Avatar

r00ty

r00ty@kbin.life
Joined
1 posts • 256 comments

I’m the administrator of kbin.life, a general purpose/tech orientated kbin instance.

Direct message

That dystopian “future”. It’s the present, isn’t it.

permalink
report
parent
reply

Yeah, I’m quite sure it’s a deliberate activity to dissuade against private email servers. Keep everyone’s email “in the club”. Once you’ve got this much working you need a whole suite of tools to deal with the HUGE amount of spam you need to filter. It can be a hell of a lot.

permalink
report
parent
reply

kbin/mbin does have some mastadonesque facilities. So it straddles the line between threadiverse and I dunno what we call the mastadon side.

permalink
report
parent
reply

IRC was “kinda” federated. You needed to convince a server already in the network to accept your server. But in the early days requirements were quite low.

BBS was not really federated (except Fidonet I guess).

Usenet, I guess it kinda was. But only ISPs were really running NNTP servers. Only they and unis really had the resources to too.

permalink
report
parent
reply

You CAN do the full list of things to get accepted there. But you only need to fail a SINGLE test to get sent to junk mail jail.

To not be put to junk you need all of the following (oh and this can and will change one day and you’ll go straight to junk)

  • SPF configured
  • DKIM configured with valid keys applied to DNS
  • DNS secured with DNSSEC, with validated keys passing all minimum requirements
  • DMARC configured for domain
  • Your mail server NOR the entire network on a DNSRBL. For example right now my mail server is hosted on OVH (moving soon) and it will go to junk, and in the hotmail/outlook headers it makes clear this is the only failure (-0.2 points, enough to go straight to junk mail jail)

Not sure if I missed any there. It’s been a while since I set all this crap up.

permalink
report
parent
reply

2.49 extra for assured personal rioter. Otherwise you’ll get the standard service where your rioter may have another riot to attend to first.

permalink
report
parent
reply

Maybe someone can make an app, so I can have someone paid at an insultingly low hourly rate to go protest/riot in my place?

permalink
report
parent
reply

Well it seems it was more to do with sanctions, if the open letter from one of the chopped developers is to be believed. In which case, I think the right thing is to move the names to contributors (they did still contribute), remove them from maintainers (some maintainers are actually paid by the foundation, I mean not a lot, but some are paid).

I still find it all a little odd. But likely there was a bit of a prod from somewhere higher as to how sanctions should be followed.

permalink
report
parent
reply

Most nuclear enabled countries have nuclear subs. I believe here in the UK our entire nuclear deterrent is based on trident missiles fired from submarines.

permalink
report
parent
reply

It’s going to be precisely the reason. If you have a dedicated wire, fibre or copper then the entire available bandwidth is available per connection (one caveat with copper is crosstalk but it is minimal and can be mitigated). With fibre the available bandwidth per strand is huge.

It’s so fast that even where there’s contention, it is rarely a problem that everyone sharing a part of the connection is downloading or uploading at once. So pretty much most of the times you test, you get the full speed.

With mobile data, the entire cell is sharing a small amount (in comparison) of spectrum. Unlike a wire, the entire spectrum cannot be used by a single tower, a pretty small number of channels are carved out for them. Also because the signals are travelling through the air, there is more of a problem of signal loss and interference to contend with, so the channels very rarely reach the maximum possible speed (forward error correction and reducing bits per symbol to reach a suitable signal to noise ratio both will reduce speed for example.

For upload (which isn’t usually much of an issue) there’s another problem of guard time between timeslots. When downloading, the cell transmitter transmits the whole time and shared the channel between all users (another thing that can slow things down) so there’s no problem of needing a guard time. But when it’s separate transmitters (phones) sending there’s going to be a guard time between different handsets timeslot and the more active transmit stations there are (phones) the more these guard times add up to wasted bandwidth. Luckily most people are downloading far more than uploading, so it’s less of an issue.

I think for these reasons caps are used to limit people from ALWAYS consuming data on the cell/mobile networks and instead using wifi wherever they can in order to keep it fast for those that do/need to.

permalink
report
parent
reply