Hi all, Below you'll find the October 2017 board report for Fineract. The board meeting is on October 18th.
The report was due last week, so I've already submitted it to the board. It can however still be changed. If I've missed something important that you'd like to have included, say so, ASAP. Note to Ed, I've removed the mention of future releases from the Releases section and added it to the Activity section. Best Regards, Myrle Krantz V.P., Apache Fineract ## Description Apache Fineract (\’fīn-,ә-,rakt\) is an open source system for core banking as a platform. Fineract provides a reliable, robust, and affordable solution for entrepreneurs, financial institutions, and service providers to offer financial services to the world’s 2 billion underbanked and unbanked. ## Issues - Respond to the board's question: "If Mifos I/O was always intended for Fineract, why wasn't it part of the project from its inception?" Before I answer, first some history about Mifos I/O: Markus has been thinking about and planning a rearchitecting of MifosX since before it entered incubation at Apache under the name Fineract in December 2015. Besides his other duties he spent much of the time up till March 2016 gathering stakeholder support. In March, the Mifos Initiative sponsored Mifos Tech Days in Amsterdam to which all Fineract committers were invited (though not all could attend). At that conference, Markus introduced his architecture, and early versions of some of the command processing code. Videos of those talks were made and put online. Then Markus and Myrle went into heads-down mode designing the security concept, the accounting module, and other basic building blocks for the new architecture. In August Mark joined us and started work on a new UI designed to be capable of being ALV2 licensed according to the ASF's constraints, unlike the current Mifos community app. During this period we wanted to work with just the three of us in order to develop a clear architectural vision before we presented it to the community. We were new to Apache (Fineract was still in incubating), and didn't realize that we were violating the spirit of the "no off-list communications" rule. The code we were producing is not part of Fineract, and we never planned to introduce it without community buy-in, so it seemed a reasonable way to work to us. Early 2017, we changed the code to a public repository. We did a renaming and restructuring at the same time, during which we gave up the commit history. For this reason, you won't see the history before that point. In March Markus, Mark, and Myrle switched to a new employer named Kuelap, founded as a for-profit with the intention of making the same social mission that Mifos pursues fiscally sustainable. We have continued to work on the source code of Mifos I/O, preparing it for deployment. Mifos I/O currently consists of nearly 30 github repositories. When we first addressed this structure on our mailing list, we understood Greg Stein's response to be a veto against spreading code across multiple repositories for a project. Not long after those comments from Greg, Greg took charge of Apache Infra. For this reason, Myrle put careful consideration into the question of why we actually wish to work this way and produced an in-depth explanation which she presented at ApacheCon in November 2016 in Seville. Before it was clear that Apache would accept a multiple repo approach, this was a technical obstacle to introducing the code to the community. In April of 2017 Fineract graduated to a top-level project at Apache. Between the funding pressures of founding a start-up, the customer pressures involved in pilots, and the time-pressures that graduating a top-level project have produced, we haven't had the time to campaign for community buy-in that we'd like. At this point the source-code is extensive. The community wouldn't work if we just threw this into everyone's lap and said: "here: this is your new code-base". Even if we got the votes, people wouldn't be able to work effectively. So we're slowly trying to collect the community and help them understand the new code well enough to make an informed decision about whether they wish to accept it. While we have done considerable development off-list, we've been quite open about this fact. We are not quietly introducing off-list code into the code-base without community buy-in. Myrle has already made clear that she hopes to get *all* committers on board before we begin the process of IP clearance and code moving. ## Activity We've added three new committers and one new PMC member. We've added new moderators to the mailing lists. GSoC projects have finished, and are waiting to be merged. These projects which were led by interns on the Mifos Initiative and Apache include two-factor authentication, a notifications framework, data import tool enhancements, credit bureau integration, GSIM/GLIM, security fixes from static analysis, and changes made to API structure for Swagger API documentation. We are looking towards participating in GCI (google code in) and digital ocean's Hacktoberfest. We've begun discussions on bringing the Fineract CN code into the Fineract code base. Kuelap is hosting a demo-server for those who want to see what it looks like. We're also considering a point release which will contain some of these changes. The Mifos Initiative has engaged a number of individual volunteers who are working across the community on the following projects but not directly contributing code: Denila Philip, product management volunteer, has taken over guiding requirements and product managing the mobile money gateway which is a critical project for Apache Fineract but has dragged on for a couple of years and through two GSoC internship cycles. Ramesh Padmanabhan, an executive-level network engineer, is conducting interviews with partners across the community to document their cloud configuration and hosting environments to help build out toolkits, documentation containerization and hosting of forthcoming cloud-native architectures. Kristen Perchal, a BCG consultant, who is conducting interviews with individuals across the community to understand their roadmap development process and document previous strengths and weaknesses of engagement with the community on roadmaps to help advance a collective community-driven roadmap process with the ongoing transitions in the software and community. Aditya Oturkar, a project manager at Analog Devices, on documenting requirements for Drag and Drop/Ad-Hoc style reporting in Apache Fineract. Mabawonku Bolarinwa, a business analyst/microfinance practitioner from Nigeria, on documenting the credit bureau integration requirements for Nigeria. ## Health report The biggest threat to our project is the long list of unmerged pull requests on the Fineract code base. The oldest open pull request is still from March 2016. This problem is worsening with the contributions of the recent GSoC interns added to the "pile". We hope that the addition of several new committers who have experience and are still actively developing on Apache Fineract 1.0 will lead to progress on this issue. Avik in particular has started reviewing and commenting on older pull requests. While adding to the backlog of pull requests a positive sign of health in the community are the contributions of code from a number of new partners in the community including several modifications made by the BOWPI team from Guatemala, GSIM/GLIM changes made by the iDT Labs team in Sierra Leone, and changes made by R116 Solutions for Mentors International in Latin America. The addition of the Fineract CN code base, though not yet voted in, is also both a sign of activity, and a contributor to the problem. We're a small community and only about a third of the committer base are active coders. Those of us who are programmers have been splitting our efforts between two code bases. We still need more committers. Should we vote in the Fineract CN codebase, we cannot afford to let pull requests there linger too. In all though, the increase in discussion around the health of our project, and the addition of Fineract CN indicates there is a baseline of interest by contributors in our project that can continue to drive the project. ## PMC changes - Currently 13 PMC members. - Last PMC addition was Avik Ganguly last week. ## Committer base changes - Currently 18 committers. - New committers: - Avik Ganguly was added as a committer on Tue Oct 03 2017 - Ippez Robert was added as a committer on Mon Aug 21 2017 - Nikhil Pawar was added as a committer on Wed Oct 04 2017 - Santosh Math was added as a committer on Fri Aug 18 2017 plus two committers not in the Apache records: Zayyad Said (who was voted in in March, and has not yet returned an ICLA) and Nayan Ambali (who has provided an ICLA, but has not yet received an Apache ID). ## Releases - Released version 1.0.0 https://dist.apache.org/repos/dist/release/fineract/1.0.0/ on June 1, 2017. - The Mifos Initiative has issued a Mifos X release based on the current Apache Fineract code. ## Mailing list activity - Subscriber counts to dev and user have been slightly but steadily increasing. The number of new subscribers who introduce themselves has seen a slight uptick, which is nice. - Email counts are not completely comparable with previous quarters because in mid-September we rearranged what e-mails are sent to which lists. Those changes may have also driven changes in subscriber count to the issues and commits mailing lists. Hopefully, those changes will also make our dev list more attractive for newcomers. - [email protected]: - 160 subscribers (up 16 in the last 3 months): - 707 emails sent to list (969 in previous quarter) - [email protected]: - 12 subscribers (up 4 in the last 3 months): - 200 emails sent to list (248 in previous quarter) - [email protected]: - 140 subscribers (up 15 in the last 3 months): - 106 emails sent to list (41 in previous quarter) ## JIRA activity - Jira activity has been down somewhat. - 44 JIRA tickets created in the last 3 months - 20 JIRA tickets closed/resolved in the last 3 months
