You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@fineract.apache.org by Ed Cable <ed...@mifos.org> on 2017/11/13 14:58:22 UTC
Re: Release Plan for Next Major Release of Apache Fineract: [Re: Need
for help in reviewing PRs]
I'm going to reply to this thread regarding the original topic but to
address Myrle's comments from Oct 20, while yes we do have an issue with a
backlog and it's unfortunate we have to close out some pull requests that
weren't merged and we do have processes we need to improve, the majority of
the pull requests that couldn't be merged aren't because the community has
ignored them.
In all fairness to Nazeer, Adi, Sander and others who were reviewing the
pull requests, they provided comments and suggested feedback on the PRs in
a timely fashion but in many cases the original contributor never replied.
I think it speaks to hopefully an easier problem we can address by
improving communications because it's all easy for us to miss github
notifications amidst the myriad of notifications flowing through. I know we
don't want to create redundant communication flows or excessive email
traffic but perhaps there are some other bots or integrations we can tap
into which will allow communications to other channels than just the email
address. We could also make it a practice of emailing on list when there
are comments but I think that's a lot of added work, creates too much noise
on the list, and defeats purpose of using the comments thread.
Ed
On Fri, Oct 20, 2017 at 2:21 AM, Myrle Krantz <my...@apache.org> wrote:
> I believe at some point we're going to have to accept that the oldest
> pull requests can no longer be made useful.
>
> I've been considering closing any pull request which has not been
> commented on in the last 3 months. It's terrible, but it's more
> honest than pretending we're going to eventually get around to it.
> And closing with the comment that we just don't have the necessary
> resources to merge is more fair to the contributor than leaving them
> in limbo forever.
>
> Best regards,
> Myrle
>
>
> On Fri, Oct 20, 2017 at 11:14 AM, Sendoro Juma <se...@singo.africa>
> wrote:
> > Dear All,
> >
> > My wish, is if we can include and merge at least all current pending
> PRs, in Fineract,
> >
> > As well as on those in community app, android clients etc... so that we
> can ship a short time after Fineract.
> >
> > Thanks
> > Sendoro
> >
> > ----- Original Message -----
> > From: "Ed Cable" <ed...@mifos.org>
> > To: "dev" <de...@fineract.apache.org>
> > Cc: "Shaik Nazeer" <na...@confluxtechnologies.com>,
> "mifos-developer" <mi...@lists.sourceforge.net>, "Avik Ganguly"
> <av...@gmail.com>
> > Sent: Friday, October 20, 2017 12:09:56 PM
> > Subject: Release Plan for Next Major Release of Apache Fineract: [Re:
> Need for help in reviewing PRs]
> >
> > Nazeer and others,
> >
> > While this is a blend of several other threads out there, can we update
> the
> > wiki with a page which outlines out all the major features/external
> > contributions we are planning on including in the next major release so
> we
> > can ensure committers can assist in reviewing and Santosh knows what he
> > needs to focus on in terms of QA.
> >
> > From the other thread, on the Apache Fineract side, I know we're aiming
> to
> > ship the work related to two-factor authentication, notifications, data
> > import tool, credit bureau integration, GSIM/GLIM, Changes made by BowPi,
> > security changes from Thisura, and changes made by Sanyam for Swagger
> APIs.
> >
> > I created a page at
> > https://cwiki.apache.org/confluence/display/FINERACT/1.1.0+-
> +Apache+Fineract
> > .
> >
> > For now I just said it was 1.1 but I believe there's enough going in that
> > we could call it a more signficant release - but I'll let others on PMC
> > determine that.
> >
> > Ed
> >
> >
> > On Tue, Oct 10, 2017 at 11:33 AM, Avik Ganguly <avikganguly010@gmail.com
> >
> > wrote:
> >
> >> Hi Nazeer,
> >>
> >> I have done one round of review for the open PRs. Please find my
> comments
> >> for the open PRs.
> >>
> >> Backlog Tracking
> >> <https://docs.google.com/spreadsheets/d/1OoyPoWKrLDieDLK3Su1
> _c6oWN77zufA9E4hTexIU0a8/edit?usp=drive_web>
> >>
> >>
> >> Regards,
> >> Avik.
> >>
> >> On Tue, Oct 3, 2017 at 11:00 AM, Shaik Nazeer <nazeer.shaik@
> >> confluxtechnologies.com> wrote:
> >>
> >>> Hi Avik,
> >>>
> >>>
> >>>
> >>> Most of them are reviewed and provided with review comments. Please
> have
> >>> a look again following the older PR(s) first as sometimes contributors
> will
> >>> update same PR without creating the new ones.
> >>>
> >>>
> >>>
> >>> Regards,
> >>>
> >>> Nazeer
> >>>
> >>>
> >>>
> >>> *From:* Avik Ganguly [mailto:avikganguly010@gmail.com]
> >>> *Sent:* 02 October 2017 00:05
> >>> *To:* Dev; Nazeer Shaik
> >>> *Cc:* Mifos software development; private@fineract.apache.org
> >>> *Subject:* Re: Need for help in reviewing PRs
> >>>
> >>>
> >>>
> >>> Hi Nazeer,
> >>>
> >>>
> >>>
> >>> Are there any particular PRs I can help in reviewing or should I do a
> >>> scan-through / follow up on the older PRs first?
> >>>
> >>>
> >>>
> >>> Regards,
> >>>
> >>> Avik.
> >>>
> >>>
> >>>
> >>> On Thu, Sep 14, 2017 at 12:00 PM, Ed Cable <ed...@mifos.org> wrote:
> >>>
> >>> Hi all,
> >>>
> >>> Once again not to sound like a broken record but echoing the same
> >>> sentiments in my previous email about the need for help in GSOC
> reviews.
> >>> The need for members of the PMC to review pull requests in general
> extends
> >>> beyond just these Google Summer of Code contributions. Nazeer, could
> use
> >>> help so there's not a backlog of pull requests to be reviewed.
> >>>
> >>> Apart from just needing more hands on deck to assist, it would be ideal
> >>> that we could have a second set of eyes review pull requests that
> Nazeer
> >>> has himself made before they can be merged. Right now with only one
> >>> maintainer, when Nazeer is writing new features, himself, he has
> nobody to
> >>> do an external review.
> >>>
> >>> Thanks,
> >>>
> >>> Ed
> >>>
> >>>
> >>>
> >>> --
> >>> *Ed Cable*
> >>> President/CEO, Mifos Initiative
> >>> edcable@mifos.org | Skype: edcable | Mobile: +1.484.477.8649
> >>> <(484)%20477-8649>
> >>>
> >>> *Collectively Creating a World of 3 Billion Maries | *http://mifos.org
> >>> <http://facebook.com/mifos> <http://www.twitter.com/mifos>
> >>>
> >>>
> >>>
> >>
> >>
> >
> >
> > --
> > *Ed Cable*
> > President/CEO, Mifos Initiative
> > edcable@mifos.org | Skype: edcable | Mobile: +1.484.477.8649
> >
> > *Collectively Creating a World of 3 Billion Maries | *http://mifos.org
> > <http://facebook.com/mifos> <http://www.twitter.com/mifos>
>
--
*Ed Cable*
President/CEO, Mifos Initiative
edcable@mifos.org | Skype: edcable | Mobile: +1.484.477.8649
<(484)%20477-8649>
*Collectively Creating a World of 3 Billion Maries | *http://mifos.org
<http://facebook.com/mifos> <http://www.twitter.com/mifos>
Re: Release Plan for Next Major Release of Apache Fineract: [Re: Need
for help in reviewing PRs]
Posted by Nazeer Shaik <na...@confluxtechnologies.com>.
Hi Avik,
Can you help us reviewing PR(s) sent by me please. I will start reviewing
the other PR(s). Since we are merging too many PR(s) we need to give
sufficient time to Santosh for testing on staging environment.
Can we plan to close the PR(s) by Monday evening, so that I will create a
release branch and update the staging environment.
Regards,
Nazeer
On Tue, Nov 14, 2017 at 3:14 PM, Myrle Krantz <my...@apache.org> wrote:
> On Mon, Nov 13, 2017 at 3:58 PM, Ed Cable <ed...@mifos.org> wrote:
> > In all fairness to Nazeer, Adi, Sander and others who were reviewing the
> > pull requests, they provided comments and suggested feedback on the PRs
> in
> > a timely fashion but in many cases the original contributor never
> replied.
> >
> > I think it speaks to hopefully an easier problem we can address by
> > improving communications because it's all easy for us to miss github
> > notifications amidst the myriad of notifications flowing through. I know
> we
> > don't want to create redundant communication flows or excessive email
> > traffic but perhaps there are some other bots or integrations we can tap
> > into which will allow communications to other channels than just the
> email
> > address. We could also make it a practice of emailing on list when there
> > are comments but I think that's a lot of added work, creates too much
> noise
> > on the list, and defeats purpose of using the comments thread.
>
> Github notifications are set by default to send e-mails. No extra
> bots or integrations necessary. If contributors aren't responding to
> comments on their PR's, we need to decide if we want to do the
> remaining work ourselves, or if we are ready to close the PR's. We
> should not just leave PR's open indefinitely. The clutter accumulates
> and makes our project unattractive. Contributors can always create a
> new PR if we close a PR and then it turns out the contributor actually
> was still interested. Closing pull requests is not the same thing as
> deleting the original code.
>
> That said, I'm going to start the following:
> 1.) find the oldest PR
> 2.) If there is a committer request for changes on a PR, I'll ping via
> github asking if the changes will still be made. If not I'll go
> straight to asking on the dev list.
> 3.) If there is no response in 72 hours, I'll ask here if anyone
> wishes to merge the PR without changes or make the changes themselves.
> 4.) If I have no response after another 72 hours, I'll close the PR.
>
> I'm starting with this pull request:
> https://github.com/apache/fineract/pull/40
>
> I'll be doing them one at a time and we have 47 open pull requests so
> this will take about a year. If we continue accumulating dead pull
> requests, I may have to parallelize.
>
> Go ahead and get your favorite PR's merged *before* I ping them. Please.
>
> Best Regards,
> Myrle
>
> > On Fri, Oct 20, 2017 at 2:21 AM, Myrle Krantz <my...@apache.org> wrote:
> >
> >> I believe at some point we're going to have to accept that the oldest
> >> pull requests can no longer be made useful.
> >>
> >> I've been considering closing any pull request which has not been
> >> commented on in the last 3 months. It's terrible, but it's more
> >> honest than pretending we're going to eventually get around to it.
> >> And closing with the comment that we just don't have the necessary
> >> resources to merge is more fair to the contributor than leaving them
> >> in limbo forever.
> >>
> >> Best regards,
> >> Myrle
>
Re: Release Plan for Next Major Release of Apache Fineract: [Re: Need
for help in reviewing PRs]
Posted by Myrle Krantz <my...@apache.org>.
On Mon, Nov 13, 2017 at 3:58 PM, Ed Cable <ed...@mifos.org> wrote:
> In all fairness to Nazeer, Adi, Sander and others who were reviewing the
> pull requests, they provided comments and suggested feedback on the PRs in
> a timely fashion but in many cases the original contributor never replied.
>
> I think it speaks to hopefully an easier problem we can address by
> improving communications because it's all easy for us to miss github
> notifications amidst the myriad of notifications flowing through. I know we
> don't want to create redundant communication flows or excessive email
> traffic but perhaps there are some other bots or integrations we can tap
> into which will allow communications to other channels than just the email
> address. We could also make it a practice of emailing on list when there
> are comments but I think that's a lot of added work, creates too much noise
> on the list, and defeats purpose of using the comments thread.
Github notifications are set by default to send e-mails. No extra
bots or integrations necessary. If contributors aren't responding to
comments on their PR's, we need to decide if we want to do the
remaining work ourselves, or if we are ready to close the PR's. We
should not just leave PR's open indefinitely. The clutter accumulates
and makes our project unattractive. Contributors can always create a
new PR if we close a PR and then it turns out the contributor actually
was still interested. Closing pull requests is not the same thing as
deleting the original code.
That said, I'm going to start the following:
1.) find the oldest PR
2.) If there is a committer request for changes on a PR, I'll ping via
github asking if the changes will still be made. If not I'll go
straight to asking on the dev list.
3.) If there is no response in 72 hours, I'll ask here if anyone
wishes to merge the PR without changes or make the changes themselves.
4.) If I have no response after another 72 hours, I'll close the PR.
I'm starting with this pull request:
https://github.com/apache/fineract/pull/40
I'll be doing them one at a time and we have 47 open pull requests so
this will take about a year. If we continue accumulating dead pull
requests, I may have to parallelize.
Go ahead and get your favorite PR's merged *before* I ping them. Please.
Best Regards,
Myrle
> On Fri, Oct 20, 2017 at 2:21 AM, Myrle Krantz <my...@apache.org> wrote:
>
>> I believe at some point we're going to have to accept that the oldest
>> pull requests can no longer be made useful.
>>
>> I've been considering closing any pull request which has not been
>> commented on in the last 3 months. It's terrible, but it's more
>> honest than pretending we're going to eventually get around to it.
>> And closing with the comment that we just don't have the necessary
>> resources to merge is more fair to the contributor than leaving them
>> in limbo forever.
>>
>> Best regards,
>> Myrle