The PeerTube instance that I have been uploading to is no more. I think there are around 70 edits of mine that are no longer accessible. If there was a catbox link, they may still work. This represents hundreds of hours of work that I did that have simply vanished. If a link is broken, let me know and I can see if I still have the files for it. Thank you all for watching the things I’ve made/
This is why Lemmy is the only decentralized service I use. Can’t imagine putting your faith
of the heartin random users to store your data. If my instance shuts down it will suck but the only real value lost is discussions and funny posts.Everyone remember, nothing uploaded to the internet is backed up. You can’t count on the “nothing on the internet is ever deleted” axiom to save you. Because SO MUCH of the internet has already been lost to history.
Ideally, you want your personal data in three locations. One copy on the computer or device you use where it can be freely accessed and edited. One copy backed up locally (external drive or physical media) that is archived but can be gotten to at a moment’s notice. And one copy off-site (a copy on another computer somewhere else or a cloud backup service) that can be requested or sent back to you should something destroy the first two backups, like a fire.
It’s also important to remember that syncing services like DropBox or iCloud are NOT backups. If a file is deleted or damaged in one location, that change will sync to all the other locations and the previous version of the file is very difficult to recover, even by the engineers running those services.
I’m not trying to make anyone feel bad here, this is as good a time as any to repeat the importance of getting into the habit of backing up irreplaceable data, or setting up some kind of automated service. There are a lot of options and they’re usually pretty cheap.
(I pay $9 USD a month to BackBlaze to backup and keep 10TB of data and changes going back one year. But even keeping a spindle of DVDs in your closet is better than nothing.)
Sorry to hear about this, Jawa, I’ve been there multiple times. Hope you can recover most (if not all) of the stuff you lost.
Assume anything you want to have saved on the Internet is already gone. Assume anything you never want published on the Internet will stay there forever.
I do follow these guidelines for stuff that is legitimately important. These were all silly edits that I made in hopes of making people smile, and maybe laugh. Nothing of true importance was lost, aside from making someone marginally happier if only for a brief moment. I can recover a good bit, but I will have to revenger them all which will take time.
I disagree that they are not important. The OC you make is like a scrapbook of that point in your life. And like you said, it makes people smile. I think you should backup your memes.
So if I save on my laptop, back up to a separate hard drive, and then share to Dropbox, if I then suffer a house fire that takes out 1 and 2 do I also lose my Dropbox files?
Synchronized files are not backup.
While synchronized files may save you if you get lucky, proper backups are things that work without that requirement of “luck”.
It also doesn’t mean synchronized files are useless. It only means they are not backup.
I get that, just these files get updated regularly so synchronicity is important than independent files. I get for a will I’d want independent files since that is (hopefully) rarely changed except maybe once a year.
I think your question is related to this part:
It’s also important to remember that syncing services like DropBox or iCloud are NOT backups. If a file is deleted or damaged in one location, that change will sync to all the other locations and the previous version of the file is very difficult to recover, even by the engineers running those services.
In your hypothetical scenario: no, your Dropbox files would not be lost. After you lose your laptop and HDD, if you were to buy a new laptop, you should be able to download your files from Dropbox just fine.
What that comment is talking about is how when your local files are synchronized with Dropbox, any changes locally get synchronized out to your Dropbox version. If you delete the file locally, a request gets sent out to the Dropbox servers to delete it there. If your file gets corrupted locally, that newer corrupted version gets sent off to the Dropbox servers to overwrite the intact version.
But if your laptop burns up, it doesn’t send off a request to delete or corrupt the file on the Dropbox servers or anything. You just lose your local copy.
Thank you for answering my question. I was half awake when I typed it and if I had thought a little longer I would have figured that out eventually. :)
You can perform one-way syncs with Dropbox and I have been doing so for the past 4 years. I have a nas server running in raid, which backs up locally, but also sends a one-way sync to two separate cloud backup solutions. One of which being Dropbox. I produce music and have made mistakes with preserving my data. Months worth of work disappeared. Never again.
If you still have them you should consider uploading them to archive.org, or similar services, so you can more easily recover them in similar cases
It’s sad, but currently inevitable, for fediverse servers to disappear with everything they hold
I will do that. What irks me is that I tried to go all in on Fediverse, and it was good for a long time. I’m just sad that this has set me back a good bit in my attempt to make people smile in a shitty world.
It’s sad reality of small projects. Compared to Google pulling the plug on a new service, this is easier to swallow. Someone’s hobby project (the server) go untenable and they had to close.
I know. It was, however, the firs PeerTube instance that I found with instant sign ups and no wait times for them.
Ugh, that sucks.
For months now, I’ve been thinking of setting up a PeerTube instance alongside my Lemmy one but haven’t gotten around to it. If I ever do, you’re welcome there and I’ll guarantee you at least a week to export your stuff if I ever shut it down.
I appreciate that! Admittedly, some stuff is on YouTube, but I hate that it’s there. If you do setup a PeerTube instance, I will make an effort to be the most dedicated uploader.
I hope you learned an important life lesson from this about always backing up your stuff.
Right. Most people in that situation will learn the following, very important lesson:
Put your stuff on Google/Meta services.
All the tut-tutting in the world won’t change more minds than a corporate service with 99.9999 uptime and a small country’s power budget in data centers. If you’re trying to OSS or crowdsource services on the Internet your one and only priority should be UX.
You can buy 2 TB hard disk for 50euro or less. Why people want to put their work on the cloud beats me.
Because the best way to ensure you always have a copy is to have multiple copies in multiple mediums.
A live copy you can access on your computer stored on a local drive.
A “cold storage” copy that lives on some kind of removable media, usually USB thumbsticks but can also be full but disconnected drives.
And finally a copy in “the cloud.”
It’s about distributing the failure points so if any one fails the others are all still available.
Like, for instance, if your house burns down you still have your cloud copy if the ither two got burned. (I personally would put cold storage copies in a fireproof safe.)
On the other hand, if your cloud copy gets deleted, and your local copies are fine… All you have go do is create a new cloud copy at a different cloud site and you’re back in business.
Well, yeah, but that’s a lot of work and Google is already doing it for you.
Which is why most people just set up a GDrive and move on with their day.
You want to compete with them, it needs to work the same way, just as reliably.
It’s not even a lot of work, it’s standard IT backup.
How do you think Google does it? With lots of redundancies and backups.
They’ve got millions of failover servers and millions of backups, but even they’re not perfect and data gets lost. 99% uptime is great, but lots of people still lose data or get locked out of their accounts. That’s not a new thing with Google. They absolutely do their best to retain customer data, but even they are not perfect.
You call it “a lot of work” but it’s literally the bare minimum you can do to ensure you actually retain copies of your data. Just basic backup redundancy. Those kind of redundancies that allow Google to pull a 99% uptime, but that doesn’t mean all their redundancies are enough when it comes to your data.
Lots of people who placed all their faith in Google have either lost complete access to their accounts or have had significant amounts of data lost. It’s not a guarantee that your data is safe with Google forever and ever amen.
Yes, it’s a lot of work because the alternative is no work. Of course Google does the same thing, they just do it in the back, where you’re not looking, financed by all the data and money they syphon from you.
I’m not saying competing for free is easy, I’m saying that’s where the bar is.
Hey, I do keep multiple copies in multiple locations. But I’m a nerd. Just one self-aware enough to understand how everybody else is doing it and what the usability bar is.
It doesn’t matter why. What matters is if you build an alternative to a server-based alternative you need to match their functionality and UX.
For the record, proper storage at scale costs a whole lot more than 50 euros. I think self-hosting is set to become way more popular in the future, but it gets harder if data keeps evaporating.
![[Meta] A lot of my OC edits are gone.](https://media.tenor.com/LqVVm2ubyXwAAAAM/spock-computer.gif){kind=link}