sorted by: new top controversial old
[-] Candelestine@lemmy.ca 1 points 1 year ago

... they can't find someone to come up with trivia questions worded in reverse word order? That's really not that hard.

Ooh! They should use ChatGPT!

[-] Candelestine@lemmy.ca 4 points 1 year ago

This comment is hilarious, and it being downvoted is sad.

[-] Candelestine@lemmy.ca 9 points 1 year ago

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.

[-] Candelestine@lemmy.ca 3 points 1 year ago

Surprised this one took so long. We've had basic hologram tech for decades now. Even with a private jet, it's not like flying cross country all the time for business is fun or anything. Being on a jet is still being on a jet, and not being able to do anything except pull out your laptop, mobile device or book.

[-] Candelestine@lemmy.ca 8 points 1 year ago

I would describe it as a cacaphonic symphony that you eventually get used to. It packs as much information into one sense as you can get from your other four put together.

Much like how you can discern an individual instrument type in a symphony, sight lets you discern individual objects from afar, and gives you a mostly accurate summary of its basic properties.

Also much like with sound, it can be very appealing or unappealing, depending. There's an intrinsic beauty to the sense itself though. Every object has color, for instance, and color is more like smell. It can give you hints about what something is, but its mostly an arbitrary blend of different "flavors" that combine to create more complex examples.

It's the super-sense, the one sense that binds them all. When one of your other four detects something, your first instinct is to locate it with sight to determine more information before you do anything else. You "look at it" first. Almost without fail.

[-] Candelestine@lemmy.ca 62 points 1 year ago

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.

[-] Candelestine@lemmy.ca 4 points 1 year ago

I'm not him, but now that I think about it, there is a tendency for many people to prefer the more generalized term.

Where scientists don't tend to use the word scientist as much, I can't recall ever seeing the term in a journal article for instance. (I don't read many, but I'll read an abstract here and there) I'm not sure why. I expect it's some categorization thing, where not all scientists perform research, so researcher is the more precise term. I'm just guessing as to the reason though, I do not have a PhD.

[-] Candelestine@lemmy.ca 10 points 1 year ago

Problem with attacking stupidity is its not necessarily fixable. We do not attack people over things they cannot change, like the color of their skin or their sexual orientation.

How do they change their innate intelligence? We're not all gifted with the same amount. Can your system apply to someone who takes 5 minutes to learn the definition of even one new word? Someone who needed remedial classes, because the average classes were beyond their ability?

We need a system that allows for them too. So, asking for intelligence is asking too much, so that the execution of the method is easily within everyone's capabilities. Thus, back to the drawing board.

[-] Candelestine@lemmy.ca 4 points 1 year ago

I agree, the cross-posting gets annoying. Why do people insist that everyone who is interested in a certain topic needs to participate in their post, so it has to go on every community?

People did not do that on reddit. They just made one post and waited for interaction.

[-] Candelestine@lemmy.ca 7 points 1 year ago

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.

[-] Candelestine@lemmy.ca 1 points 1 year ago

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?

[-] Candelestine@lemmy.ca 1 points 1 year ago

Correct, if you only use Lemmy, you would not need to worry. But most people still use google products.

view more: next ›

Candelestine

joined 1 year ago