this post was submitted on 21 Jul 2023
454 points (98.9% liked)
Lemmy
523 readers
2 users here now
Everything about Lemmy; bugs, gripes, praises, and advocacy.
For discussion about the lemmy.ml instance, go to [email protected].
founded 4 years ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
My thoughts exactly. You should not be choosing TLDs that are volatile to upsets like this. Stick with the tried and true .com or .net, or one of the new TLDs that are not bound to a nation (unless you can comply with the stipulations) or particular type of organization.
Or if you absolutely have to, choose the TLD of a country you live in.
Agreed. I went with lemmy.ca since I'm Canadian and the instance is in my country.
I also heard Lemmy should perform a little quicker for me too this way.
No, practically speaking the domain name should have no effect on access time. DNS has so many layers of caching that as long as SOMEONE has accessed the website nearby (including you), the domain lookup will be local and therefore fast.
Anyway, DNS lookup times, even slow ones, are still not going to be noticable to the end use originally.
No, I meant the instance itself. The server. The one who runs lemmy.ca is here in Canada with me.
It's like when playing a game; You choose servers closet to you for the lowest ping time.
The other reason I neglected to mention was I like to support local. 😎
It makes a difference for a game, but it's not really significant for a website.
The server load and resources will have a much bigger impact on performances than geographic proximity.
And you spread that server load by selecting different servers. While what you're saying is technically true, in a practical sense if everyone picked a more local server that would be one way to achieve what you're saying.
No because the population is not even close to being uniformly distributed geographically.
you don't need a uniform distribution. if the server distribution mirrors the population distribution (and why wouldn't it?), that will still achieve the desired effect.