JustEnoughDucks

I would be interested to see a figure of people with home servers that have had that happen to them. DoS & pwned yes, especially 15+ years ago before there were good resources, TLS, reverse proxies, or authentication front ends.

I would be very interested to see any stat whatsoever of selfhosters that have gottened murdered specifically because of their server.

It is extremely important to note that in those days, people just opened their, often out-of-date, servers completely to the internet via a DMZ or port forwarding, let ssh be open to the internet, didn’t harden ssh at all, and most people didn’t use a VPN for downloading.

That is literally like saying that people who light wall torches in their wooden home burned their house down, so let’s not use lightbulbs or electricity.

What is the difference between a paid service and a paywalled service in this case?

So you have absolutely no devices that are a different resolution than you download? You don’t direct play 4k on a 1080p screen for example.

Me too, and the new one I didn’t even realize this change happened. I saw there were no breaking changes, updated, and saw “oh, it isn’t synced anymore” so I reselected the folders, it ran a sync check on everything, which took a while, and everything works fine again.

I didn’t even realize there was a difference until now, but I guess there is a start/stop sync switch.

I mean, jellyfin is absolutely even.more of a security nightmare than Plex, with multiple unfixed CVEs IIRC (software, not website or forum)

I use jellyfin also, but I only trust it not exposed to the internet at all. That is one very big area of improvement for them.

That and subtitle syncing.

Except that proton released their LLM AI so maybe they will start doing that?

Or if your internet enters the house in a dead zone.

I have a brick house and our internet comes inside literally in one far corner with the most walls around it, so if the access point in there. Half of our house gets no internet.

I went for a cloud gateway ultra and then one access point centrally in the house where everything can reach.

Immich!

Keeping personal photos off someone else’s server and stopping google and apple from training their AI on your nudes.

I think the issue is more that large tech firms can absolutely deal with external security in their applications. The amount of times gmail or Microsoft 365 has been hacked and leaked a bunch of client data is statistically zero when looking at their attack area.

Joe Dirt self hosting a mail server for his neighbors on a salvaged rack server is 1000x more likely to get hacked or lose a ton of his neighbors’ data than a big tech firm.

That is kind of the trade off for community hosting. There are very very few backup and security-literate people in communities.

Then you need to know 2 different codes.

They would need to hack the server password database or your password app AND have physical access to your device.

It is the same concept as using biometric + TOTP or password.

Something you have, something you know, something you are: those are the 3 general “factors”

What’s the difference between an unpopular opinion and a wrong opinion?

Without MFA, hundreds of thousands more accounts if not millions would be completely compromised. That is just a fact because most people choose horrible and/or completely the same password for everything. Bank account details, credit card info, social security or government ID numbers, etc…

It doesn’t have to be as bad as email or SMS. TOTP has been a standard for a very long time and there are a dozen apps for it. Simply enter the app, copy the code, done. SMS and email are less secure anyways.

American companies seem particularly allergic to TOTP for some reason…

Also with the excalidraw plugin, hand drawing images and such is also possible.

It is not as good for flowcharts and diagrams since there are only like 5 non-specified font sizes, but also usable for notes

Some drives are worse than others and higher capacities get worse and worse, in my experience, Seagate drives are extremely loud.

If you get helium drives (like wd red plus > 8TB i think),or 2nd hand hgst/ WD enterprise drives) they are significantly quieter.

But, having an ssd is cheaper probably. I have an SSD for the boot drive and all databases, configuration folders, etc… In docker so general IO is fast, then media, documents, pictures, etc… On the big HDDs.

KNX.

Everything is decentrally programmed, and you can do extra automations and stuff from home assistant, but KNX devices are wired (generally) and will always Just Work™. More expensive that the cheaper retrofit options, but if you factor in manual overrides or getting the “better” wireless smart devices it is comparable. They generally also have a manual override at the panel. For core functions like lights, HVAC, roll shutters or blinds, etc… That is honestly the best option (unless you want every light to be an RGB light for some reason, then you still need smart bulbs)

This is similar to what I do.

I have a USB drive with the whole bootloader + decryption keyfiles on it. I remove it while it is running as everything is stored in RAM and already booted.

Downside being it has to be plugged in to update the boot partition during an upgrade.

Bitwarden

My company switched over to it to use with sharepoint for our quality system instead of synology because all files need to be tracked and we were already integrated with Microsoft every other way. That was two months ago.

Since then, multiple people have come forward with problems about syncing documents.

I, myself had multiple times already in this short time where I would make changes to a file, save it, one drive would sync and tell me the changes were pushed, colleagues got the previous version while their one drives told them everything was synced, and then I had to open my version again from the Onedrive folder to see that it was the new version, manually save it again, and then manually pause and resume syncing, then FINALLY it would push the changes.

It isn’t common, but when you have hundreds of thousands of files and there is a 0.1% chance that it silently fails syncing some files with absolutely no indication, even in the admin logs, that happens many many times

There is also leantime.io that I have been hosting for 5 years or so. It is a bit more than planka or tarallo as far as scope I think, but it has integrated kanban, gannt charts, and hour logging which is all I need for my personal projects.

Your cloud example is exactly right and exactly what we want to NOT HAPPEN.

They shoved the cloud so much down our throats so that they can force you into monthly income-sucking unneeded subscriptions. That is it. That is the single reason everyone did it.

The result is now the average user has a much worse experience overall. One literally has to fight with Microsoft products to save things on their own computer. IoT and smart products literally won’t function without connections to their “cloud”. Phones come without SD card compatibility and with low flash memory to force you into cloud subscriptions. Now every damn piece of software is a way overpriced subscription that almost all originally started as “switching to cloud infrastructure” (fucking adobe creative cloud).

The “cloud” has had so many data breaches and people data have been stolen, siphoned off, lost due to bugs, and sold to earn even more cash on the side.

A huge portion of the general corporatization and bad enshittification of digital services and software in general can be attributed to “the cloud shoving down our throats” that you describe.

AI is looking to do the same thing except castrate peoples’ digital skills, critical thinking skills, transcription skills, and writing skills in order to siphon more and more of your income off in the form of AI subscriptions while they double dip and sell everything you ever say to it and triple dip in mining everything you say to it as R&D that you pay to do

Companies need to do the fucking R&D themselves with their revenue of a small country and stop forcing regular people to pay to be their alpha and beta testers and focus groups, and people gobble that boot up so hard because LLMs have a few small areas where they are slightly useful and can save 10 minutes per day and make them not have to critically think, so people will literally sell their data, their already small income, and their soul to save 10 minutes, and in 10 years the digital experience will be even more shitty and degraded than it got after “the cloud.”

Your usecase is the exact definition as using LLMs as accessibility and to actually better the user experience for certain people which is not the goal of any AI company or 99% of LLM integrations

TD;DR

Non-consentual cloud shoving has caused newer generations to think that paying corporations every month to save files is normal and that your data is not yours and always corporate property ™®©, along with the decimation of understanding simple file structures. You can actually talk to teachers and professors and they unanimously say that tech literacy has nosedived.

Now with the LLM shoving, they are trying to force the new generation to have to pay subscriptions to think, write, compose, draw, and get information by stripping them of those skills.

Also the Node 804 is worth looking into with an entire separate chamber for HDDs in order to keep them cool without exposing them to GPU and cpu heat, plus it is a lot shorter instead, sometimes easier to fit places (mATX motherboard only)