TOC-Members,

 

This email is a follow up of a conversation I started with dims.

 

I am asking you to rethink the current Sandbox process – mostly regarding speed. The idea of having a way to collaborate across organisations under a neutral foundation is key ideal of the CNCF. However, this is proving to get harder; mostly regarding the speed of acceptance.

 

Let me share an example. We have put together an industry consortium to define a common, vendor-neutral standard for feature flagging (https://openfeature.dev ) and brought together a consortium spanning key industry players and end users (see interested parties: https://github.com/open-feature/community/blob/main/interested-parties.md ). We submitted for sandbox early this year and given the current backlog and progress on evaluating projects it is very likely that the project will get accepted until late this year.

 

Pulling these activities off, getting buy in form key stakeholder and driving momentum to move them forward is a major effort. A key part of proces is being able to operate under a neutral entry like the CNCF. If this process is taking a very long time, it has negative impact on these initiatives. People might lose interest, change roles or jobs, etc.

 

The sandbox process is an essential part of evolving the cloud native landscape, but it is broken and needs to evolve.

 

I am proposing to make this a continuous process; maybe involving TAGs (again), who can support handling to workload and defining a set of criteria allowing project to prepare for being accepted quickly. Just some ideas on criteria:

• Obviously, the project being cloud native
• Clear goals and roadmap
• A community engagement plan.
• A team/consortium that can the delivery on the project’s goals.

 

Immediate steps then should be to get the current backlog down and define a “service level” by when project should be able to expect a response.

 

I am willing to support on improving the process, if needed/wanted.

 

 

This email may contain confidential information. If it appears this message was sent to you by mistake, please let us know of the error. In this case, we also ask that you do not further forward the content and delete it. Thank you for your cooperation and understanding. Dynatrace Austria GmbH (registration number FN 91482h) is a company registered in Linz whose registered office is at 4020 Linz, Austria, Am Fünfundzwanziger Turm 20.

Just a thought, could the Sandbox process be amended to be run by a designed to be nimble, yet to be created committee/council/board?

Chris Short

He/Him/His

Sr. Developer Advocate, AWS Kubernetes (GitOps)

TZ=America/Detroit

On Apr 30, 2022, at 12:18, Davanum Srinivas <davanum@...> wrote:



CAUTION: This email originated from outside of the organization. Do not click links or open attachments unless you can confirm the sender and know the content is safe.

Alois. 

Big thanks for bringing this up!

Some observations from me:

- there are 21 applications currently in the queue, a bunch of them are resubmissions where the TOC has a set of questions and they came back with answers.

- The process needs to be consistent (irrespective of when a project was submitted or who is evaluating it) to prevent angst/worry on the part of the submitters.

- TOC's are not particularly healthy as they are understaffed, we do push some of the submissions to talk to either TAGs or k8s SIGs for example and come back with written responses and endorsements for example.

- Getting time on calendars for everyone on the TOC is challenge, so we have/will run into quorum issues trying to schedule additional calls to get through the backlog

- "People might lose interest, change roles or jobs, etc." << I get it, but i'd rather not accept something that is ephemeral and doesn't really have folks who can drive things for the longer term, Sorry.

- On a good day, TOC is able to handle about 10-12 submissions. Trying to do more will be just rubber stamping instead of actually looking through, reading and watching the stuff the submitters have requested.

I do get the need for speed and I agree that we need to do better. So let's have this conversation and see how we can proceed next. 

Alois, I am happy to add this to the TOC agenda and walk through the issues and work through possible solutions, 

thanks,

Dims

On Sat, Apr 30, 2022 at 11:25 AM Reitbauer, Alois via lists.cncf.io <alois.reitbauer=dynatrace.com@...> wrote:

TOC-Members,

 

This email is a follow up of a conversation I started with dims.

 

I am asking you to rethink the current Sandbox process – mostly regarding speed. The idea of having a way to collaborate across organisations under a neutral foundation is key ideal of the CNCF. However, this is proving to get harder; mostly regarding the speed of acceptance.

 

Let me share an example. We have put together an industry consortium to define a common, vendor-neutral standard for feature flagging (https://openfeature.dev ) and brought together a consortium spanning key industry players and end users (see interested parties: https://github.com/open-feature/community/blob/main/interested-parties.md ). We submitted for sandbox early this year and given the current backlog and progress on evaluating projects it is very likely that the project will get accepted until late this year.

 

Pulling these activities off, getting buy in form key stakeholder and driving momentum to move them forward is a major effort. A key part of proces is being able to operate under a neutral entry like the CNCF. If this process is taking a very long time, it has negative impact on these initiatives. People might lose interest, change roles or jobs, etc.

 

The sandbox process is an essential part of evolving the cloud native landscape, but it is broken and needs to evolve.

 

I am proposing to make this a continuous process; maybe involving TAGs (again), who can support handling to workload and defining a set of criteria allowing project to prepare for being accepted quickly. Just some ideas on criteria:

• Obviously, the project being cloud native
• Clear goals and roadmap
• A community engagement plan.
• A team/consortium that can the delivery on the project’s goals.

 

Immediate steps then should be to get the current backlog down and define a “service level” by when project should be able to expect a response.

 

I am willing to support on improving the process, if needed/wanted.

 

 

This email may contain confidential information. If it appears this message was sent to you by mistake, please let us know of the error. In this case, we also ask that you do not further forward the content and delete it. Thank you for your cooperation and understanding. Dynatrace Austria GmbH (registration number FN 91482h) is a company registered in Linz whose registered office is at 4020 Linz, Austria, Am Fünfundzwanziger Turm 20.

--

Davanum Srinivas :: https://twitter.com/dims

On 4/30/22 09:18, Davanum Srinivas wrote:
> - there are 21 applications currently in the queue, a bunch of them are
> resubmissions where the TOC has a set of questions and they came back
> with answers.

FWIW, this part of the process could be made considerably more efficient
for both the projects and the TOC as well.

Right now, when sandbox projects are sent out and come back with
answers, there's no paper trail for the questions and the answers.  This
forces TOC members to re-evaluate the project from scratch.

My suggestion for a simple process that would solve this.  If a project
needs to answer questions or get inspected by a TAG, what happens is:

1. The TOC writes the questions in an issue in the TOC repo
2. The project or the TAG (depending) answer those questions in the TOC
repo.
3. Evaluation is resumed whenever the questions are answered.

This would spare TOC members from effectively starting over each time a
project re-applies.


--
-- Josh Berkus
    Kubernetes Community Architect
    OSPO, OCTO

[Edited Message Follows]

+1 to pushing more to TAGs. Perhaps each proposed project can be assigned to a TAG and a member of the TAG can lead a technical review and guide the project's leads on criteria for acceptance. That could help TOC reviews go more smoothly, make them more likely to succeed, and ensure projects and contributors don't get lost or feel unsupported along the journey.

It could also give first-time CNCF/TAG contributors an idea of where to start - they could pick an open project for the TAG and review and present it to the group.

As a start in this direction a couple months ago I created a label in TAG App Delivery to track project review requests from TOC: https://github.com/cncf/tag-app-delivery/issues?q=label%3Atoc-review

I, for one, would love to see the sandbox process be faster and improve.

With regard to moving more work to the TAGs, two things come to mind.

First, when TAGs did more in the past they were inconsistent across each other and added their own criteria. This was a problem I don't want to see again. For example, I remember when one project lead was proposing a project for sandbox. He was frustrated because his project was criticized for not meeting a graduation criteria and for a criteria that was of the TAGs own making. I don't think we want this to happen again.

Second, sandbox projects don't get or need an in depth technical analysis. That shows up for incubation. I'm wondering, what would a TAG do here that wouldn't be repeated by the TOC when they go to look at it?

Having been through sandbox reviews twice now and having given advice to some projects that wanted to go for sandbox I've learned a few areas that could use some improvement...

1. I've answered a lot of questions about things not in the docs. Things that provide context to the CNCF, what sandbox is, what I think the TOC is looking for, and how to communicate well to the TOC. I think this could be better documented.
   
2. TAGs have a unique intersection where they have experts in an area and they work with the projects. I (and the rest of the TOC) don't scale on advising projects for sandbox. The TAGs may be able to do that. While I wouldn't require it, it could be useful for those who want to submit a sandbox project to present to the appropriate TAG and get guidance from them. For those who need it, getting some mentoring from a TAG could be useful.
   

These are just my initial thoughts. Happy to hear agreement, disagreement, or things built upon this.

Cheers,

Matt

On Sun, May 1, 2022, at 11:09 AM, Matt Young wrote:

We have been thinking about this in TAG Observability as well, and have work in flight that’s related:

* Form Program: Annual Sandbox Review [1]

    * Create summary slides [2]

[1] https://github.com/cncf/tag-observability/issues/80

[2] 

https://github.com/cncf/tag-observability/issues/83

Will have details Tuesday as part of TAG update.

Matt

On Sun, May 1, 2022 at 9:41 AM Josh Gavant <joshgavant@...> wrote:

[Edited Message Follows]

+1 to pushing more to TAGs. Perhaps each proposed project can be assigned to a TAG and a member of the TAG can lead a technical review and guide the project's leads on criteria for acceptance. That could help TOC reviews go more smoothly, make them more likely to succeed, and ensure projects and contributors don't get lost or feel unsupported along the journey.

It could also give first-time CNCF/TAG contributors an idea of where to start - they could pick an open project for the TAG and review and present it to the group.

As a start in this direction a couple months ago I created a label in TAG App Delivery to track project review requests from TOC: https://github.com/cncf/tag-app-delivery/issues?q=label%3Atoc-review

A little history: the current process was supposed to be super-lightweight, to reflect the very, very low bar for Sandbox projects - essentially, is it cloud native. I don’t remember the exact number but I’m pretty sure we got through a lot more than 12 applications in the first meeting.

Maybe it’s worth the TOC revisiting what that low bar really should be so that it’s easier and quicker to assess?  Here’s a suggestion that would make it super lightweight but I think still be in line with the CNCF mission. 

One of our reasons to exist is to enable multiple organisations to have a neutral place to collaborate, even if the project is little more than at the paper napkin stage. Based on this, we could define the bar for Sandbox as: a project needs to have support from minimum two CNCF member organisations who consider themselves stakeholders in the project. That could mean they’re involved in building it, or interested in using it. The onus is on the project to find those stakeholders before applying. The TOC’s approval would simply be a check that they agree that it’s a cloud native project and that they don’t have any other objection to it being included 

On Sun, 1 May 2022 at 19:29, alexis richardson <alexis@...> wrote:

+1 

On Sun, 1 May 2022, 19:27 Matt Farina, <matt@...> wrote:

I, for one, would love to see the sandbox process be faster and improve.

With regard to moving more work to the TAGs, two things come to mind.

First, when TAGs did more in the past they were inconsistent across each other and added their own criteria. This was a problem I don't want to see again. For example, I remember when one project lead was proposing a project for sandbox. He was frustrated because his project was criticized for not meeting a graduation criteria and for a criteria that was of the TAGs own making. I don't think we want this to happen again.

Second, sandbox projects don't get or need an in depth technical analysis. That shows up for incubation. I'm wondering, what would a TAG do here that wouldn't be repeated by the TOC when they go to look at it?

Having been through sandbox reviews twice now and having given advice to some projects that wanted to go for sandbox I've learned a few areas that could use some improvement...

1. I've answered a lot of questions about things not in the docs. Things that provide context to the CNCF, what sandbox is, what I think the TOC is looking for, and how to communicate well to the TOC. I think this could be better documented.
   
2. TAGs have a unique intersection where they have experts in an area and they work with the projects. I (and the rest of the TOC) don't scale on advising projects for sandbox. The TAGs may be able to do that. While I wouldn't require it, it could be useful for those who want to submit a sandbox project to present to the appropriate TAG and get guidance from them. For those who need it, getting some mentoring from a TAG could be useful.
   

These are just my initial thoughts. Happy to hear agreement, disagreement, or things built upon this.

Cheers,

Matt

On Sun, May 1, 2022, at 11:09 AM, Matt Young wrote:

We have been thinking about this in TAG Observability as well, and have work in flight that’s related:

* Form Program: Annual Sandbox Review [1]

    * Create summary slides [2]

[1] https://github.com/cncf/tag-observability/issues/80

[2] 

https://github.com/cncf/tag-observability/issues/83

Will have details Tuesday as part of TAG update.

Matt

On Sun, May 1, 2022 at 9:41 AM Josh Gavant <joshgavant@...> wrote:

[Edited Message Follows]

+1 to pushing more to TAGs. Perhaps each proposed project can be assigned to a TAG and a member of the TAG can lead a technical review and guide the project's leads on criteria for acceptance. That could help TOC reviews go more smoothly, make them more likely to succeed, and ensure projects and contributors don't get lost or feel unsupported along the journey.

It could also give first-time CNCF/TAG contributors an idea of where to start - they could pick an open project for the TAG and review and present it to the group.

As a start in this direction a couple months ago I created a label in TAG App Delivery to track project review requests from TOC: https://github.com/cncf/tag-app-delivery/issues?q=label%3Atoc-review

I should add, that’s not intended as a criticism -  the number of very early stage applications from individuals and single vendors has increased, which over time opened up the question for the TOC of whether it’s really right to commit CNCF resources for these projects. 

Those discussions naturally move us away from the original intention that the process should involve very little assessment or subjective judgement (e.g. the intention was to avoid a complicated definition of what is “mature enough” for sandbox) 

On Mon, 2 May 2022 at 18:48, Liz Rice <liz@...> wrote:

A little history: the current process was supposed to be super-lightweight, to reflect the very, very low bar for Sandbox projects - essentially, is it cloud native. I don’t remember the exact number but I’m pretty sure we got through a lot more than 12 applications in the first meeting.

Maybe it’s worth the TOC revisiting what that low bar really should be so that it’s easier and quicker to assess?  Here’s a suggestion that would make it super lightweight but I think still be in line with the CNCF mission. 

One of our reasons to exist is to enable multiple organisations to have a neutral place to collaborate, even if the project is little more than at the paper napkin stage. Based on this, we could define the bar for Sandbox as: a project needs to have support from minimum two CNCF member organisations who consider themselves stakeholders in the project. That could mean they’re involved in building it, or interested in using it. The onus is on the project to find those stakeholders before applying. The TOC’s approval would simply be a check that they agree that it’s a cloud native project and that they don’t have any other objection to it being included 

On Sun, 1 May 2022 at 19:29, alexis richardson <alexis@...> wrote:

+1 

On Sun, 1 May 2022, 19:27 Matt Farina, <matt@...> wrote:

I, for one, would love to see the sandbox process be faster and improve.

With regard to moving more work to the TAGs, two things come to mind.

First, when TAGs did more in the past they were inconsistent across each other and added their own criteria. This was a problem I don't want to see again. For example, I remember when one project lead was proposing a project for sandbox. He was frustrated because his project was criticized for not meeting a graduation criteria and for a criteria that was of the TAGs own making. I don't think we want this to happen again.

Second, sandbox projects don't get or need an in depth technical analysis. That shows up for incubation. I'm wondering, what would a TAG do here that wouldn't be repeated by the TOC when they go to look at it?

Having been through sandbox reviews twice now and having given advice to some projects that wanted to go for sandbox I've learned a few areas that could use some improvement...

1. I've answered a lot of questions about things not in the docs. Things that provide context to the CNCF, what sandbox is, what I think the TOC is looking for, and how to communicate well to the TOC. I think this could be better documented.
   
2. TAGs have a unique intersection where they have experts in an area and they work with the projects. I (and the rest of the TOC) don't scale on advising projects for sandbox. The TAGs may be able to do that. While I wouldn't require it, it could be useful for those who want to submit a sandbox project to present to the appropriate TAG and get guidance from them. For those who need it, getting some mentoring from a TAG could be useful.
   

These are just my initial thoughts. Happy to hear agreement, disagreement, or things built upon this.

Cheers,

Matt

On Sun, May 1, 2022, at 11:09 AM, Matt Young wrote:

We have been thinking about this in TAG Observability as well, and have work in flight that’s related:

* Form Program: Annual Sandbox Review [1]

    * Create summary slides [2]

[1] https://github.com/cncf/tag-observability/issues/80

[2] 

https://github.com/cncf/tag-observability/issues/83

Will have details Tuesday as part of TAG update.

Matt

On Sun, May 1, 2022 at 9:41 AM Josh Gavant <joshgavant@...> wrote:

[Edited Message Follows]

+1 to pushing more to TAGs. Perhaps each proposed project can be assigned to a TAG and a member of the TAG can lead a technical review and guide the project's leads on criteria for acceptance. That could help TOC reviews go more smoothly, make them more likely to succeed, and ensure projects and contributors don't get lost or feel unsupported along the journey.

It could also give first-time CNCF/TAG contributors an idea of where to start - they could pick an open project for the TAG and review and present it to the group.

As a start in this direction a couple months ago I created a label in TAG App Delivery to track project review requests from TOC: https://github.com/cncf/tag-app-delivery/issues?q=label%3Atoc-review

Replying top-level as my thoughts jump across the thread.


I didn't run the numbers, yet I believe that the pace of submissions
has picked up. That alone can increase backlog.

We tried SIGs (now TAGs) doing due diligence for projects. The level
of scrutiny, and the closeness to the guidance material available, was
different across TAGs. In effect, this meant inconsistent processes
which is arguably unfair. And in cases of disagreements, TOC is pulled
in automatically anyway.
A clear delegation from TOC might be possible, yet project advancement
is one of the main tasks of TOC and arguably what votees expect TOC to
do. In any case, it does change any of the underlying desires.

What TAGs could provide is an initial proving ground, though: Projects
could give a presentation and go through questions and feedback in a
more limited scope, allowing them to polish their submittal.


While I know that the current sandbox process is designed to be very
low barrier, I am still not convinced that this is an obviously
desirable design goal. It is true that a neutral playing field is good
and helps some projects grow. It is also true that "CNCF project"
holds immense marketing value and many efforts are ephemeral, in
particular if largely driven by perf & marketing.
Back when sandbox criteria were relaxed, I was of the opinion that
they should remain more stringent. I have come to wonder if four
levels wouldn't be more appropriate: An initial runway on which
projects can be put; but also pruned more aggressively if they do not
show growth/adoption/the usual. E.g. once submitted they have three?
six? twelve? months to show certain progress or are removed outright.
Medium term, this might also allow for a smaller jump towards
Incubating, which is currently significant.


Orthogonally, I believe we can manage expectations better. One
possible approach would be to create dashboards and reports of the
underlying data to help manage expectations and keep ourselves honest.
What are the average and median times a project takes from stage X to
stage Y? How has this changed over time?
Another would be to rework the process & documentation; e.g.
Incubation had distinct requirement docs which TAGs copied together
and deduplicated back during the DD trials.



Having seen things from both sides now, and since CNCF started, I can
understand both the frustrations about some timelines better and also
understand how a few dedicated people are trying to do their best with
the time they have. On all sides.


Best,
Richard

Replying top-level as my thoughts jump across the thread.


I didn't run the numbers, yet I believe that the pace of submissions
has picked up. That alone can increase backlog.

We tried SIGs (now TAGs) doing due diligence for projects. The level
of scrutiny, and the closeness to the guidance material available, was
different across TAGs. In effect, this meant inconsistent processes
which is arguably unfair. And in cases of disagreements, TOC is pulled
in automatically anyway.
A clear delegation from TOC might be possible, yet project advancement
is one of the main tasks of TOC and arguably what votees expect TOC to
do. In any case, it does change any of the underlying desires.

What TAGs could provide is an initial proving ground, though: Projects
could give a presentation and go through questions and feedback in a
more limited scope, allowing them to polish their submittal.


While I know that the current sandbox process is designed to be very
low barrier, I am still not convinced that this is an obviously
desirable design goal. It is true that a neutral playing field is good
and helps some projects grow. It is also true that "CNCF project"
holds immense marketing value and many efforts are ephemeral, in
particular if largely driven by perf & marketing.
Back when sandbox criteria were relaxed, I was of the opinion that
they should remain more stringent. I have come to wonder if four
levels wouldn't be more appropriate: An initial runway on which
projects can be put; but also pruned more aggressively if they do not
show growth/adoption/the usual. E.g. once submitted they have three?
six? twelve? months to show certain progress or are removed outright.
Medium term, this might also allow for a smaller jump towards
Incubating, which is currently significant.


Orthogonally, I believe we can manage expectations better. One
possible approach would be to create dashboards and reports of the
underlying data to help manage expectations and keep ourselves honest.
What are the average and median times a project takes from stage X to
stage Y? How has this changed over time?
Another would be to rework the process & documentation; e.g.
Incubation had distinct requirement docs which TAGs copied together
and deduplicated back during the DD trials.



Having seen things from both sides now, and since CNCF started, I can
understand both the frustrations about some timelines better and also
understand how a few dedicated people are trying to do their best with
the time they have. On all sides.


Best,
Richard