Hi Tom, There are some internal delays anyway. But I thought we will be having for some info from Tal's work and assess the same, before we can make a final decision.
Thanks, /Shen -----Original Message----- From: Thomas Nadeau [mailto:tnadeaua...@gmail.com] Sent: Thursday, June 07, 2018 6:30 PM To: dev@ariatosca.incubator.apache.org Cc: smar...@apache.org Subject: Re: Updated: Discuss The Future Of AriaTosca - Meeting Minutes Guys, Its been a week (and well past “the end of this week”). Do you have any updates ? —Tom > On May 30, 2018, at 1:01 AM, S Shenbaga Rajan <s.shenbaga.ra...@ericsson.com> > wrote: > > Hi Sunil, > > Give us couple of days. We are discussing around this topic. > I would come back by end of this week. > > Thanks, > /Shen > > -----Original Message----- > From: Suneel Marthi [mailto:smar...@apache.org] > Sent: Wednesday, May 30, 2018 5:30 AM > To: dev@ariatosca.incubator.apache.org > Subject: Re: Updated: Discuss The Future Of AriaTosca - Meeting > Minutes > > @DJ - any updates - we are due to submit the incubator report , let us know > by tomorrow as to what Ericsson's decision is. > > > > On Fri, May 25, 2018 at 6:11 PM, Suneel Marthi <smar...@apache.org> wrote: > >> Thanks for the call this week. >> >> @DJ, please let us know by next week about Ericsson's decision. >> >> >> On Mon, May 21, 2018 at 11:17 AM, Thomas Nadeau >> <tnad...@lucidvision.com> >> wrote: >> >>> >>> Discussing the Future of AriaTosca >>> >>> Monday, May 21, 2018 10-11AM EST >>> >>> >>> Attendees: Tal, Ido, Shen, DJ, Tom Nadeau, Sunil >>> >>> Tom: Does anyone want to proceed forward with Aria and how? >>> DJ: Ericsson has been looking into this project/area. They started >>> buildign the missing parts of Aria and wish to contirbute them back >>> to the community. Initially we discussed these things with Arthur from >>> Cloudify >>> (with Tal) and to help with our contirbutions. Other than that, we were >>> always under the assumption that the communit was moving forward and >>> we thought we could contribute a few things. >>> >>> Most recently we have built a lot of components around Aria >>> (including Tosca 1.1). Issue is there are no active members that can >>> commit our PRs. >>> We are ready to contribute but need to know what the interest of the >>> community. >>> >>> Tom: Tal, Ido? Opinions? >>> >>> Ido: Speaking on behalf of Cloudify, currently we do not have any >>> committers in the Aria project. We are not actively working on Aria >>> at the moment - as a company we have other initiatives to work on, >>> and other engagements such as LF and ONAP community. I don't know >>> if that is going to change. >>> >>> Tal: I'd like to suggest a few options to move forward. Neither >>> Cloudify nor Ericsson can take leadership of the project. You can >>> take Apache incubation projects into "The Attic". Different status. >>> Can be unfrozen in the future. Taking it out of the attic is much >>> easier than starting a new project. Another option is to fork the >>> project and take it to github (outside of the Apache org). >>> >>> Tom: Thats just forking the code on github. >>> >>> Tal: Yes. We would tell our mentors in ASF that we are leaving the >>> project and will continue on github. >>> >>> Tom: They can fork independently and collaborate (Cloudify and Ericsson). >>> >>> Tal: Ericsson chose this project because they thought there was an active >>> community around an orchestrator. One option is to look into the Cloudify >>> "community edition". This does not support TOSCA proper but is an option. >>> >>> DJ: The important part of your last point was to use a pure TOSCA parser >>> implementation. This is why we were in favor of Aria. >>> >>> I would like to hear from you (Tal) about moving this to github. >>> >>> Can you explain why this project was started in Apache? >>> >>> Tal: Pros (to github): more flexibility (including management). Cons: >>> the management complications. >>> >>> Right now Aria is being used by the ONAP project in LF. The fact >>> that the code belonged to a foundation made it attractive to that >>> organization. >>> WIth that said, Aria does not have a big place. >>> >>> There are many successful reasons for how projects are successful on >>> github with healthy communities. >>> >>> I worked hard to get support within Apache for Aria and got mild >>> support. The community at large did not support this much. >>> >>> Sunil: This is exactly the question I asked to Arthur a year ago. >>> >>> Tal: This made sense because it helped us join the open-o and onap >>> communities. We hoped those folks would contribute back to the project. >>> Unfortunately those folks haven't, and I am not hopeful that those >>> folks will. >>> >>> I am actually working on a next-generation of Aria. I am rewriting >>> everything in scratch in Go to create a stateless parser (TOSCA). I have a >>> lot of faith in TOSCA but the approach needs to be a lot simpler. That >>> might be out there on the community to collaborate with. >>> >>> DJ: As you already pointed out, we are trying to understand if your >>> new project would have the same attraction. >>> >>> Sunil: Attracting contributors is not necessarily related to Apache. >>> Tensorflow is a massive project that is just on github without any >>> organizational affiliation. >>> >>> Shen: Cloudify started this project on Apache. If Ericsson wants to >>> reboot Aria is this possible? >>> >>> Shen introduces himself as a Product Manager at Ericsson. >>> >>> Suneel: Once the project is part of ASF its owned by ASF. The >>> problem is that the active commiters are no longer interested in >>> participating in the project. Chicken & egg problem. >>> >>> Shen: Is there a way forward? >>> >>> Suneel: Ido what are Cloudify's plans going forward? >>> >>> Ido: See comment above. >>> >>> Suneel: Cloudify has no stake currently in participating in the project. >>> >>> Ido: I am just saying this is the current status. >>> >>> Tal: There might be interest in Aria but is there interest in >>> keeping it in Apache? >>> >>> Tom: This is one remedy to the chicken&egg problem. >>> >>> Ido: I think it would be better to keep it under Apache than >>> starting a github project. Why can't we just add Ericsson to the Committers >>> list? >>> >>> Suneel: Do we have any PRs that need to be merged? >>> >>> DJ: We have many things which we wish to contribute. >>> >>> Shen: We have many improvements. >>> >>> Tom: Are they ready as a PR? >>> >>> Shen: Not yet. But we have many changes that are to be made ready. >>> >>> Tom: I think I merged about 6 PRs for DJ/Ericsson. >>> >>> Suneel: The problem is that there are no active committers that can >>> merge the PRs but there are no PRs that can merge/review the PRs. >>> >>> TOM: There might be a way forward by putting the project in the >>> attic while in parallel working on github. >>> >>> How does putting a project into the attic work? >>> >>> Suneel: I never saw a project moved out of the attic. But your >>> proposal of working on the forked version on github and making the >>> case based on a significant set of contributions to bring it out of >>> the attic/rebooting the project. >>> >>> Tal: Does ASF have a short cut/emergency way to advance committers? >>> >>> Suneel: Could Tom/Tal review/merge PRs ? >>> >>> Tal: I have no interest on moving forward on the project. >>> >>> Suneel: Consensus is to move the project into the Apache Attic. >>> When/if you guys would like to bring back the code from github, we can >>> restart the >>> project. We have no committers to review/commit PRs. If Tal/Tom could >>> lower the bar for committers, this could work too. >>> >>> TOm: There are no existing ASF comitters left in Cloudify's employ >>> >>> Tal: What is Cloudify's position Ido? You've been very diplomatic. >>> >>> Ido: Ericsson can continue with their own copy of Aria and develop it on >>> their own, but this is not a community. But there was no real community >>> when Cloudify was working on this actively. We can decide to >>> terminate the project and Ericsson can continue working. >>> >>> Tal: This is not only about control of the project, but you should >>> have goals such as growing the community. If you only want control, >>> you can simply fork the code on git. Is the community part on your roadmap >>> at all? >>> >>> Shen: Yes, this was on our roadmap. But now that the situation has >>> changed... It was our plan. For now the plan is to work on the code. What >>> I am hearing from suneel without existing committers we cannot move >>> forward within the organization (ASF). >>> >>> Suneel: we have committers but none have a stake in the project. to >>> move forward we need people to be active in the project including >>> add new committers. >>> >>> Tal: We can rubber stamp PRs. But we can accept them and then we can >>> vote to accept DJ etc... as a new committers. That is the nuclear option. >>> >>> Suneel: The question is: is this really worth it for a single vendor >>> to run this project? >>> >>> Ido: If you want to graduate from apache incubation, Ericsson will >>> need much more community. >>> >>> Tom: Community is important to move this forward. >>> >>> Tal: Graduation is not necessarily an indication of success. Don't >>> treat graduation as some ultimate goal. >>> >>> I think we have a bunch of ideas here. Maybe we can meet again in a >>> few weeks. The ball appears to be in Ericsson's court. Suneel >>> suggested the "nuclear option". Tom and I don't really care. I >>> would go with the nuclear option. My advice to the E// folks is to >>> think about this a lot for yourselves. It is a lot of work taking a >>> project with code you did not originally write. If you want to come back >>> Tom and Tal will support that. >>> >>> Suneel: That is fine with me but is Ericsson really ready to take >>> this over. >>> >>> Tom: Just post to the mailing list. >>> >>> Suneel: The other option is again to fork it on github and use that. >>> >>> Tal: There is an AriaTosca website that is owned by Cloudify. If >>> you go forward with that, Ericsson should have some discussion with >>> Cloudify. The existing site points at that. >>> >>> Tom: The site is incubator.ariatosca.apache.org The ariatosca.org >>> is owned by Cloudify. >>> >>> DJ: We will have a discussion internally and let the community know. >>> >>> Tom: Please push that to the mailing list whatever you decide what to do. >>> >>> Suneel: would you also like to look at what Tal is looking at? >>> >>> DJ: yes. >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> >>>> On May 18, 2018, at 2:39 PM, Thomas Nadeau >>>> <tnad...@lucidvision.com> >>> wrote: >>>> >>>> >>>> Sorry for the fits and starts with the last invite. That was >>>> the >>> first meeting invite I’d created with the BJeans system. >>>> This one should be correctly scheduled for Monday, May 21 at >>>> 10AM >>> EST and have a more appealing headline/title. >>>> >>>> —Tom >>>> >>>> >>>> >>>> >>>>> Begin forwarded message: >>>>> >>>>> From: Thomas Nadeau via BlueJeans Network <inv...@bluejeans.com >>> <mailto:inv...@bluejeans.com>> >>>>> Subject: Updated: Discuss The Future Of AriaTosca >>>>> Date: May 18, 2018 at 2:35:40 PM EDT >>>>> To: Thomas Nadeau <tnad...@lucidvision.com <mailto: >>> tnad...@lucidvision.com>> >>>>> Reply-To: "Thomas Nadeau" <tnad...@redhat.com <mailto: >>> tnad...@redhat.com>> >>>>> >>>>> <http://bluejeans.com/> >>>>> Thomas Nadeau <mailto:tnad...@redhat.com> has updated the >>>>> information >>> for your video meeting. >>>>> >>>>> >>>>> To join from a Red Hat Deskphone or Softphone, dial: 84336. >>>>> Join Meeting <https://bluejeans.com/7897462078?src=htmlEmail> >>> >>>>> (Join from computer or phone) >>>>> >>>>> Connecting directly from a room system? >>>>> >>>>> 1.) Dial: 199.48.152.152 or bjn.vc >>>>> 2.) Enter Meeting ID: 7897462078 >>>>> >>>>> >>>>> Just want to dial in on your phone? >>>>> >>>>> 1.) Dial one of the following numbers: >>>>> 408-915-6466 <tel:408-915-6466> (US) >>>>> See all numbers <https://www.redhat.com/en/conference-numbers> >>>>> 2.) Enter Meeting ID: 7897462078 >>>>> 3.) Press # >>>>> >>>>> Want to test your video connection? >>>>> https://bluejeans.com/111 <https://bluejeans.com/111?ll=en> >>>> <Mail Attachment.ics> >>>> <invite.ics> >>>> >>> >>> >>
