Hi Nick, Could you please give PMCs and initial committers write access to the Wiki page? https://wiki.apache.org/incubator/MilagroRescueRoadmap I will update the wiki page to add our proposals.
The first problem to solve is lack of write access permissions for committers at document sites and repositories. We need write access to contribute ether it is documents or codes. Regards, Go Yamamoto -----Original Message----- From: Go Yamamoto [mailto:yamamoto...@lab.ntt.co.jp] Sent: Saturday, April 21, 2018 3:34 AM To: dev@milagro.incubator.apache.org Subject: RE: Milagro Rescue Roadmap Hi Nick and team, Thank you for clarifying the problem. I think we can discuss by agreeing criteria that define what is work in Apache and what is donation. In my opinion the difference is in the place where the ticket/review processes are held. The works at Apache will be a bunch of codes that resolves tickets at Apache and that are reviewed and accepted by the Apache project community. All the other works can be thought as donations. The codes in various milagro* repose are technically matured to work as the MFA product as described in the project proposal as far as the engineers in NTT play with them, and I suppose so is in Miracl too. That is why no significant changes are happening in recent months. However it is obvious that we need a lot of works to develop it as a matured open source software that is widely used and is accepted as a fundamental software. To start the works at Apache, I propose an immediate action list. (1) Prepare Milagro JIRA a. If we can user https://issues.apache.org/jira/projects/MILAGRO/, then provision accounts for committers and announce it as the project JIRA. (2) Prepare write access rights to committers at the official repository (3) Bind Milagro JIRA with the official repository at GitHub (4) Commit selected codes from various milagro* repos and accept as donation. a. We need to discuss a roadmap to accept codes. (5) Ticket problems at Milagro JIRA for further developments. I propose to use Milagro JIRA not only for ticketing problems in the development but in the management processes. For example, I am an initial committer of this project but I do not have a write access to the official repository at GitHub, so it blocks for NTT to join the editorial works at the official repository. If we have JIRA, I can place a ticket that says “I need write-access” and we can discuss how to resolve it. Regards, Go Yamamoto -----Original Message----- From: Nick Kew [mailto:n...@apache.org] Sent: Thursday, April 19, 2018 11:28 PM To: dev@milagro.incubator.apache.org Subject: Re: Milagro Rescue Roadmap On Tue, 17 Apr 2018 19:46:36 +0900 "Go Yamamoto" <yamamoto...@lab.ntt.co.jp> wrote: > Hi Milagro team, > > We need to talk to update the wiki page. Indeed. > I have a question, what is our definition of "place elsewhere"? > > Because of the nature of the pull-request driven workflow, the code is > updated at each participant's fork repository first. > Does anybody think it is developed in a place elsewhere than Apache > for this situation? The issue isn't that individual developers work in their own forks from time to time. Or even most of the time. That's fine. What's not fine is that great swathes of work happen elsewhere and appear as a big code dump! That's a donation, and the idea of incubation is that we *start* with a donation and work in the open thereafter! The issue is that nothing seems to be happening at all at apache. The last commit that does anything (and that just fixing a URL) was on October 17th. And nothing before that since 2016, when we thought activity was initial project setup! Meanwhile, as John pointed out, development seems to be happening elsewhere: various milagro* repos at github. The bundle John cited as an example: https://git1-us-west.apache.org/repos/asf?p=incubator-milagro-mfa-server.git ;a=log looks to me like a lot more than a developer's working branch! > If yes, then is it a problem to solve in managing this project? It's a roadmap we want right now. Why is nothing happening at Apache? Does the project really want to work at Apache (bearing in mind, working elsewhere and dumping code to Apache is not a solution)? Are all the team on board? I'm looking to you (everyone@milagro) to think through how work can migrate to apache, and commit to migrating. Then flesh out that roadmap! Note that on the infrastructure side, we can do more now than when Milagro first entered incubation. I think gitbox is likely to feature in the roadmap, if folks will use it! -- Nick Kew
