Why do the instances keep going down? It makes me think that this is not a reliable social network, but the alternatives are not as good.
Its on its infancy and remember this is not run by a big corporation its run by volunteers. If you dont like feel free to go back to reddit.
Edit: Hey guys so yeah this comment was too harsh against newcomers to the platform and I would like to apologize.
While you are correct, your tone is quite dismissive and unwelcoming. If it’s to be a replacement, it’s reliability is relevant. On most of the update posts ive seen, users are generally appreciative of the work done.
Yeah, if it’s a roll of the dice whether the platform will be accessible on any day, people will eventually go somewhere else.
Lemmy.world isn’t meant to replace Reddit. The fediverse and various implementations of the ActivityPub protocol are. Lemmy and Kbin appear to be some of the frontrunners, but lemmy.world isn’t a lynchpin. Other Lemmy instances are up. And it’s a bit naive to expect mostly volunteer servers to have the same uptime as a well-founded corporation. And let’s remember, reddit used to go down a lot more early in its life compared to late in life (and it’d still go down then).
Plus, it’s still a bit entitled to expect uptime just because they have an intent to provide what they believe is a better service than reddit. It’s not only about folks ingesting content, but also those who create and moderate.
It’s reliability has been pretty good all things considered.
I don’t disagree, but discussing reliability and access shouldn’t be met with suggestions to leave the community.
It may be entitled to expect uptime for a volunteer service but it’s also entitled of the project as a whole to expect users if the service is not usable.
I’ve had good reliability. My home instance advises when updated are planned and are normally for minutes. The timeline is generally rough estimates, with the caveat that the admin is doing it outside their usual work hours. The users thank them for the update and take an interest.
Maybe we should just sticky this post 😫
I’m talking out of my ass, but software devs rarely think about scalability, backup, and high availability.
No, they absolutely do. The problem is generally affording scalability and high availability. Those aren’t just free with the right code. Those are essentially features that need to be paid for by the instance owners. The fact that Lemmy works at the scale it’s at is a sign that they pay attention to those things.
The Lemmy.world server runs on a 16mhz 386dx and has 16mb of ram. It just does that sometimes.
Is this a rhetorical question?
