You should also be aware of heapoverflow.ml for programming questions. Unfortunately, it doesn’t look terribly active at the moment.
It is just as you said. As an example you can look at heapoverflow.ml which got domain high jacked. It is down but the content is still available on other instances:
Someone should setup a “HeapOverflow” instance specifically for this…
Heapoverflow?
I think by default people choose instances related to their region. That way you can communicate with likeminded people in the timeframe they’re the most active.
If not that then you look at the topics instance specialises in. For example:
If you don’t care about specialization and prefer simply more content you choose general instance such as beehaw or sopuli
I think https://heapoverflow.ml/ was trying to do this!
Done! HeapOverflow.ml
Not the author but I tested some alternatives. Seems like alternative git forges often support IPv6, rather few Lemmy instances do, but many other alternatives to HN/reddit do. I could not find an easy accessible list of Mastodon instances to check that. Overall the state of IPv6 in 2022 is pretty sad…
lemmygrad.ml: NO
lemmy.ml: YES szmer.info: NO
lemmy.eus: NO
wolfballs.com: NO
lemmy.ca: NO
collapse.cat: NO
feddit.de: NO
sopuli.xyz: NO
beehaw.org: NO
mander.xyz: NO
fapsi.be: YES
midwest.social: NO
lemmy.juggler.jp: NO
mandacaru.caatinga.digital: NO
tabinezumi.net: YES
baraza.africa: NO
forum.nogafam.es: NO
buckeyestate.social: NO
lm.korako.me: YES
stammtisch.hallertau.social: NO
Heapoverflow.ml: NO
eope.xyz: NO
heapoverflow.ml: NO
lemmy.tedomum.net: YES
lemmy.schuerz.at: NO
lemmy.cat: NO
forum.nobigtech.es: NO
lemmy.mesh.party: YES
lemmy.shrieker.net: NO
lem.ph3j.com: NO
lobste.rs: YES
tildes.net: YES
raddle.me: YES
slashdot.org: NO
saidit.net: YES
steemit.com: YES
20-things.com: YES
www.4chan.org: NO
hexbear.net: YES
codeberg.org: YES
notabug.org: YES
savannah.gnu.org: YES
savannah.nongnu.org: YES
sourceforge.net: NO
bitbucket.org: YES
===
18/46 support IPv6
==
import socket
import json from urllib.request import urlopen
LEMMY_INSTANCE_LIST_URL = "https://lemmymap.feddit.de/instances.json"
OTHER_DOMAINS = [
# reddit & HN alternatives
"lobste.rs",
"tildes.net",
"raddle.me",
"slashdot.org",
"saidit.net",
"steemit.com",
"20-things.com",
"www.4chan.org",
"hexbear.net",
# Twitter alternatives
# TODO find list of mastodon instances
# Github & Sourcehut alternatives
"codeberg.org",
"notabug.org",
"savannah.gnu.org",
"savannah.nongnu.org",
"sourceforge.net",
"bitbucket.org",
# ...
]
def get_lemmy_instances():
with urlopen(LEMMY_INSTANCE_LIST_URL) as response:
body = response.read()
obj = json.loads(body) return [instance["domain"] for instance in obj["instance_details"]]
domains = []
domains.extend(get_lemmy_instances())
domains.extend(OTHER_DOMAINS)
total = 0
for domain in domains:
print(domain, ": ", end="", sep="")
try:
addrinfo = socket.getaddrinfo(domain, "http", socket.AF_INET6)
print("YES")
total += 1
except Exception:
print("NO")
print("===")
print("{}/{} support IPv6".format(total, len(domains)))
print("===")
Yep this is a serious issue we’ve been dealing with, that we need to fix before the next release.
You mean that you want to stay within lemmy.ml, connected to your user, and replace the content with the content from heapoverflow.ml? Then you would be able to explore instances more easily and find new communities to subscribe to.
There is no way to do that at the moment. Lemmy.ml only fetches some of the content from other instances - such as the content that lemmy.ml’s users are subscrribed to, and the content that is manually requested by pasting the original Url into the searchbox.
I can’t think of an obvious way that this can be implemented without fetching a large amount of unsued content from the other instance, and that approach would become very resource inefficient as instances become larger.
Another approach could be to create instance relays or explorers that are modified instances that will focus their resources on discovery by fetching a lot of content that might live for a week or two in the instance, and then build tools into it such as filtering by instance. This instance would subscribe to every community of the instances in its list. You would still need to leave your instance’s site to explore instances, unless the relay is somehow connected to the original instance, but at least this would be an efficient way to explore.
As of now, what I do is that I manually explore instances by accessing their site in a new tab, and manually subscribe to new communities by copying and pasting the community Url into mine.
here are the js-clients docs:
https://join-lemmy.org/api/
also: this should probably have been posted to !javascript@heapoverflow.ml =/
I didn’t know heapoverflow’s domain got hijacked, do you perhaps know more ? and if they got a new domain ?
We had a stackoverflow alternative called heapoverflow.ml, which was it’s own lemmy server with communities for different programming languages etc. But after a year or something and not much use it was shut down unfortunately :(
honestly, !bash@heapoverflow.ml would’ve been most appropriate, considering my answer was xargs
Linux --> /c/linux@heapoverflow.ml
I would just ask it in the js & webdev community =]
heapoverflow.ml seems to be down.
Done! I created a generic crypto sub. Maybe I’ll have to make more implementation-specific ones, but hopefully everyone can get along and live under one roof =/
If the people want heapoverflow, I’ll resurrect it