Hacker Newsnew | past | comments | ask | show | jobs | submit | maples37's commentslogin

Do you have any tips and tricks to share? I'm running a self-hosted instance on an old desktop PC in my basement for me and a couple family members. Performance is kinda meh, and I don't think it's due to resource constraints on the server itself. This is after following the performance recommendations in the admin console to tweak php.ini settings.


Thie was a few years ago so I can't say I exsctly remember, but thinking about PHP performance is certainly one of the good routes to think about.


In my case, I want file/photo syncing, calendar syncing, and contact syncing.

Nextcloud provides all 3 in a package that pretty much just works, in my experience (despite being kinda slow).

The Notes app is a pretty nice wrapper around a specific folder full of markdown files, I mostly use it on my phone, and on my desktop I just use my favorite editor to poke at the .md files directly.

Oh, and when a friend group wanted a better way to figure out which day to get together, I just installed the Polls app with a few clicks and we use that now.

I am a bit disappointed in the performance, but I've been running this setup for years and it "just works" for me. I understand how it works, I know how to back it up (and, more importantly restore from that backup!)

If there's another open-source, self-hosted project that has WebDAV, CalDAV, and CardDAV all in one package, then I might consider switching, but for now Nextcloud is "good enough" for me.


Ok so it's just the convenience of being a package, thank you for explaining.


If they make their own distro, though, they're not really gaining more control. They're just enabling even more choice for someone who's looking for alternatives.

Let's say, hypothetically, that Valve releases SteamOS to the general public, and it's received generally well, and it becomes much more common for people to use "that Linux thing" than it is today. Then let's say, hypothetically, that Valve turns evil and... I dunno, starts charging money for updates? At that point you've got a large population already using Linux, I'm sure there would be a pretty big migration to Ubuntu or some other mainstream Linux desktop.


I was about to comment that SteamOS is based on Arch, but after looking at the actual graphs, they've got SteamOS as its own separate category.

I wonder how much of that is "hackers and experimentalists", versus random gamers* preferring Arch Linux's bleeding-edge latest-and-greatest packaging approach versus Ubuntu's seemingly-slower-paced development?

* though I suspect even the most casual 25% of PC gamers are probably significantly more tech-savvy than the average PC user of the population in general.


and for me, the one (rather niche, I might add) game that didn't "just work" was working just fine after trying a different Proton version - which is literally as simple as opening the "Properties" page and using a drop-down menu.


You wouldn't be the first to suggest pneumatic delivery of burritos

https://idlewords.com/2007/04/the_alameda_weehawken_burrito_...


Is there any reason to believe that deleting a ChatGPT conversation is anything more than "UPDATE conversations SET status='user_deleted WHERE conversation_id=..."?


Don’t get me wrong, I am highly skeptical. I also am genuinely curious because it seems to be in their best interest to delete these records for a few reasons:

1. Adherence to their own customer-facing policy. 2. Corporate or government customers would CERTAINLY want their data handling requirements to be respected. 3. In my experience at $MEGA_CORP, we absolutely delete customer data or never maintain logs at all for ML inference products. 4. They’re a corporation with explicit goal of making money. They’re not interested in assisting LE beyond minimum legal requirements.

But still I wonder what the reality is at OpenAI.


> They’re a corporation with explicit goal of making money. They’re not interested in assisting LE beyond minimum legal requirements.

Having a good relation with LE and the state is beneficial to companies, it puts them in a good position to sell them services, get preferential treatment, quid pro quo, etc.

Look at how Google, Microsoft, Amazon etc cozied up to LE and the government. They get billion dollar contracts, partnerships with LE and the more they cooperate willingly, the less likely they'll be sued into cooperation or punished with loss of contracts.


Apparently they have a "unexplained crashes must have an explanation determined" policy ever since there was a trend of uninvestigated unexplained crashes that were canaries in the mine for a security issue.

https://blog.cloudflare.com/however-improbable-the-story-of-...

> But [the Cloudbleed sensitive information disclosure security incident] wasn’t the only consequence of the bug. Sometimes it could lead to an invalid memory read, causing the NGINX process to crash, and we had metrics showing these crashes in the weeks leading up to the discovery of Cloudbleed. So one of the measures we took to prevent such a problem happening again was to require that every crash be investigated in detail.

Since then, they have a "no crashes go uninvestigated" policy, which for the scale Cloudflare operates at, seems pretty impressive.


Yes, and we set up all the tooling for that and I would look at the output every single day and keep an eye on what was happening. Any team that didn't fix a crash quickly got a personal message from me. That responsibility has been taken over by others now.


Some alternators ironically require electricity to make electricity. They don't have permanent magnets inside, but instead use electromagnets. So from a stone cold battery, if there's not enough power to get those electromagnets functional, you don't have a way of converting that rotational energy into electricity.

https://en.wikipedia.org/wiki/Alternator#By_excitation

I do wonder how much current that requires, though. In a pinch, could a duct-taped string of AAs be enough to get you going?


As of 2013, manual cars (at least Mazdas) can still be roll-started, as long as the engine computer has enough power to function.

My CX-5 even has a wireless-pushbutton start, not a physical-key-in-the-ignition start, but I've still been able to roll-start it when the battery is too dead to crank the starter motor but still has enough juice for the electronics (lowest I've seen is ~8v if I recall correctly, but don't quote me on that).

The process is pretty much the same: put the car's ignition into the "ON" position (in my case, press the pushbutton twice without touching the pedals -- once to ACC mode, then once to move from ACC to ON), then it's the same as normal: clutch-in, shift to your preferred gear, get rolling, and pop the clutch. Engine computer sees "oh, looks like the engine's spinning, let's add gas and spark" and you're good to go.

Anecdotally, I've seen the described behavior of the engine computer ("detects spinning and adds gas/spark, even if the initial motion wasn't from the starter motor") on automatic transmission vehicles, too. On a 2008 Chevrolet, I found that if you revved the engine up a bit (for inertia), turned the key to OFF, then quickly turned the key back to ON (without turning all the way to START), the engine computer will catch it and keep it running.


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

Search: