If either of those figures is actually accurate from an end-user standpoint, then the entire downtime must be coming during my primary periods of usage.

It’s been this way for weeks, actually. I haven’t seen a graph of the uptime, but I’m sure one would look extremely ugly, based on my own user experience.

This right here is an alt, and despite the fact that I don’t prefer to comment from it, since I won’t necessarily check in soon to see replies, it’s seeing some heavy use.

The attacks a few weeks ago weren’t a one-off, they never stopped. It seems down maybe half the time or so?

One of the many ways we (all of Lemmy) are not quite ready for the mainstream yet, we still have basic technical/security issues to resolve. Soon, though.

We’re getting there, still in the very early stages here. One thing I’ve noticed is how extremely techy the initial community here was, something I personally collided with like a bit of a wrecking ball. People in general, not just techy people, tend to assume others will approach things similarly to how they naturally do. So they don’t necessarily always see problems that others might stumble over, ahead of time.

Now that we’ve started growing more rapidly, these problems of scale, where they now have to anticipate problems they did not have to anticipate before, all are coming due. So, growing pains.

This is why I have not been inviting people to Lemmy yet, I’ve been waiting until it’s more polished for the mainstream. It’s also why the graph is trending down. We’re literally not ready yet for the mainstream, in many, many different ways.

Also useful to remember, we’re only done getting big growth spikes if spez is done pissing off reddit. I doubt he is.

Why does the purpose the data was collected for matter? Either the data is suitable or not. The motive of the pollsters who gathered it is irrelevant, isn’t it?