Meeting 2016 09 23

Call for meeting

Friday 2016/09/23 at 20:00 UTC

Agenda

Feel free to add to this before the meeting!

• Conservancy Application update
• Pump.io code development
  • 1.0.0 release (celebrate)
  • Does utml-to-jade imply semver-major?
    • If so I (strugee) would like to release 2.0.0 soon
    • But generally it seems unsustainable
  • Express 3.x update
  • Telemetry (#1187)
  • Removing plugins (for now) (#1194)
    • Other option: explicitly state that they won't be semver-major
  • Code coverage (#1193)
  • ActivityStreams 2.0?
  • (If we have time) weird code in lib/ and test/
• Community governance and sustainability
  • Node adoption (sponsoring, administering)
  • Adoption of non-node related websites (e.g. OpenFarmGame, ih8.it)
  • Governance (non-profit status/application)
  • Funding
• Other stuff
  • T-shirt designs

Log

[22:04:33] <larjona> #############################################################

[22:04:33] <larjona> BEGIN LOG

[22:04:33] <larjona> #############################################################

[22:04:34] <evanpro> Hey

[22:04:42] <evanpro> Let's do this!

[22:04:45] <larjona> Dear all, welcome to this Pump.io community meeting!

[22:04:45] <larjona> Agenda is here:

[22:04:45] <larjona> https://wiki.debian.org/Teams/Publicity/Meetings/2016-09-23

[22:04:45] <larjona> First, as always, roll call meanwhile people is coming. Who's here? Say hello!

[22:04:45] -*- larjona is here ;)

[22:04:55] <evanpro> evanpro is here

[22:04:58] -*- strugee is here

[22:05:13] <larjona> pumabot is on strike

[22:05:21] <strugee> larjona: that's the publicity team :P

[22:05:35] <larjona> oh sorry

[22:06:07] <larjona> agenda is here: Dear all, welcome to this Pump.io community meeting!

[22:06:07] <larjona> Agenda is here:

[22:06:07] <larjona> https://wiki.debian.org/Teams/Publicity/Meetings/2016-09-23

[22:06:07] <larjona> First, as always, roll call meanwhile people is coming. Who's here? Say hello!

[22:06:07] -*- larjona is here ;)

[22:06:10] <strugee> pumabot: meeting agenda

[22:06:11] <pumabot> strugee: there isn't currently a meeting going on.

[22:06:21] <evanpro> Uh

[22:06:29] <strugee> larjona: same thing

[22:06:30] <larjona> agenda is here: https://github.com/e14n/pump.io/wiki/Meeting-2016-09-23

[22:06:33] <strugee> :)

[22:06:37] -*- strugee is still here

[22:06:57] -*- evanpro is here

[22:07:14] <larjona> sorry, copypaste problems + lag...

[22:07:24] <strugee> no worries

[22:07:53] <larjona> so, first topic: Conservancy Application update

[22:07:59] <larjona> Any news? I have no news

[22:08:04] <evanpro> I have no news either

[22:08:08] <evanpro> Hold on, lemme check

[22:08:44] <strugee> FWIW bkuhn told me at a conference that Conservancy has a large backlog, so it may be a while

[22:09:07] <evanpro> Yeah it's always really long

[22:09:19] <evanpro> So I just pinged them by email to confirm that the application was received

[22:09:23] <larjona> thanks

[22:09:28] <strugee> thx!

[22:09:37] <larjona> I guess we can go to next topic

[22:09:43] <larjona> Topic: Pump.io code development

[22:09:47] <evanpro> Awesome

[22:09:51] <evanpro> We have a lot of these

[22:09:51] <larjona> 1.0.0 release

[22:09:56] <larjona> hey!!

[22:10:03] <evanpro> \o/

[22:10:05] <evanpro> Nice job

[22:10:12] <strugee> \o/ whoooooooooooo

[22:10:24] <larjona> I think some people already upgraded their production servers successfully

[22:10:30] <strugee> there's not much to discuss here, I just added it to the agenda so we can celebrate :)

[22:10:35] <larjona> ah, fine

[22:10:51] <strugee> evanpro: I noticed e14n.com is running the new release :)

[22:10:54] <strugee> also we got on LWN!

[22:10:56] <evanpro> Is it?

[22:11:02] <evanpro> I guess so

[22:11:07] <strugee> according to the Server header

[22:11:14] <bkuhn> strugee: it shouldn't be THAT long but it will definitely be a month or two

[22:11:16] <evanpro> OK, I'll get the rest of the sites running it

[22:11:24] <larjona> thanks bkuhn!

[22:11:26] <strugee> bkuhn: noted, thanks

[22:11:31] <evanpro> bkuhn: thanks!

[22:11:43] <larjona> evanpro: that's good news (e14n servers upgrading soon :)

[22:11:45] <evanpro> strugee: I don't think there's anything important to do except push the new code, right?

[22:11:52] <strugee> nope

[22:12:02] <strugee> it's mostly just bugfixes, real easy

[22:12:20] <evanpro> I might take the opportunity to move to Docker

[22:12:33] <evanpro> We have an official docker image don't we?

[22:12:37] <strugee> nope

[22:12:41] <strugee> there are one or two unofficial ones

[22:12:58] <strugee> #789

[22:12:58] <pumabot> Issue 789: Docker image (OPEN - https://github.com/e14n/pump.io/issues/789)

[22:13:13] <evanpro> We should move to an official one

[22:13:19] <strugee> definitely

[22:13:30] <larjona> this may be linked to next subtopic? Release procedure (e.g. signing?)

[22:13:37] <evanpro> Sure

[22:13:41] <strugee> oh actually I think that item is outdated

[22:13:55] <strugee> it's been on there for many months

[22:13:59] <larjona> ah

[22:14:00] <evanpro> X it off

[22:14:14] <strugee> I signed the git tag, FWIW

[22:14:20] <evanpro> Great

[22:14:26] <evanpro> Did you npm publish it?

[22:14:29] <strugee> yep

[22:14:34] <evanpro> :thumbsup:

[22:14:38] <evanpro> OK!

[22:14:40] <larjona> ok, who can be in charge of the official Docker image, then? I guess that person needs to build+sign+upload it somewhere?

[22:14:42] <strugee> http://pump.io is updated as well

[22:14:50] <evanpro> I saw that! Nice!

[22:15:01] <evanpro> So, question about docker image

[22:15:13] <evanpro> it should probably have the same "namespace" as the github repo

[22:15:29] <evanpro> We talked about setting up a "pump-io" or "pump.io" organization

[22:15:42] <strugee> gitlab.com/pump.io exists

[22:16:42] <evanpro> Oh

[22:16:47] <evanpro> So we'd do it on Gitlab?

[22:16:50] <strugee> idk

[22:16:55] <strugee> all it has is the ReadTheDocs repo

[22:17:00] <evanpro> OK

[22:17:08] <evanpro> I actually like gitlab

[22:17:16] <strugee> I didn't mean that as "it exists, therefore..." I was just pointing it out

[22:17:19] <strugee> I'd be fine with GitLab

[22:17:20] <evanpro> OK

[22:17:29] <-- drymer left (Ping timeout: 240 seconds)

[22:17:30] <evanpro> Let's do this

[22:17:36] <evanpro> Let's start off with github.com

[22:17:45] <evanpro> strugee: I just added you to the team

[22:17:56] <strugee> awesome, thanks. was going to ask you about that anyway :)

[22:17:56] <evanpro> We can transfer the pump.io repo to that organization

[22:18:12] <evanpro> And I think there are ways to mirror to gitlab

[22:18:26] <evanpro> If there's a strong reason to move to Gitlab we can reverse that direction

[22:18:46] <strugee> GitLab has a built-in GitHub importer but it can't mirror out of the box AFAIK

[22:18:51] <evanpro> OK

[22:18:57] <strugee> I can look into it though

[22:19:37] <larjona> so...?

[22:20:14] <evanpro> So I just moved e14n/pump.io to pump-io/pump.io

[22:20:26] <evanpro> Making it officially confusing

[22:20:51] <evanpro> But github does the redirection automatically

[22:21:44] <evanpro> I gave you two admin access

[22:21:48] <evanpro> So we should be good to go

[22:21:50] <strugee> ok so is the plan to use https://github.com/pump-io/pump.io, and try to mirror to GitLab?

[22:21:57] <evanpro> Yes if that's OK

[22:22:01] <strugee> ok

[22:22:15] <strugee> sounds good

[22:22:33] <evanpro> That name is terrible but that's how github does their organization names

[22:22:40] <evanpro> No dots

[22:22:52] <evanpro> I guess we could do pump-io/pump-io

[22:23:01] <strugee> eh

[22:23:01] <evanpro> At least it would look better

[22:23:07] <evanpro> OK, we'll see

[22:23:18] <strugee> I think it looks ok. if you want to go ahead but

[22:23:22] <evanpro> I don't think you can have hyphens in the name for Docker

[22:23:28] <evanpro> It's been a while

[22:23:38] <evanpro> OK, so, Docker image to come soon

[22:23:51] <evanpro> One thing I'd like to do is move away from the /etc/pump.io.json for configuration...

[22:23:56] <evanpro> ...and use yargs

[22:24:17] <evanpro> It's a nice library that can take configuration data from the command line, a config file, or environment variables

[22:24:21] <evanpro> I find it really useful

[22:25:08] <larjona> So, our canonical url for the repo now is https://github.com/pump-io/pump.io or this is only for the Docker image?

[22:25:11] <pumabot> Issue #1202 "Use yargs for command-line options" opened by evanp - https://github.com/pump-io/pump.io/issues/1202

[22:25:15] <evanpro> https://github.com/pump-io/pump.io/issues/1202

[22:25:26] <evanpro> No, that's for everything

[22:25:31] <strugee> larjona: that's the canonical URL for everything

[22:25:32] <larjona> ok, thanks

[22:25:43] <evanpro> Like I said, github.com will redirect the old URLs automatically

[22:25:57] <strugee> so I just checked and apparently GitLab themselves mirror to GitHub, but only the git repo itself

[22:26:07] <strugee> their issue tracker is on gitlab.com

[22:26:15] <evanpro> RIght

[22:26:24] <evanpro> I don't think you can mirror the issues

[22:26:28] <strugee> right

[22:26:28] <evanpro> Just the git repo

[22:26:35] <strugee> is that what we want to do?

[22:27:24] <evanpro> Well

[22:27:32] <evanpro> Hmmm

[22:27:42] <evanpro> So I guess I was trying to soft-pedal any move to gitlab

[22:27:49] <strugee> right

[22:27:50] <strugee> we can always do it later

[22:27:51] <evanpro> Like, we could maintain a presence at gitlab.com

[22:27:57] <evanpro> By mirroring the repos

[22:28:17] <strugee> ok

[22:28:25] <evanpro> Currently I'd rather stay at github.com but I could be convinced otherwise

[22:28:29] <strugee> I'd be fine with that as long as we don't take issues/MRs on GitLab

[22:28:38] <evanpro> Oh, really?

[22:28:46] <evanpro> Oh, I see what you mean

[22:28:49] <evanpro> Like, do it in one place

[22:28:56] <strugee> exactly

[22:29:05] <evanpro> We'd probably have to put something up like "JUST A MIRROR DON'T GIVE ANY ISSUES"

[22:29:19] <strugee> no you can just straight up disable issues and MRs

[22:29:23] <strugee> so the option doesn't even appear

[22:29:33] <evanpro> Oh great

[22:29:43] <evanpro> So, if you want to do that, I'm cool with it

[22:29:47] <strugee> sounds good

[22:30:04] <evanpro> Let's keep moving -- I don't want to take up the whole meeting with this

[22:30:08] <strugee> sounds good

[22:30:33] <strugee> next item: Does utml-to-jade imply semver-major?

[22:30:43] <strugee> we discussed this last month

[22:31:23] <strugee> evanpro: you mentioned that we might want to do a semver-major release (i.e. 2.0.0) because some people modify templates

[22:31:26] --> drymer joined #pump.io

[22:31:30] <evanpro> Yeah

[22:31:34] <evanpro> That's pretty much the case

[22:31:40] <strugee> if we're going with that the next release will be 2.0.0

[22:31:43] <evanpro> Yep

[22:31:48] <evanpro> I think we said that was what we'd do

[22:31:59] <strugee> OK

[22:32:10] <strugee> does that mean anytime the templates are modified we need to bump?

[22:32:17] <evanpro> Ummmmmmmmmm

[22:32:20] <evanpro> I don't know

[22:32:41] <evanpro> I'd say if they change in a way that someone who has templates that work with 1.0.0 that won't work with the next version, yes.

[22:32:50] <strugee> in particular the Express 3.x merge rewrites almost 100% of the templates because the indentation changed

[22:33:03] <evanpro> Yeah

[22:33:13] <evanpro> So, would that be part of the 2.0.0 change?

[22:33:23] <strugee> it seems to me that anyone who's modifying templates will have a source-based install, and will notice because git will show merge conflicts

[22:33:24] <evanpro> Also, I forget why we're upgrading to 3.x and not 4.x

[22:33:38] <strugee> 4.x is after 3.x

[22:33:43] <evanpro> OK

[22:33:59] <strugee> I just figured it made sense to break it up as much as possible

[22:34:15] <strugee> evanpro: not sure

[22:34:17] <-- evanpro left (Quit: evanpro)

[22:34:38] -*- strugee waits for Evan to come back

[22:36:26] --> evanpro joined #pump.io

[22:36:34] <strugee> evanpro: welcome back

[22:36:36] <evanpro> Sorry my laptop batter died

[22:36:42] <strugee> no worries

[22:36:45] <evanpro> So what's going to go into 2.0.0?

[22:37:00] <strugee> so we could have Express 3.x go into 2.0.0

[22:37:05] <strugee> but that's a lot of change for one release

[22:37:16] <evanpro> Well, that's what major releases are for

[22:37:20] <strugee> true

[22:37:30] <strugee> ok so are we going to do that?

[22:37:42] <evanpro> Can we make that the release the one where we deprecate everything before node 4?

[22:37:50] <strugee> yeah, that's the plan

[22:37:54] <evanpro> Woot

[22:37:57] <evanpro> Good

[22:38:07] <strugee> we can't drop support entirely because everyone still ships 0.10 but definitely we can deprecate it

[22:38:20] <strugee> man Node 0.10 is terrible

[22:38:53] <evanpro> Oh really

[22:38:58] <evanpro> :(

[22:39:03] <strugee> I know :(

[22:39:07] <strugee> nodecompat.com has the details

[22:39:36] <strugee> it's getting better, but not quite there yet

[22:39:48] <evanpro> Whyyyyyyyyyyy

[22:39:53] <strugee> evanpro++

[22:39:54] <pumabot> evanpro has 1 point

[22:39:56] <evanpro> At least Debian isn't at 0.6 any more

[22:40:01] <strugee> truly

[22:40:13] <evanpro> OK, so let's keep going

[22:40:16] <strugee> Debian Stretch will have Node 4 though

[22:40:25] <evanpro> Telemetry

[22:40:28] <strugee> yes

[22:40:31] <strugee> so

[22:40:32] <evanpro> That's ofirehose.com and pumplive.com right

[22:41:05] <strugee> I actually was not aware of pumplive.com

[22:41:15] <evanpro> pumplive.com did stats

[22:41:22] <evanpro> Like, everything comes through ofirehose...

[22:41:29] <evanpro> ...and pumplive would count things that came through there

[22:41:33] <strugee> ok

[22:41:44] <evanpro> But it's broken and I can't remember why

[22:41:47] <evanpro> I would love someone to take those over

[22:41:49] <strugee> that fufills some of the goals of what I was proposing, but not all

[22:42:05] <evanpro> The firehose is great for people who want all public posts

[22:42:14] <strugee> I'd love to run a server that would collect (voluntarily-reported) data from nodes

[22:42:18] <evanpro> Riiiiiight

[22:42:30] <evanpro> So, there are two schools of thought on that

[22:42:32] <strugee> there's a list in #1187

[22:43:41] <evanpro> So

[22:43:46] <evanpro> This is going to sound keeeeeeeerazy

[22:43:48] <evanpro> But

[22:43:58] <evanpro> The server could send that telemetry data to ofirehose.com

[22:44:21] <evanpro> Like, I'm a server, I'm at "pumpdog.me", and here's my data.

[22:44:27] <evanpro> So anyone could see it.

[22:44:29] <evanpro> Again, opt-in.

[22:44:35] <strugee> I'd be okay with that

[22:44:44] <evanpro> It'd be just another activity stream.

[22:45:18] <evanpro> And I'll be honest

[22:45:33] <evanpro> People get a little tetchy about having telemetry be opt-out

[22:45:39] <evanpro> But your data is much much better

[22:45:53] <evanpro> I'm pretty sure ofirehose.com is on by default in pump.io

[22:46:02] <evanpro> And we had telemetry in StatusNet

[22:46:16] <evanpro> Mostly the people who didn't want it just turned it off

[22:46:39] <strugee> we could make the telemetry be an opt-in setting and if it's not on, just don't send that particular data to OFirehose

[22:46:47] <evanpro> Right

[22:46:48] <pumabot> Pull Request #1203 "Update the URL of the git repo" opened by larjona - https://github.com/pump-io/pump.io/pull/1203

[22:46:59] <evanpro> OK

[22:47:16] <strugee> so is that the plan? send to OFirehose but only when the admin turns it on?

[22:47:31] <evanpro> Well

[22:47:38] <strugee> to be clear

[22:47:38] <evanpro> I'd make it opt-out but it's your call

[22:47:44] <evanpro> I'm probably not going to use the data muh

[22:47:46] <evanpro> much

[22:47:57] <strugee> regular OFirehose data would still be opt-out

[22:48:08] <strugee> you'd make it opt-out? OK

[22:48:14] <evanpro> Right

[22:48:22] <evanpro> Well

[22:48:24] <evanpro> Hmmm

[22:48:28] <strugee> I think we'd get backlash

[22:48:30] <evanpro> If it's public, then opt-in

[22:48:33] <evanpro> That's probably the best

[22:48:45] <evanpro> OK let's keep going

[22:48:46] <strugee> OK

[22:48:52] <evanpro> I have to leave in 12 minutes rain or shine

[22:48:58] <strugee> ok

[22:49:04] <evanpro> What's up with plugins?

[22:49:20] <strugee> so tl;dr we got a PR extending the plugins API

[22:49:49] <strugee> and I want to get rid of plugins for the time being because essentially every time we upgrade a library internally it may break plugins

[22:50:01] <larjona> #1194

[22:50:13] <larjona> pumabot does not love me today

[22:50:14] <strugee> since we have a lot of stuff to upgrade it seems like a good idea to say "plugins are unsupported right now" and support them when we have less churn in the codebase

[22:50:26] <strugee> larjona: yeah, didn't work for me either. not sure why pumabot is so broken today

[22:50:49] <evanpro> strugee: sorry, I don't get it

[22:50:57] <evanpro> Like, no plugins?

[22:51:04] <evanpro> I forgot we even had them honestly

[22:51:08] <strugee> evanpro: yeah

[22:51:11] <strugee> no plugins

[22:51:19] <evanpro> But they're in 1.0.0?

[22:51:23] <strugee> or make the admin turn on some option

[22:51:28] <strugee> yes, but they're undocumented

[22:51:34] <larjona> https://github.com/pump-io/pump.io/pull/1194

[22:51:37] <evanpro> Were the in 0.3.x?

[22:51:42] <strugee> yeah

[22:52:08] <strugee> > There's rudimentary support for plugins in 0.3.x. I'm really dubious about them; I wouldn't count on continued support.

[22:52:24] <strugee> evanpro: ^^^ is you June 2013

[22:52:30] <evanpro> ha ha

[22:52:35] <evanpro> That sounds like me

[22:52:43] <evanpro> I can't remember why I added them

[22:52:58] <strugee> someone asked for them basically :P

[22:53:06] <evanpro> ha ha

[22:53:31] <strugee> basically plugins are just given the app variable to attach routes, etc.

[22:53:38] <evanpro> Ohhhhhhhhhhhhhhhh

[22:53:50] <strugee> that's why they break when we upgrade e.g. Express internally

[22:53:52] <evanpro> Right

[22:54:05] <evanpro> Well, that's a versioning issue, then

[22:54:24] <strugee> so the way I see it there are three options

[22:54:30] <strugee> 1. don't support plugins

[22:54:37] <evanpro> That plugin API is an external interface...

[22:54:42] <strugee> 2. bump semver-major every time we upgrade a library internally

[22:54:52] <evanpro> Uh

[22:55:00] <strugee> 3. keep plugins, but make it really clear that we don't support them and they may break

[22:55:07] <evanpro> I'd say instead to bump it when the programming API changes for plugins

[22:55:18] <strugee> so option 2?

[22:55:26] <evanpro> So for connect 2 to express 3, I think that's... almost the same?

[22:55:36] <evanpro> Especially for adding routes

[22:55:41] <strugee> no

[22:55:49] <strugee> Express 2 or Conecct 2?

[22:55:55] <strugee> s/Conecct/Connect/

[22:56:00] <evanpro> Heh

[22:56:02] <strugee> Express 2 to Express 3 is huge

[22:56:14] <evanpro> Right

[22:56:45] <evanpro> OK, so, I'd say that the plugin API would change between pump.io 1.x and pump.io 2.x

[22:57:13] <evanpro> But it's still just the app object

[22:57:22] <evanpro> And then after that figure out a more long-term plugin strategy

[22:57:30] <strugee> ok so basically

[22:58:02] <larjona> Friends, 3 minutes to o'clock

[22:58:04] <strugee> keep the current code and we'll switch to a different plugin strategy at 3.0 or something?

[22:58:11] <evanpro> Yes

[22:58:13] <evanpro> That's basically it

[22:58:21] <strugee> okay

[22:58:23] <strugee> larjona++

[22:58:23] <pumabot> larjona has 10 points

[22:58:32] <evanpro> If we're semvering anyway

[22:58:35] <evanpro> We get one freebie

[22:58:38] <strugee> sounds good

[22:58:43] <evanpro> OK, next!

[22:58:56] <strugee> Code coverage

[22:59:03] <larjona> I'd jump to network status/node adoption

[22:59:23] <evanpro> I'm going to stretch another few minutes

[22:59:29] <strugee> mostly all I wanted to ask with this is whether Coveralls is OK, since everyone seems to use it

[22:59:35] <evanpro> Great!

[22:59:41] <evanpro> Fine here

[22:59:44] <strugee> if there's any issues with that then I vote we table it and move on

[22:59:46] <strugee> ok, sounds good

[22:59:50] <evanpro> Let's talk AS2

[22:59:58] <larjona> Ok, AS2

[23:00:03] <strugee> paroneayea and I talked about this last week

[23:00:08] <evanpro> Great

[23:00:27] <evanpro> We need to support to ActivityPub

[23:00:49] <evanpro> Ideally I think we'd do this:

[23:00:55] <strugee> the current plan I have (which IIRC paroneayea approved, but don't quote me) is to store AS2 objects internally and output AS1 objects externally, for compat

[23:01:14] <evanpro> Yeah, that's the kind of plan someone without millions of activities in AS1 would make up

[23:01:35] <strugee> then if the client or the remote server sends an X-Pump-AS2 header or something, indicating that it understands AS2, it'll get AS2 output

[23:01:44] <evanpro> What?

[23:01:56] <evanpro> -1

[23:02:02] <evanpro> Here's what I think we should do

[23:02:11] <strugee> I'm open to suggestions :)

[23:02:12] <evanpro> 1) Different endpoints for ActivityPub

[23:02:17] <evanpro> Completely separate

[23:02:27] <evanpro> 2) Store internally as AS1 for now

[23:03:00] <evanpro> 3) Leave all the AS1 endpoints live for a while

[23:03:14] <evanpro> We have people who've written lots of code to those endpoints

[23:03:19] <evanpro> And we have lots of data stored in AS1

[23:03:48] <evanpro> identi.ca alone has millions of activities

[23:03:54] <strugee> right

[23:03:58] <evanpro> Probably another million or so in other sites

[23:04:31] <strugee> so what happens when the ActivityPub endpoint receives AS2 that can't be downgraded?

[23:04:37] <evanpro> It'd probably be possible to just convert AS2 data to AS1 right at the surface, and then process it as AS1, and then re-convert to AS2 for output

[23:04:50] <evanpro> How could the AS2 not be downgraded?

[23:05:15] <strugee> off the top of my head, if it contains e.g. a Link, those can't be represented as AS1

[23:05:16] <evanpro> Oh, like, with unrecognized types?

[23:05:53] <strugee> that too

[23:06:13] <evanpro> So, two possibilities

[23:06:33] <pumabot> Issue #1199 "Meanwhile items without titles aren't terribly useful" larjona added label: "webui" - https://github.com/pump-io/pump.io/issues/1199

[23:06:33] <evanpro> One is to reject the submissions (this is what I think we should do)

[23:06:48] <evanpro> And second is to accept it and store it

[23:06:58] <evanpro> Either with some kind of marker like "this one is AS2"

[23:07:14] <evanpro> Or convert as much as possible and mark the unrecognized properties

[23:07:49] <strugee> I vote we reject stuff

[23:07:52] <evanpro> Me too

[23:08:19] <strugee> ok so here's what I've got, tell me if it's what you meant:

[23:08:54] <strugee> we add an ActivityPub endpoint which takes AS2 objects and downgrades them to AS1, which is used internally

[23:09:01] <strugee> if that can't be done activities are rejected

[23:10:06] <strugee> original API endpoints will continue to output AS1, ActivityPub endpoints will upgrade AS1 objects back to AS2

[23:10:31] <-- evanpro left (Read error: Connection reset by peer)

[23:11:44] <larjona> :(

[23:11:52] <strugee> indeed

[23:12:13] <strugee> I'm going to assume that's the plan, will email Evan to confirm

[23:13:32] <larjona> so, what do we do now?

[23:13:43] <strugee> I guess we finish up the meeting

[23:13:50] <strugee> too bad we didn't get to node adoption, sorry

[23:13:55] <strugee> I didn't think AS2 would take so long

[23:14:00] <larjona> No problem

[23:14:23] <larjona> It's late, let's end the meeting, we can follow up by email or whatever

[23:14:25] --> e_s_p joined #pump.io

[23:14:29] <larjona> ah!

[23:14:30] <e_s_p> argh

[23:14:34] <strugee> ah! cool

[23:14:35] <e_s_p> wifi

[23:14:45] <e_s_p> OK I have last question

[23:15:10] <e_s_p> larjona I need the link to the spreadsheet with all the pump server data

[23:15:17] <larjona> ah, just a sec

[23:15:59] <strugee> http://lacaja.larjona.net/shared/vSm84lM8lLw6WTX4k44QX_Ezx14RjIqug0xTvehwcKz

[23:16:09] <larjona> thanks strugee

[23:16:14] <strugee> 'course :)

[23:17:02] <larjona> Anything else?

[23:17:11] <larjona> do you want some pizza?

[23:17:29] <e_s_p> I'm about to leave to get some!

[23:17:35] <e_s_p> so hungry

[23:17:38] <larjona>

[23:17:38] <larjona> .....

[23:17:38] <larjona> .:.--|--.:.

[23:17:38] <larjona> .: .'\o | o /'. '.

[23:17:38] <larjona> // '. \ o| / o '.\

[23:17:38] <larjona> //'.o'. \ |o/ o.-'o\

[23:17:38] <larjona> || o '-.'.|/.-' o ||

[23:17:38] <larjona> ||--o--o-->|<o-----o-||

[23:17:38] <larjona> \ o _.-'/|'-._o o//

[23:17:38] <larjona> \.-' o/ |o\ o '-.//

[23:17:38] <larjona> '.'.o / o| \ o.'.'

[23:17:38] <larjona> `-:/.__|__o:-'

[23:17:38] <larjona> "--=--"

[23:17:38] <larjona>

[23:17:38] <larjona>

[23:17:51] <larjona> thanks for the meeting!

[23:17:52] <larjona> #############################################################

[23:17:52] <larjona> END LOG

[23:17:52] <larjona> #############################################################