perigrin · October 25, 2016 16:10
diff --git a/matrix-request.log b/matrix-request.log
 10:06 -!- dcg_ [[email protected]] has joined #magnet-srb
 10:07  dcg_: perigrin:  ether: hi
 10:09  dcg_: Are you the correct people to approach about permission to connect a bridge from a matrix Home Server to irc.perl.org
 10:09  dcg_: If you are not familiar with matrix mor information can be found at matrix.org
 10:12  dcg_: Currently I'm involved with a perl based project http://ledgersmb.org and we have moved from using clients directly connected to #ledgersmb@freenode to matrix clients (riot.im mainly) for development and support discussion.
 10:13  dcg_: As we are now looking to add dancer to our application we would like to bridge at least #dancer to a matrix room as well.
 10:17  perigrin: sungo would be the one to talk to. im at breakfast currently can I get back to you in a bit?
 10:18  perigrin: if it operates as a bot you dont need permission, if you need server level interconnect we will have to take it to the network opers
 10:18  dcg_: Perigrin hi.
 10:19  dcg_: Basically it works by connecting as multiple ghost users.
 10:19  dcg_: One ghost per real matrix user
 10:20  dcg_: On some networks this doesn't cause any problems. On others, such as freenode this requires special network side config to allow multiple users on a single ip
 10:22  dcg_: Recently ipv6 support has been activated allowing one user per ipv6 address which has simplified things connecting to freenode although that may be overkill in this instance.
 10:23  perigrin: yeah gimme 15 to finish up here
 10:23 -!- Irssi: #magnet-srb: Total of 14 nicks [2 ops, 0 halfops, 0 voices, 12 normal]
 10:24  dcg_: No problem.
 10:43  perigrin: right so from a technical persepctive I don't think there will be an issue
 10:43  perigrin: but yeah definitely need to clear that with the network opers
 10:43 -!- Irssi: #magnet-srb: Total of 14 nicks [2 ops, 0 halfops, 0 voices, 12 normal]
 10:44  dcg_: perigrin ok. Thanks.
 10:45  dcg_: Do I need to do anything further right now or will the network operators see this discussion and respond here?
 10:46  perigrin: I'm filling sungo in on the relevant details and will let you know. Nothing you need to do currently.
 10:46  dcg_: BTW the best way to contact us is via
 10:47  dcg_: https://matrix.to/#/#ledgersmb:matrix.org
 10:48  dcg_: And select one of the available clients.
 10:48  dcg_: riot.im is probably the easiest as it's a full featured web client
 10:50  dcg_: I'm dcg_mx over there and John is the other dev that will be working on providing and maintaining the bridge
 10:50  dcg_: I'll keep my android irc client running here too
 10:53  perigrin: dcg_: okay short answer is no. The setup you're describing would mean we wouldn't be able to use our normal controls for dealing with individuals if there was an issue.
 10:53  dcg_: Ok.
 10:54  dcg_: Can you describe (here or in a pm) what controls you use. We can likely address any issues fairly easily
 10:55  perigrin: Also we're not interested (or really capable) of making one-off configuation changes to allow *your* network configuation without opening it up for everyone (as best I understand it)
 10:55  perigrin: dcg_: namely we use /kill and /kline to enforce network policy, and if everyone is sharing a host that becomes severly limited.
 10:57  dcg_: Ok. I believe that's been handled correctly with the move to ipv6
 10:57  dcg_: Is your network v6 enabled?
 10:58  perigrin: alh: actually you may know the answer to that better than I do ... I don't *think* we are
 10:58  perigrin: I know *I* aint.
 10:58  alh: I'm guess know
 10:58  alh: WOW
 10:58  alh: I'm guessing no.
 11:00  dcg_: Hmm. Ok. I know the matrix.org team had a working solution for that with ipv4 on freenode
 11:00  dcg_: What I'm not sure of is if it required specific changes on the freenode end.
 11:01  dcg_: I'll see if I can get the bridge developer to drop in here. He can more easily clarify how that was set up.
 11:03  dcg_: Perigrin for our immediate needs we would only have a small number of users on the bridge and all would be trusted by us (we'd have to create accounts for them on our server anyway).
 11:04  dcg_: If one required Kline etc then we could collectively assume responsibility and accept the consequence of all being blocked.
 11:05  dcg_: To be honest I'd be extremely surprised if measures like that would ever be required with this team.
 11:05  dcg_: There are likely only a max of 5 users involved here.
 11:06  perigrin: Honestly I would too. But it's more about precedent and policy which is currently to deny all such requests becasue of the tangle of issues it involves.
 11:08  perigrin: I will copy this conversation to sungo, but as of right now the answer is no.
	10:06 -!- dcg_ [[email protected]] has joined #magnet-srb
	10:07 dcg_: perigrin: ether: hi
	10:09 dcg_: Are you the correct people to approach about permission to connect a bridge from a matrix Home Server to irc.perl.org
	10:09 dcg_: If you are not familiar with matrix mor information can be found at matrix.org
	10:12 dcg_: Currently I'm involved with a perl based project http://ledgersmb.org and we have moved from using clients directly connected to #ledgersmb@freenode to matrix clients (riot.im mainly) for development and support discussion.
	10:13 dcg_: As we are now looking to add dancer to our application we would like to bridge at least #dancer to a matrix room as well.
	10:17 perigrin: sungo would be the one to talk to. im at breakfast currently can I get back to you in a bit?
	10:18 perigrin: if it operates as a bot you dont need permission, if you need server level interconnect we will have to take it to the network opers
	10:18 dcg_: Perigrin hi.
	10:19 dcg_: Basically it works by connecting as multiple ghost users.
	10:19 dcg_: One ghost per real matrix user
	10:20 dcg_: On some networks this doesn't cause any problems. On others, such as freenode this requires special network side config to allow multiple users on a single ip
	10:22 dcg_: Recently ipv6 support has been activated allowing one user per ipv6 address which has simplified things connecting to freenode although that may be overkill in this instance.
	10:23 perigrin: yeah gimme 15 to finish up here
	10:23 -!- Irssi: #magnet-srb: Total of 14 nicks [2 ops, 0 halfops, 0 voices, 12 normal]
	10:24 dcg_: No problem.
	10:43 perigrin: right so from a technical persepctive I don't think there will be an issue
	10:43 perigrin: but yeah definitely need to clear that with the network opers
	10:43 -!- Irssi: #magnet-srb: Total of 14 nicks [2 ops, 0 halfops, 0 voices, 12 normal]
	10:44 dcg_: perigrin ok. Thanks.
	10:45 dcg_: Do I need to do anything further right now or will the network operators see this discussion and respond here?
	10:46 perigrin: I'm filling sungo in on the relevant details and will let you know. Nothing you need to do currently.
	10:46 dcg_: BTW the best way to contact us is via
	10:47 dcg_: https://matrix.to/#/#ledgersmb:matrix.org
	10:48 dcg_: And select one of the available clients.
	10:48 dcg_: riot.im is probably the easiest as it's a full featured web client
	10:50 dcg_: I'm dcg_mx over there and John is the other dev that will be working on providing and maintaining the bridge
	10:50 dcg_: I'll keep my android irc client running here too
	10:53 perigrin: dcg_: okay short answer is no. The setup you're describing would mean we wouldn't be able to use our normal controls for dealing with individuals if there was an issue.
	10:53 dcg_: Ok.
	10:54 dcg_: Can you describe (here or in a pm) what controls you use. We can likely address any issues fairly easily
	10:55 perigrin: Also we're not interested (or really capable) of making one-off configuation changes to allow your network configuation without opening it up for everyone (as best I understand it)
	10:55 perigrin: dcg_: namely we use /kill and /kline to enforce network policy, and if everyone is sharing a host that becomes severly limited.
	10:57 dcg_: Ok. I believe that's been handled correctly with the move to ipv6
	10:57 dcg_: Is your network v6 enabled?
	10:58 perigrin: alh: actually you may know the answer to that better than I do ... I don't think we are
	10:58 perigrin: I know I aint.
	10:58 alh: I'm guess know
	10:58 alh: WOW
	10:58 alh: I'm guessing no.
	11:00 dcg_: Hmm. Ok. I know the matrix.org team had a working solution for that with ipv4 on freenode
	11:00 dcg_: What I'm not sure of is if it required specific changes on the freenode end.
	11:01 dcg_: I'll see if I can get the bridge developer to drop in here. He can more easily clarify how that was set up.
	11:03 dcg_: Perigrin for our immediate needs we would only have a small number of users on the bridge and all would be trusted by us (we'd have to create accounts for them on our server anyway).
	11:04 dcg_: If one required Kline etc then we could collectively assume responsibility and accept the consequence of all being blocked.
	11:05 dcg_: To be honest I'd be extremely surprised if measures like that would ever be required with this team.
	11:05 dcg_: There are likely only a max of 5 users involved here.
	11:06 perigrin: Honestly I would too. But it's more about precedent and policy which is currently to deny all such requests becasue of the tangle of issues it involves.
	11:08 perigrin: I will copy this conversation to sungo, but as of right now the answer is no.