KEP-0007 - Community forum #2011

castrojo · 2018-04-06T18:39:05Z

This is an initial proposal for establishing a community forum for Kubernetes.

bacongobbler · 2018-04-06T19:41:49Z

keps/sig-contributor-experience/0007-20180403-community-forum.md

+- We're losing too much information in the Slack ether, and most of it does not show up in search engines. 
+- Mailing lists remain mostly the domain of existing developers and require subscribing, whereas an open forum allows people to drive by and participate with minimal effort.
+  - There's an entire universe of users and developers that we could be reaching that didn't grow up on mailing lists and emacs. :D
+  - Specifically, hosting our lists on google groups has some issues:


tacking on a +1 here: google groups is also limited to strictly google accounts. I cannot use my @microsoft.com domain to subscribe to the mailing list, for example. However, I know that's being changed with a slow roll-out to groups.io.

Excellent point, I'll add this as an item when the KEP moves to the discussion step.

bacongobbler · 2018-04-06T19:52:09Z

keps/sig-contributor-experience/0007-20180403-community-forum.md

+
+### Risks and Mitigations
+
+- One more thing to check everyday(tm)


Yeah, this is the one thing I'm most concerned about. Org members are already feeling pretty strained with the number of support-oriented communication channels available today.

Would it be worthwhile to consider directing org members to chime in PRIMARILY at the forum, while keeping other forms of support-oriented communication like kubernetes-users, stack overflow and the mailing list left to the userbase? That would address your point above where "over time, as community interaction and knowledge base builds, people will end up with a better experience than in #kubernetes-users on slack and will naturally gravitate there". If users notice org members chiming into their questions in one place, they'll likely gravitate there naturally.

What do you think about that?

I think the mailing list integration will be the key feature here. If it can clearly replace the google groups then the decision will be easier. The initial goal of this KEP is for a prototype so we can test this feature and determine how viable it is.

I've tried (and failed) in the past to artificially segregate traffic on a medium and it doesn't really work. This whole section is the hardest part of the whole thing afaict. Joe has reminded me that basically "you can't add a new thing without killing an old thing", so I am hopeful that we can get some decent data points for viability from a prototype.

I will stress that discourse IS NOT a replacement for mailing lists and we should review their mailing list mode. If the primary usage of discourse is for users to get support and dev work still happens on mailing lists, I think that would be the best approach.

If discourse is not a replacement for mailing lists then we will need it to be a replacement for slack. One or the other will need to go in this case because we we have already hit communication overload. But this decision can come post test of discourse.

Open ended questions that we may need to answer:
If Discourse can't replace mailing lists, is it work migrating the very large and vibrant slack community?

That, to me, is a very high bar to meet.

This KEP specifically outlines that is does not replace. IMO it should be considered a POC for a period of time and re-evaluated. The purpose of a KEP is not to live with decisions forever, but to have a living document of ideas that evolves as the project evolves.

I'm an NAK on removing anything until there has been sufficient signal gathered to make an informed decision.

jberkus · 2018-04-13T17:33:34Z

keps/sig-contributor-experience/0007-20180403-community-forum.md

+- As part of a generic community portal, this gives people a place to go where they can discuss Kubernetes, and a sounding board for developers to make announcements of things happening around Kubernetes that can reach a wider audience.
+- We would be in charge of our own destiny, (aka. The Slack XMPP/IRC gateway removal-style concerns can be partially addressed)
+  - Software is 100% Open Source
+  - We'd have full access to our data, including the ability to export all of it.


Is it worth mentioning the ability to data mine community data here, in order to actively monitor community health beyond github?

I mention the datamining in another section, but I can add some more detail there.

jberkus · 2018-04-13T17:37:03Z

keps/sig-contributor-experience/0007-20180403-community-forum.md

+  - Link to important SIG announcements
+  - Any related user-facing announcements for our mentorship programs
+- Job board
+  - This might be difficult to do properly and keep classy, but leaving it here as a discussion point.


I'd be in favor of a moderated job board, and willing to help moderate it.

The main reason to moderate it is because otherwise we'll get bot-based spamming of the board.

We should add more bots and spam and thus more moderation hours to the drawbacks section @castrojo

Moderation is handled by a community trust system, what happens is we pick the 5 or so admins, and over time trusted users get more moderation powers. I don't expect that there will be extra moderation for people not engaged in the forum.

castrojo · 2018-04-13T18:06:55Z

Some feedback from Sarah Novotny during a call: Add a research section with some feedback from other OSS communities, ask them specifically how things are going, get some proper lessons learned added to the KEP itself.

parispittman · 2018-04-20T15:29:55Z

fyi for public view: the test period on this is three months and then Jorge/ContribEx will evaluate next steps (potentially decommissioning another service or platform to reduce the number we have).

/lgtm
/approve

cblecker · 2018-04-20T17:08:12Z

keps/sig-contributor-experience/0007-20180403-community-forum.md

+
+## Summary
+
+Kubernetes is large enough that we should take a more active role in growing our community. We need a place to call our own that can encompass users, contributors, meetups, and other groups in the community. Is need for something between email and real time chat that can fulfill this? The primary purpose of this KEP is to determine whether we can provide a better community forum experience and perhaps improve our mailing list workflow. 


nit: grammar -- "Is need for something.."

cblecker · 2018-04-20T17:12:30Z

keps/sig-contributor-experience/0007-20180403-community-forum.md

+  - The motivation of having searchable information that is owned by the Kubernetes community comes from voiced concerns about having so much of Kubernetes depend on Slack. 
+  - You are encouraged to propose a KEP for real-time communication if you would like to champion that, this KEP is not about Slack. 
+- This does not replace Stack Overflow or kubernetes-users for user support. 
+  - However inevitably users who prefer forums will undoubtedly use it for support.


We can talk about this later, but having a single "official" support stream would be preferred in my opinion. Discourse might be better for that, or we might want to redirect users back to SO, etc.

agree, generally prefer SO for support related questions.

cblecker · 2018-04-20T17:13:37Z

keps/sig-contributor-experience/0007-20180403-community-forum.md

+  - "Jill's neat K8s project on github" is too small to have it's own official k8s presence, but it could be a post on a forum. 
+- Events section for meetups and Kubecon
+- Sub boards for meetup groups
+- Sub boards for non-english speaking community members


This would be great, but we probably need to have language-specific moderators.

cblecker · 2018-04-20T17:16:44Z

keps/sig-contributor-experience/0007-20180403-community-forum.md

+
+### Risks and Mitigations
+
+- One more thing to check everyday(tm)


Open ended questions that we may need to answer:
If Discourse can't replace mailing lists, is it work migrating the very large and vibrant slack community?

That, to me, is a very high bar to meet.

cblecker · 2018-04-20T17:20:33Z

keps/sig-contributor-experience/0007-20180403-community-forum.md

+
+## Graduation Criteria
+
+There will be a feedback subforum where users can directly give us feedback on what they'd like to see. Metrics and site usage should determine if this will be viable in the long term. 


The key thing I see missing here is a proposed timetable for review. At what point do we sit down and decide:

This is a better solution than what we have, and we should deprecate (mailing lists|Slack|StackOverflow|etc).

This solution is not better than what we have, and we don't want to support Yet Another Tool. We should shut the project down.

We don't have enough information to draw a conclusion, so we need to extend our evaluation period.

cblecker · 2018-04-20T17:24:40Z

/hold
I gave a bit of feedback. I read the document as-is first before reading the comments, and @parispittman noted what the timeline is. I'd suggest it's valuable to have that evaluation timeline right in the document, as there's no indication in the doc of what that timeline is.

@parispittman @castrojo Thoughts?

parispittman · 2018-04-20T17:47:53Z

@cblecker thats exactly why I put it in the comment - didn't see it in the KEP but wanted it to be known that this is an evaluation period and not the implementation of it forever.

@castrojo - will you update it with that?

cblecker · 2018-04-20T18:07:05Z

/lgtm

@jbeda @jberkus @grodrigues3 @parispittman Any last thoughts before merging? (you're either listed as a reviewer or approver on this KEP)

timothysc

/lgtm

timothysc · 2018-04-20T19:08:09Z

keps/sig-contributor-experience/0007-20180403-community-forum.md

+
+### Risks and Mitigations
+
+- One more thing to check everyday(tm)


This KEP specifically outlines that is does not replace. IMO it should be considered a POC for a period of time and re-evaluated. The purpose of a KEP is not to live with decisions forever, but to have a living document of ideas that evolves as the project evolves.

I'm an NAK on removing anything until there has been sufficient signal gathered to make an informed decision.

jberkus · 2018-04-20T19:29:29Z

keps/sig-contributor-experience/0007-20180403-community-forum.md

+- Determine if this is a better solution than what we have, and figure out where this would fit in the ecosystem
+  - There is a strong desire that this would replace an existing support venue, SIG Contributor Experience will weigh the options.  
+- If this solution is not better than what we have, and we don't want to support yet another tool we we would shut the project down.
+- If we don't have enough information to draw a conclusion, we may decide to extend the evaluation period.


Having written policies for moderation, adding topics, etc. should be part of this graduation criteria somehow.

I was thinking since we have to have moderation policies for the mailing lists anyway (which we haven't written yet) that it would make sense to have one set of moderation policies across the project so we're consistent across the board. I'll make that more explicit next time I PR.

jberkus · 2018-04-20T19:35:37Z

@cblecker

On a third review, I feel strongly that having written moderation and management policies should be a graduation criterion.

jberkus · 2018-04-20T20:00:39Z

/lgtm

k8s-ci-robot · 2018-04-20T20:00:46Z

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: cblecker, jberkus, parispittman, timothysc

The full list of commands accepted by this bot can be found here.

The pull request process is described here

Needs approval from an approver in each of these files:

~~keps/OWNERS~~ [cblecker,timothysc]

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

grodrigues3 · 2018-04-20T21:13:29Z

looks good to me

cblecker · 2018-04-20T21:14:31Z

/hold cancel
Happy Friday, @castrojo!

KEP-0007 - Community forum

k8s-ci-robot requested review from cblecker and idvoretskyi April 6, 2018 18:39

Propose KEP-0007 - a Community forum for Kubernetes

7b6094e

bacongobbler reviewed Apr 6, 2018

View reviewed changes

Spelling fixes to pass bot, add more information on google groups

9d2f5a8

jberkus reviewed Apr 13, 2018

View reviewed changes

Add links to other OSS communities using Discourse

2ab4dc2

k8s-ci-robot assigned parispittman Apr 20, 2018

k8s-ci-robot added the lgtm "Looks good to me", indicates that a PR is ready to be merged. label Apr 20, 2018

cblecker reviewed Apr 20, 2018

View reviewed changes

k8s-ci-robot added the do-not-merge/hold Indicates that a PR should not merge because someone has issued a /hold command. label Apr 20, 2018

Be more explicit about the graduation criteria

140ce8c

k8s-ci-robot removed the lgtm "Looks good to me", indicates that a PR is ready to be merged. label Apr 20, 2018

k8s-ci-robot assigned cblecker Apr 20, 2018

k8s-ci-robot added lgtm "Looks good to me", indicates that a PR is ready to be merged. approved Indicates a PR has been approved by an approver from all required OWNERS files. labels Apr 20, 2018

timothysc approved these changes Apr 20, 2018

View reviewed changes

k8s-ci-robot assigned timothysc Apr 20, 2018

jberkus reviewed Apr 20, 2018

View reviewed changes

Add moderation policy requirement

9e4e30c

k8s-ci-robot removed the lgtm "Looks good to me", indicates that a PR is ready to be merged. label Apr 20, 2018

k8s-ci-robot assigned jberkus Apr 20, 2018

k8s-ci-robot added the lgtm "Looks good to me", indicates that a PR is ready to be merged. label Apr 20, 2018

k8s-ci-robot removed the do-not-merge/hold Indicates that a PR should not merge because someone has issued a /hold command. label Apr 20, 2018

k8s-ci-robot merged commit d09814f into kubernetes:master Apr 20, 2018

parispittman mentioned this pull request Aug 7, 2018

archiving k-users@ ml and move to discuss.k8s.io #2492

Closed

5 tasks

calebamiles pushed a commit to calebamiles/community that referenced this pull request Sep 5, 2018

Merge pull request kubernetes#2011 from castrojo/community-forum

67f86eb

KEP-0007 - Community forum

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

KEP-0007 - Community forum #2011

KEP-0007 - Community forum #2011

castrojo commented Apr 6, 2018

bacongobbler Apr 6, 2018

castrojo Apr 9, 2018

bacongobbler Apr 6, 2018 •

edited

Loading

castrojo Apr 9, 2018

caniszczyk Apr 9, 2018

parispittman Apr 13, 2018 •

edited

Loading

cblecker Apr 20, 2018

timothysc Apr 20, 2018

jberkus Apr 13, 2018

castrojo Apr 13, 2018

jberkus Apr 13, 2018

jberkus Apr 13, 2018

parispittman Apr 13, 2018

castrojo Apr 20, 2018

castrojo commented Apr 13, 2018

parispittman commented Apr 20, 2018

cblecker Apr 20, 2018

cblecker Apr 20, 2018

grodrigues3 Apr 20, 2018

cblecker Apr 20, 2018

cblecker Apr 20, 2018

cblecker Apr 20, 2018

cblecker commented Apr 20, 2018

parispittman commented Apr 20, 2018

cblecker commented Apr 20, 2018

timothysc left a comment

timothysc Apr 20, 2018

jberkus Apr 20, 2018

castrojo Apr 20, 2018

jberkus commented Apr 20, 2018

jberkus commented Apr 20, 2018

k8s-ci-robot commented Apr 20, 2018

grodrigues3 commented Apr 20, 2018

cblecker commented Apr 20, 2018


		### Risks and Mitigations

		- One more thing to check everyday(tm)


		## Summary

		Kubernetes is large enough that we should take a more active role in growing our community. We need a place to call our own that can encompass users, contributors, meetups, and other groups in the community. Is need for something between email and real time chat that can fulfill this? The primary purpose of this KEP is to determine whether we can provide a better community forum experience and perhaps improve our mailing list workflow.


		## Graduation Criteria

		There will be a feedback subforum where users can directly give us feedback on what they'd like to see. Metrics and site usage should determine if this will be viable in the long term.

KEP-0007 - Community forum #2011

KEP-0007 - Community forum #2011

Conversation

castrojo commented Apr 6, 2018

Choose a reason for hiding this comment

Choose a reason for hiding this comment

bacongobbler Apr 6, 2018 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

parispittman Apr 13, 2018 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

castrojo commented Apr 13, 2018

parispittman commented Apr 20, 2018

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

cblecker commented Apr 20, 2018

parispittman commented Apr 20, 2018

cblecker commented Apr 20, 2018

timothysc left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

jberkus commented Apr 20, 2018

jberkus commented Apr 20, 2018

k8s-ci-robot commented Apr 20, 2018

grodrigues3 commented Apr 20, 2018

cblecker commented Apr 20, 2018

bacongobbler Apr 6, 2018 •

edited

Loading

parispittman Apr 13, 2018 •

edited

Loading