Hi Saad
Thanks for all your input and pertinent questions. Apologies for taking so long to respond.
I think this would be better discussed verbally in person (happy to do that in a public place). Otherwise you might end up with partial and inadequate email responses. That and the round-trip times when people are busy with other things make for a very drawn-out conversation. Lets do that at the next Storage WG meeting. If you want to chat before then, feel free to send out a zoom invite to the group.
I'd love to chat about this at the next CNCF Storage WG meeting, but I'm happy keeping the the conversation going over email until then as well.
In very brief summary in the mean time:
The CNCF is not in the business of picking winners. Any project can join the CNCF. Sandbox in particular has intentionally very low barriers to entry. "Looks OK to 2 TOC members" is essentially it, provided that the project falls within the general mandate of the CNCF. Incubation and Graduation have higher, explicit, and fairly specific and objective requirements (activity, adoption etc). If a project meets the requirements, it's in, modulo any specific objections by the TOC. Here is an example of a specific objection by the TOC:
https://github.com/cncf/toc/pull/168#pullrequestreview-171778524
Hope that helps.
Q
As much as I'd like to believe that the CNCF Sandbox is not a "real endorsement" by the CNCF and therefore it's ok to have a very low bar for entry, the fact is joining the CNCF sandbox has real benefits: the CNCF Sandbox Guidelines state the purpose of the CNCF Sandbox is to 1) "Encourage public visibility", 2) "Facilitate alignment with existing projects", 3) "Nurture projects", and 4) "Remove possible legal and governance obstacles to adoption and contribution" (as mentioned in the other thread). These are tangible benefits of being part of the CNCF, even at the Sandbox level.
Therefore, we should be more discriminating about the projects we choose to accept in to the CNCF. We have to understand the benefits to users of the CNCF ecosystem.
Extending the Sandbox charter to add a formal requirement for all external projects to first get a CNCF SIG to sponsor them seems like a great way to do this; given that the purpose of these SIGs per charter is already to "Strengthen the project ecosystem", "Identify gaps in the CNCF project portfolio", and "Reduce some project workload on TOC". My 2 cents, while it is of course critical to understand scope and charter, is deciding who does what that important this early on ?
A SIG (like storage) can demonstrate capability by formally recommending the next candidate project, which paves the way for a more formal process by example. In other words, ask for forgiveness, not permission :)
I like the attitude! But given how big of a step the formation of CNCF SIGs is, and the concurrent conversation about inducting OpenEBS before Kubecon EU, we should proceed very intentionally.
Sorry, I wasn't involved in those discussions.
Based on the published SIG charter, the purpose of these SIGs is to "Scale contributions by the CNCF technical and user community" with goals including "Strengthen the project ecosystem", "Identify gaps in the CNCF project portfolio", and "Reduce some project workload on TOC".
To that end, extending the charter to add a formal requirement for all external project to first get a CNCF SIG to sponsor them seems reasonable and in line? Doing so would "Strengthen the project ecosystem", "Identify gaps in the CNCF project portfolio", and "Reduce some project workload on TOC". In addition, as I mentioned before, it would give these new CNCF SIGs clear long-term ownership of something concrete, and help ensure that the CNCF decision making process is open, transparent, and inclusive.
It sounds like you're firmly against that, and if so, I'd like to understand why? What am I missing?
And I apologize if I'm coming across as combative. My intention is to ensure we're doing the right things for our users and building healthy communities. So please bear with me as I try to make sense of this.
On Thu, Apr 11, 2019 at 5:03 PM Alexis Richardson <alexis@...> wrote:
The purpose is documented in the SIG charter that we voted on recently.
On Thu, Apr 11, 2019 at 5:01 PM Saad Ali <saadali@...> wrote:
>
> Sure, SIGs can do what they want. And the final decision is still up to TOC.
>
> But would you support adding a formal requirement for all external project to first get a CNCF SIG to sponsor them? If not, why not, and what do you see as the purpose of these new CNCF SIGs?
>
> On Thu, Apr 11, 2019 at 4:44 PM Alexis Richardson <alexis@...> wrote:
>>
>> The SIGs can do whatever they like! But the final decision is not for
>> delegation.
>>
>> On Thu, Apr 11, 2019 at 4:32 PM Saad Ali <saadali@...> wrote:
>> >
>> > Thanks Alexis.
>> >
>> > Deciding which projects join the CNCF is one of the most important decisions the CNCF TOC is responsible for. And as far as I can tell, the only major requirement for joining the CNCF sandbox currently is support of "2 TOC sponsors".
>> >
>> > My understanding is that these CNCF SIGs are being formed, at least in part, to help the TOC with that decision making. But from everything I've read, it doesn't seem like these SIGs will have any real ownership of anything.
>> >
>> > So I propose that that in order to be inducted in to the CNCF (at any level, including Sandbox), an external project must get a CNCF SIG to sponsor it first? Each SIG would define the process for project consideration, establish criteria for the categories of project to be considered, etc.
>> >
>> > This would give these new CNCF SIGs clear long-term ownership of something concrete, and we would ensure that the CNCF decision making process is open, transparent, and inclusive.
>> >
>> > Of course, the TOC would still retain the right to veto the SIG decision, but it would do so publicly and provide justification for doing so.
>> >
>> > Thoughts?
>> >
>> > On Wed, Apr 10, 2019 at 5:45 PM Alexis Richardson <alexis@...> wrote:
>> >>
>> >> Saad
>> >>
>> >> Unlike K8s TOC & SIGs, the CNCF TOC will own final decisions on
>> >> project acceptance.
>> >>
>> >> alexis
>> >>
>> >> On Wed, Apr 10, 2019 at 5:37 PM via Lists.Cncf.Io
>> >> <saadali=google.com@...> wrote:
>> >> >
>> >> > Thanks Quinton and Alex! This is very helpful.
>> >> >
>> >> > Based on that, it seems to me like the responsibilities for this SIG are essentially to collect and disseminate storage related information in a consumable manner to 1) the CNCF TOC, 2) CNCF end users, and 3) CNCF projects (at all stages from pre-incubation through to graduation)?
>> >> >
>> >> > If so, then the concrete work this group will be responsible for is conducting research (surveys, etc.) and the output will be documents summarizing the results of the research for various audiences?
>> >> >
>> >> > The Storage Whitepaper produced by this group is out. User survey is in progress. What kind of work concretely do you see going forward?
>> >> >
>> >> > In Kubernetes the TOC tries to avoid making unilateral decisions as much as possible. Instead the Kubernetes TOC empowers the SIGs to make decisions.
>> >> >
>> >> > Right now the CNCF TOC is considering incubating a new storage project. Is the recommendation for accepting or rejecting that proposal be in the purview of this proposed SIG? If so, will the CNCF TOC sponsors have to get the approval of a SIG for any sponsored projects?
>> >> >
>> >> > On Wed, Apr 10, 2019 at 2:01 PM Alex Chircop <alex.chircop@...> wrote:
>> >> >>
>> >> >> Hi Saad,
>> >> >>
>> >> >>
>> >> >>
>> >> >> Thanks for throwing your hat into the ring ! :-)
>> >> >>
>> >> >>
>> >> >>
>> >> >> (apologies for the long email, but the info will hopefully benefit the rest of the mailing list too)
>> >> >>
>> >> >>
>> >> >>
>> >> >> The CNCF SIGs are being setup to help the TOC as the CNCF continues to scale with the growing list of projects and members.
>> >> >>
>> >> >> All the detail for the proposal/formation of the CNCF SIGs is available here: https://github.com/cncf/toc/blob/master/sigs/cncf-sigs.md
>> >> >>
>> >> >>
>> >> >>
>> >> >> In summary, the general objectives are:
>> >> >>
>> >> >> Strengthen the project ecosystem to meet the needs of end users and project contributors.
>> >> >> Identify gaps in the CNCF project portfolio. Find and attract projects to fill these gaps.
>> >> >> Educate and inform users with unbiased, effective, and practically useful information.
>> >> >> Focus attention & resources on helping foster project maturity, systematically across CNCF projects.
>> >> >> Clarify relationship between projects, CNCF project staff, and community volunteers.
>> >> >> Engage more communities and create an on-ramp to effective TOC contribution & recognition.
>> >> >> Reduce some project workload on TOC while retaining executive control & tonal integrity with this elected body.
>> >> >> Avoid creating a platform for politics between vendors.
>> >> >>
>> >> >>
>> >> >>
>> >> >> and the specific responsibilities of the SIG include:
>> >> >>
>> >> >>
>> >> >>
>> >> >> Project Handling:
>> >> >>
>> >> >> Understand and document a high level roadmap of projects within this space, including CNCF and non-CNCF projects. Identify gaps in project landscape.
>> >> >> For projects that fall within the CNCF, perform health checks.
>> >> >> Perform discovery of and outreach to candidate projects
>> >> >> Help candidate projects prepare for presentation to the TOC
>> >> >> Every CNCF project will be assigned to one suitable SIG by the TOC.
>> >> >>
>> >> >> End User Education (Outbound Communication)
>> >> >>
>> >> >> Provide up-to-date, high quality, unbiased and easy-to-consume material to help end users to understand and effectively adopt cloud-native technologies and practises within the SIG’s area, for example:
>> >> >>
>> >> >> White papers, presentations, videos, or other forms of training clarifying terminology, comparisons of different approaches, available projects or products, common or recommended practises, trends, illustrative successes and failures, etc.
>> >> >> As far as possible, information should be based on research and fact gathering, rather than pure marketing or speculation.
>> >> >>
>> >> >> End User Input Gathering (Inbound Communication)
>> >> >>
>> >> >> Gather useful end user input and feedback regarding expectations, pain points, primary use cases etc.
>> >> >> Compile this into easily consumable reports and/or presentations to assist projects with feature design, prioritization, UX etc.
>> >> >>
>> >> >> Community Enablement
>> >> >>
>> >> >> SIGs are open organizations with meetings, meeting agendas and notes, mailing lists, and other communications in the open
>> >> >> The mailing list, SIG meeting calendar, and other communication documents of the SIG will be openly published and maintained
>> >> >>
>> >> >> As Trusted Expert Advisors to the TOC
>> >> >>
>> >> >> Perform technical due diligence on new and graduating projects, and advise TOC on findings.
>> >> >> Be involved with, or periodically check in with projects in their area, and advise TOC on health, status and proposed actions (if any) as necessary or on request.
>> >> >>
>> >> >> See Example Responsibilities of a CNCF SIG.
>> >> >>
>> >> >>
>> >> >>
>> >> >> The first set of SIGs as per the proposal and the TOC members who will act as liaison with the SIG are:
>> >> >>
>> >> >> (from: https://docs.google.com/presentation/d/1BUmTO5PFt7NZ9jVMMR3r1W7k8NANltNNJJqFCZdbS0I/edit#slide=id.g4e24cc378e_1_39)
>> >> >>
>> >> >> Matt: Traffic (networking, service discovery, load balancing, service mesh, RPC, pubsub, etc)
>> >> >>
>> >> >> Envoy, Linkerd, NATS, gRPC, CoreDNS, CNI
>> >> >>
>> >> >> Jeff: Observability (monitoring, logging, tracing, profiling, etc.)
>> >> >>
>> >> >> Prometheus, OpenTracing, Fluentd, Jaeger, Cortex, OpenMetrics,
>> >> >>
>> >> >> Liz + Joe: Security/Governance (auth, authorization, auditing, policy enforcement, compliance, GDPR, cost management, etc)
>> >> >>
>> >> >> SPIFFE, SPIRE, Open Policy Agent, Notary, TUF, Falco,
>> >> >>
>> >> >> Michelle + Alexis: App Dev, Ops & Testing (PaaS, Serverless, Operators, CI/CD, Conformance, Chaos Eng, Scalability and Reliability measurement etc.)
>> >> >>
>> >> >> Helm, CloudEvents, Telepresence, Buildpacks, (CNCF CI)
>> >> >>
>> >> >> Brendan + Brian: Core and Applied Architectures (orchestration, scheduling, container runtimes, sandboxing technologies, packaging and distribution, specialized architectures thereof (e.g. Edge, IoT, Big Data, AI/ML, etc).
>> >> >>
>> >> >> Kubernetes, containerd, rkt, Harbor, Dragonfly, Virtual Kubelet
>> >> >>
>> >> >> Xiang: Storage (Block and File Stores, Databases, Key-Value stores etc)
>> >> >>
>> >> >> TiKV, etcd, Vitess, Rook
>> >> >>
>> >> >>
>> >> >>
>> >> >> Hope this helps,
>> >> >>
>> >> >>
>> >> >>
>> >> >> Kind Regards,
>> >> >>
>> >> >> Alex
>> >> >>
>> >> >>
>> >> >>
>> >> >>
>> >> >>
>> >> >>
>> >> >>
>> >> >>
>> >> >>
>> >> >> From: Saad Ali <saadali@...>
>> >> >> Date: Wednesday, 10 April 2019 at 21:38
>> >> >> To: Quinton Hoole <quinton@...>, cncf-wg-storage <cncf-wg-storage@...>, "cncf-toc@..." <cncf-toc@...>, Alex Chircop <alex.chircop@...>
>> >> >> Subject: CNCF Storage SIG - Mission/Purpose/Ownership?
>> >> >>
>> >> >>
>> >> >>
>> >> >> Hi CNCF Storage WG,
>> >> >>
>> >> >>
>> >> >>
>> >> >> Following up on the CNCF Storage WG meeting this morning: I'm considering throwing my hat in to the ring as a possible TL for the proposed CNCF Storage SIG. I'm currently Co-chair/TL of SIG Storage.
>> >> >>
>> >> >>
>> >> >>
>> >> >> That said, I'd first like to understand what this new CNCF SIG will own?
>> >> >>
>> >> >>
>> >> >>
>> >> >> The Kubernetes Storage SIG, for example, has clear ownership of the Kubernetes volume sub-system. This, CNCF storage workgroup, so far has produced whitepapers on the storage landscape, surveyed users, and hosted presentations from 3rd party storage projects. While all valuable, these are short term tasks.
>> >> >>
>> >> >>
>> >> >>
>> >> >> My goal here is to make sure we act intentionally with a clear mission and purpose. A group without something to own will, at best, waste time, and, at worst, cause unnecessary conflict. To that end, I want to make sure the SIG has an obvious, long-term ownership of something concrete.
>> >> >>
>> >> >>
>> >> >>
>> >> >> Thoughts? What do we think will be the long term sustainability strategy for this new SIG? What will it own?
>> >> >>
>> >> >>
>> >> >>
>> >> >> Thanks,
>> >> >>
>> >> >>
>> >> >>
>> >> >> Saad Ali
>> >> >>
>> >> >>
>> >> >
You received this message because you are subscribed to the Google Groups "cncf-wg-storage" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cncf-wg-storage+unsubscribe@....
To post to this group, send email to cncf-wg-storage@....
To view this discussion on the web visit https://groups.google.com/d/msgid/cncf-wg-storage/CABBBJP3Mo3V%3DBXt_TDbJuFQDPwKbtM_15N5vc06ptOjk2%2BkdkA%40mail.gmail.com.
For more options, visit https://groups.google.com/d/optout.
You received this message because you are subscribed to the Google Groups "cncf-wg-storage" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cncf-wg-storage+unsubscribe@....
To post to this group, send email to cncf-wg-storage@....
To view this discussion on the web visit https://groups.google.com/d/msgid/cncf-wg-storage/CAHy7msgw-HvGcdHKqkx3zVm_616rtM4ENjxT%3DD%2B3%2BBw_Ppdnkw%40mail.gmail.com.
For more options, visit https://groups.google.com/d/optout.
|
