|
Re: Helping Projects (was Re: [cncf-toc] What's the point? (or,"What's the Emperor wearing?"))
Jessica Frazelle <me@...>
Bounties on bug fixes comes to mind.
--
|
|
|
|
Re: Helping Projects (was Re: [cncf-toc] What's the point? (or,"What's the Emperor wearing?"))
Jessica Frazelle <me@...>
Pay the maintainers and high contributors who don't have other means to make money.
On Thu, Sep 20, 2018 at 05:13 Chris Aniszczyk <caniszczyk@...> wrote:
--
|
|
|
|
Re: What's the point? (or,"What's the Emperor wearing?")
alexis richardson
Matt I kicked off a new thread on the "helping projects" topic you & Jessie have homed in on. Re: "favourites" I think this ought to be simple: 1) Graduated & Incubated projects get lots of help & pref. 2) Sandbox get much less, and zero marketing, per the guidelines I hope we can scale (1) in line with overall CNCF growth & budget. I think Sandbox projects will be many, for a number of reasons that I am happy to bore everyone with. So their budget needs to be limited. They are not "favourites". They may fail to get to incubation. But, they need a neutral home & various forms of help that I think can be boxed into Categories/WGs Incubated / Graduated projects ARE favoured. There is real DD, and TOC care here. The CNCF is making bets & should back them up. "But you say you are not a kingmaker organisation", some people complain. On the contrary - we do not pick winners, and can back more than one horse. The clear example right now is in service mesh where we see multiple high quality projects. There are and will be others. a
|
|
|
|
Re: Helping Projects (was Re: [cncf-toc] What's the point? (or,"What's the Emperor wearing?"))
If you have ideas on how we can help projects more, please give us ideas here or take a look at the servicedesk where we document the services CNCF offers for projects: https://github.com/cncf/servicedesk This was recently updated and we appreciate any feedback.
On Thu, Sep 20, 2018 at 2:08 AM, alexis richardson <alexis@...> wrote:
--
Chris Aniszczyk (@cra) | +1-512-961-6719
|
|
|
|
Helping Projects (was Re: [cncf-toc] What's the point? (or,"What's the Emperor wearing?"))
alexis richardson
Jessie & Matt Thank-you for raising and elaborating on the topic of how CNCF helps projects. The topic was aired (again) at the last GB offsite, whose slides were shared a couple of months ago. Here they are again as ref: https://docs.google.com/presentation/d/1BFEUAnbsboyRx9qh3jyRjN-V7ukZtA9-OTjIbknUBn4/edit#slide=id.g3e55b0b2a9_0_0 Personally I do not want "CNCF could do more to help projects" to be chiselled on my gravestone. I think the TOC, Contributors and project leads can ALL show some direction here, and I think we have to do so. And so - Jessie & Matt & anyone else! Please could you help me & Chris do something more about this. Let's (again) make a list of what we think Incubated & Graduated projects really need, by talking with the project leads & others. This list needs to be in a shared public doc. Not everyone wants the same things, but there are themes. Such as: * docs & project management * community management * project services (hosting, CI, etc) * github wrangling * community / contributor summits * AR/PR help (esp: for projects that don't have 1-2 companies driving them) * maintainer psychotherapy & creche ;-) WDYT? a
|
|
|
|
FYI: CNCF ServiceDesk Docs Update
Hey all, one of the feedback items from the last CNCF Maintainer Survey was improved documentation for the service desk used by CNCF projects: We have recently completed this work after some beta testing / feedback from projects. Thanks and feel free to contribute ideas on how we can help more. -- Chris Aniszczyk (@cra) | +1-512-961-6719
|
|
|
|
Re: What's the point? (or,"What's the Emperor wearing?")
Matt Klein <mklein@...>
Then when Beyond neutral IP, I agree this is the most important thing the foundation provides, and IMHO CNCF does not do enough of it (there are so many things that high velocity projects need I could go on and on and on). Unfortunately, this is the area in which decisions will need to be made. There are limited funds and they will need to be allocated across the increasing portfolio of projects in the foundation, so clearly some decision criteria will need to be developed (whether based on graduation level or something else) to decide how funds are allocated and for what purpose (and unfortunately this in and of itself will involve politics). I'm happy to see this conversation happening, as I ultimately I do believe that some level of choosing "favorites" is going to be required for the foundation to realize its full benefit to its most popular member projects. On Wed, Sep 19, 2018 at 10:23 AM, Jessica Frazelle via Lists.Cncf.Io <me=jessfraz.com@...> wrote: Yeah that was what I meant with shared IP.
|
|
|
|
Re: What's the point? (or,"What's the Emperor wearing?")
Jessica Frazelle <me@...>
Yeah that was what I meant with shared IP.
toggle quoted messageShow quoted text
I think there are ways to limit the effect of gamification of the system and I think you all have done a great job of this so far especially with the format for sandbox projects.... I think this is more a culture problem in that, if people see the leaders of projects and in the community pushing for more projects to be added to the foundation at a very past pace then we have lost a culture of "making the best tool of high quality for the job" and we lost a culture of innovation.... we merely have the culture of "winning". Which I am still unsure as to what people seek to win but I digress. Regardless, I think the right measures are in place with the sandbox to limit kingmaking. I think the focus should now be put on promoting (and I don't mean marketing) but leading the ecosystem to focus on building tools that innovate technology in the cloud native space with a high importance on quality and fostering collaboration. Then when those projects need help with things the foundation can offer, money for infrastructure, a place for shared IP it seems like it would be obvious that they should be projects in the foundation, of course they need to qualify etc. Just my 2 cents.
On Wed, Sep 19, 2018 at 1:12 PM Matt Farina <matt@...> wrote:
--
Jessie Frazelle 4096R / D4C4 DD60 0D66 F65A 8EFC 511E 18F3 685C 0022 BFF3 pgp.mit.edu
|
|
|
|
Re: What's the point? (or,"What's the Emperor wearing?")
Matt Farina
In addition to what Jess said, a foundation provides a vendor neutral (or as close as we can get) location for a project. Competitors on products and services have a place to work together on projects that are not controlled by a single vendor in its governance.
I think this is one of the biggest reasons for a foundation. Not everything needs this. But, some foundational things benefit from it. Then there is being frank on the politics. People are trying to game the system for career advancement, getting money for startups, more tag lines in marketing campaigns, for bragging rights, and more. In this highly competitive marketplace with money being thrown to the degree it is, I don’t know if we can avoid this all together. But, we can put a target on being useful and building useful things. For this I mean Jakob Nielsen’s definition whose gist is useful = usability + utility. Projects and people can be promoted for the useful.
|
|
|
|
Re: COMMERCIAL: Re: [cncf-toc] What's the point? (or,"What's the Emperor wearing?")
jesse.noller@...
+1
Agreed and well put.
From:
<cncf-toc@...> on behalf of "Jessica Frazelle via Lists.Cncf.Io" <me=jessfraz.com@...>
On Tue, Sep 18, 2018 at 18:04 Nick Chase <nchase@...> wrote:
In my opinion the foundations role should be a space for shared IP.
And I agreed that people are putting too much importance on being in the foundation. You can do open source without a foundation. The foundations role should not be marketing projects and creating non-organic growth but helping the projects have a space to work without worrying about IP or licensing. It should also help the communities of those projects get things they need like money for CI infrastructure or other things and making sure those projects communities are healthy.
thats what a foundation is in imho.
--
|
|
|
|
Re: RFC: Dragonfly
sheng.zou@...
Really excited to see this news. It's a trend to embrace the CNCF ecosystem (k8s, docker) currently, and we are trying to build our own platform based on k8s. In image distribution part, Dragonly is the best choice for us compared with other solutions, it saves us a lot of time and efforts, and we truely believe Dragonfly will be an important role in container tech in the near future.
Looking forwarding to that day Dragonfly be a member of CNCF family, which help more people, and meanwhile make CNCF ecosystem better.
|
|
|
|
Re: Scaling TOC community while minimizing politics
alexis richardson
Matt I agree! These are both good areas for a Category WG to be responsible for. One Q -- I wasn't sure what you meant by "Then, separate from the documentation effort, come up with a list of value propositions for both having a project in the CNCF and keeping it under control by some other entity. These can even serve as a checklist to see if a proposed project has it’s goals in the right place when they want to join. Do the CNCFs value props for a project help them in addition to being a cloud native project." In any case, shall we move this to a working doc? a
On Tue, Sep 18, 2018 at 4:59 PM Matt Farina <matt@...> wrote:
|
|
|
|
Re: What's the point? (or,"What's the Emperor wearing?")
alexis richardson
The foundation exists to foster cloud native technology and adoption by end users The basis for this is that certain open source projects solve problems specific to cloud native, as defined by the cncf Being a neutral home for those projects and providing core services and help has to be the absolute baseline. All projects however nascent have the right to ask for that. Helping with adoption requires users and the cncf to have a conversation in which technologies are recommended, as are patterns and other best practices eg for interop, scale, what have you If you combine the above then you need: 1. a high bar for projects that are being recommended to customers. Incubation and graduation have a high bar. It could be higher 2. a low bar for new (sandbox) projects that are high risk but need a neutral venue and core services 3. commitment to prune failures out of the sandbox, in a timely manner 4. clarity around this model We continue to fail at 4. Marketing of projects is something everyone will debate. It is hard to separate user engagement from marketing - the activities are so similar. I recommend review of the Sandbox marketing language. If you feel the CNCF is handling this badly please speak out. Sandbox is a very limited blessing but the projects are entitled to be excited about joining the Sandbox. Today's discussion was about how we scale CNCF and the TOC. The WGs and Categories are possible ways to federate work. That all obviously need more thought and care. Even if we had no sandbox we would still have to resolve this issue.
On Wed, 19 Sep 2018, 00:02 Joseph Jacks, <j@...> wrote:
|
|
|
|
Re: What's the point? (or,"What's the Emperor wearing?")
Bryan Cantrill <bryan@...>
I agree with Jessie -- and for whatever it's worth, here were my reasons to be involved with the CNCF during its inception over three years ago: I expanded on that ~five months later: Three years later, that's still the appeal for me: serving our industry by serving the open source projects that represent the foundation for cloud-native infrastructure. I think we have had some successes in that regard -- but not without our share of stumbles. The conversation this morning was a promising start to a conversation that sorely needs to be had -- and is likely without easy answers. - Bryan
On Tue, Sep 18, 2018 at 3:30 PM, Jessica Frazelle via Lists.Cncf.Io <me=jessfraz.com@...> wrote:
|
|
|
|
Re: What's the point? (or,"What's the Emperor wearing?")
Joseph Jacks <j@...>
I agree with Jessie!
On Sep 18, 2018, at 3:30 PM, Jessica Frazelle via Lists.Cncf.Io <me=jessfraz.com@...> wrote:
|
|
|
|
Re: What's the point? (or,"What's the Emperor wearing?")
Jessica Frazelle <me@...>
On Tue, Sep 18, 2018 at 18:04 Nick Chase <nchase@...> wrote: First off, please take this in the sprit in which it's intended. It's In my opinion the foundations role should be a space for shared IP. And I agreed that people are putting too much importance on being in the foundation. You can do open source without a foundation. The foundations role should not be marketing projects and creating non-organic growth but helping the projects have a space to work without worrying about IP or licensing. It should also help the communities of those projects get things they need like money for CI infrastructure or other things and making sure those projects communities are healthy. thats what a foundation is in imho.
|
|
|
|
What's the point? (or,"What's the Emperor wearing?")
Nick Chase
First off, please take this in the sprit in which it's intended. It's not meant to be snarky or argumentative, (though it will probably sound that way), it's just meant to start a conversation.
I've been thinking a lot about the conversation about working groups from this morning's meeting, and I think we're missing a fundamental issue. What I heard was a lot of talk about how "we don't want to be kingmakers" and "people put more importance on being a CNCF project than they should". Well, if being a CNCF project doesn't mean anything ... why do it? In fact, why have a foundation at all? If the purpose of the CNCF is just to foster cloud native computing, and not to validate a project's existence, then why handle projects at all? Why not just create standards, or even just recommendations, as W3C did/does(?). I guess what I'm saying is that while nobody likes politics -- and believe me I DESPISE them -- if you're going to have a foundation that is supposed to mean something, then ... it should mean something. So my feeling is that we either bite the bullet and get tough about letting projects in -- even if that means asking them to perhaps work together, or create a joint API and then manage the API -- or we drop the pretenses and just create a directory anybody can add themselves to. See, I told you it would sound snarky, but really, I am only trying to start the discussion. Somebody please, tell me what I'm missing here. ---- Nick
|
|
|
|
Scaling TOC community while minimizing politics
Matt Farina
I really liked the honesty around politics, companies (both big and startup), and the ways extending the community can be gamed. It’s great to see reflection on keeping things on track. I had a few ideas to throw at the wall to see if they or a derivative stick. 1. For new projects, working groups, categories, and analysis of an area we could have a template containing the areas the TOC wants details on and an example of one being filled out. This will lead people who are doing the work in the direction the TOC wants. For example, with a new working group there appears to be a desire for concrete measurable goals that can be tested for done. A template with an example could illustrate and communicate that. 2. The “cloud native” landscape is exploding. Big companies, startups, and everyone in between in getting going on it. This space is prone to churn. What the CNCF brings in and oversees could be considered separate from an effort to document the entire landscape as it changes. It may be worth completely separating these efforts entirely and not trying to overlap them. For example, documenting a landscape could be a matter of coming up with criteria for inclusion and then keeping the landscape up to date for things that meet that. Then having a process to catch projects, products, and services that are no longer maintained and pruning them. Defining the category and pruning criteria is the hard part. Then, separate from the documentation effort, come up with a list of value propositions for both having a project in the CNCF and keeping it under control by some other entity. These can even serve as a checklist to see if a proposed project has it’s goals in the right place when they want to join. Do the CNCFs value props for a project help them in addition to being a cloud native project. These are just ideas to start a conversation.
|
|
|
|
CNCF TOC Meeting: 9/18/2018
Here's the agenda deck for tomorrow: We will be covering a backlog of community presentation decisions, reminder on SAFE WG voting and hearing from the netdata community. See you tomorrow! Chris Aniszczyk (@cra) | +1-512-961-6719
|
|
|
|
Re: linus
luke@...
wow, times change. +1
On Sun, Sep 16, 2018 at 10:00 PM Richard Hartmann <richih@...> wrote: Thanks for sharing; unexpected, but positive.
|
|
|