you (the issue author) review and merge. Of course, you can just hit "merge" without looking at the changes, but then it's kinda on you when unreviewed stuff ends up in main.

Dragon likely wouldn't be able to get it, unfortunately.

The lander would easily fit into the unpressurized cargo bay (the "trunk"), that is typically used to launch various vacuum-bound payloads alongside pressurized cargo inside the capsule. However, for a return from orbit, the trunk cannot be used - it is not protected by a heat shield, and is ejected before re-entry.

You are correct that the return payload mass of a Dragon would technically allow it. But you'd need to somehow get the captured object _inside_ the capsule, which may be possible via the EVA front hatch for something smaller, but not 1m in diameter like the Venera lander.

Starship should be able to do it, since it is fully protected by heat shielding and returns in one piece, not ejecting any modules in orbit. But that's quite a while from being operational yet.

I've only enabled tab groups on my profile after updating to 138 today. That first attempt where drag-and-drop didn't work was already on 138. But I can't reproduce it anymore, so I assume that was a fluke. Looks good so far!

It's available for everybody since FF 137, just randomly not enabled.

Go to about:config and then set browser.tabs.groups.enabled to true.

With the horizontal tab layout, you can drag-and-drop the group name to move the whole group, same as singular tab. Although the first time I tried it it didn't work, now that I keep trying with different click timing etc it seems reliable - not sure if there's maybe a drag handling bug hiding in there somewhere.

In the stream commentary for some of competitive Geoguessr I've watched, they definitely often mention the color and shape of the car (visible edges, shadow, reflections), so I assume pro players know which cars were used where very well.

I agree. The author isn't saying that the best devs never complain about something. He's saying they never leave it at complaining and throw their hands up, but dig in until the find the underlying reason for the behavior, or a way to work around it, as long as the problem remains in their way.

It was a live stream, presumably stream operators had instructions to not show the kaboom live on camera, should one occur.

"Completely successful" would've been reaching orbit. They didn't expect to get that far, but this was still a far cry from a complete success. Only milestones cleared are countdown, launch, and stable flight in fixed attitude - which still leaves a lot of big ones on the way to orbit (controlled pitchover, max-Q, stage sep, second stage ignition and operation, payload fairing separation, orbital insertion, payload separation at the very least).

Surveyor and other probes have landed autonomously in the years leading up to Apollo, and after it as well. We definitely do have the technology, but having not used it for a couple decades, we've gone a bit rusty with it.

According to IM's website, it landed within 250 _meters_ of the intended site - could there be a unit confusion (m / mi) somewhere in the press chain?

https://www.intuitivemachines.com/im-2

Self-hosted FreshRSS + NetNewsWire on Mac and iOS/iPadOS is my combo of choice, works great. After Reader died I migrated to Feedly for a while, but self-hosting the backend and not having ads injected + being able to debug and properly manage feeds is way nicer.

While the telemetry was still going, you could see Ship engines going out one by one. Earlier when there was video there was what looked like flames visible inside one of the flap hinges, definitely shouldn't be there on ascent. Presumably something failed internally and caused the Ship to shut down before reaching target trajectory, at which point either FTS or the failure itself caused it to blow up, as seen on the Insta reel.

Separation is much closer to the launch pad in Texas, the booster barely makes it downrange at all before turning around. This being filmed from the Bahamas with this much lateral velocity, gotta be the Ship breaking up. Likely the FTS triggered after enough engines failed that it couldn't make orbit / planned trajectory.

Inadvertently perfect timing for this footage. Glowing and backlit by the setting sun, against clear and already darkening evening sky... couldn't plan the shot any better if you tried.

Let's hope no debris came down on anyone or anything apart from open water.

First attempt maybe, but they have streamed New Shepard launches many times before, so the company is not exactly a newbie at this overall. Of course, an orbital launch provides a very different level of challenges compared to a vertical hop to 100km.

We're really spoiled by SpaceX, RocketLab and other live streams. In comparison this was pretty unspectacular - they did have camera(s?) on the booster and on the second stage, but only showed them for a few seconds during the stream, and the picture was breaking up constantly. The stream itself, video and timeline animation, was multiple seconds delayed behind the audio callouts in the background. The animated position of the first stage was clearly wrong, it was already showing almost on the ground when the re-entry burn was called out. Lots of things to work on for future launches, if they want these streams to be as watchable as people are used to.

Although that's all details, not to detract from the feat of reaching orbit on their first attempt. That's not something that happened very often and worth significant praise.

Since he explicitly mentioned a hydraulic clutch and needed to provide a separate fluid reservoir for it, it's a very safe bet to say the car is a manual.

I suppose they didn't bother with that, and the system basically only works for cruising in 5th gear, which is what the car was likely doing for the absolute majority of the 6000km drive. You'd lose out on having "self-driving" in slower traffic where gear shifting would be necessary, but it's still hugely useful.

Watched this mini video essay by BPS.space (Joe Barnard), and it resonated with me a lot. So many personal projects get side tracked by trying to optimize some detail or another, get bogged down by that detail, and never get finished in any form. I really like his approach to explaining the phenomenon and how it affects a ton of people in the maker / hacker / engineer space.

We heavily use it for end-to-end testing. The API is integrated into our Playwright tooling, so we run e2e tests that verify entire user flows including email "steps" - such as the new user registration flow ("Open landing page - click through to registration - fill out and submit form - find confirmation email in Mailpit - click link - log in with new user credentials").

What about Mailpit did you find complicated to setup? For us, the Docker image works out of the box, and about all the configuration needed is pointing the SMTP client that does the mail sending to localhost instead of the normal outgoing mail servers.

Mailpit works amazingly for us. Thanks to it having a very simple API, we've been able to integrate it into Playwright e2e tests, and can easily verify things like complete new user registration and password reset flows in seconds. And the UI is handy for local dev work.

There is still a dedicated wiper button on the steering wheel.

>German public transport fascinates with its ease of use and quality

Ease of use maybe, although my parents and grandparents would like to argue differently. They are not as quick to work their smartphone, and the ticket machines are being removed everywhere to be replaced by apps that are much cheaper to run. This works fine for the younger generations, but older and less tech-savvy people are getting left behind.

Quality though, no way. Every single time I tried to give ÖPNV a chance in the last 3-4 years I was either different degrees of late or didn't arrive at all without switching to some alternative method of transport on the way. Doesn't even matter if I tried local routes (Frankfurt and Darmstadt) or longer inter-city connections to Munich or Leipzig, it's all completely broken. People in my company routinely book connections several hours earlier than they need to be places to have a chance of arriving in time, and often are still late. Trains are overbooked, connections are late or often cancelled altogether, seat reservarions don't work more often than they do, WiFi on the trains never works... Many, many things have to change for me to reconsider my default of taking the car everywhere, and I don't think they will in any sort of a relevant timeline.

Do you have a source for that number? The whole rocket weighs 550 metric tons fully fueled on the pad, I am having a hard time coming up with a way for that to create 50x as much greenhouse gas.

According to Everyday Astronaut's article (https://everydayastronaut.com/rocket-pollution/), a Falcon 9 emits a total of about 600 tons of various gases during a launch. The difference vs the rocket's actual mass seems to be accounted for by water vapor, which comes from ground support equipment and not the rocket itself.