Kapil - Welcome to the project and thank you for your comments and engagement. Again, I think that the reflections we get from newbies is like holding up a mirror. I'll try to make just a few points here but you've definitely got me wondering if we've got the right materials written up.
1) Minor point but to keep things separated - please use a different thread title. You are raising some big questions, no longer talking about failed builds; creating a new thread highlights the new topic) This listserv is our "permanent record". 2) You have some good ideas - each of these types of activities (onboarding calls, tech syncs, roadmap discussions, etc.) take time to organize, to host, and we have tried them... repeatedly and with some mixed success - if you're interested in helping, let me know. 3) Please note that the PMC (project management committee) is responsible for steering the project. PMC members are here on this list. The Chair (me) writes the quarterly report with input and submits it to the Board of the ASF [1] . Lots of what PMCs do can be found at the PMC guide [2] (links below) 4) With more than a little hubris I started this project with the ambition of disrupting financial systems such that they served people better - unbanked, underbanked, everyone - by effectively changing the game - away from bespoke systems. The biggest lever I could imagine was an open source system that allowed small community banks and lenders to have the same level of data processing sophistication as larger banking institutions. And, ultimately, the way in which banking services are delivered globally - mostly governed by a set of strict rules - indicates at a pattern level that all of the functionality is deterministic. Why couldn't this all be done as open source software? 5) The problem of offering such an open source project is that you need a profitable Company to stand behind it. The ASF calls these Vendors and each vendor here has a role (indirectly) to play in bringing requirements and helping to build the Commons. As an Apache project, being Vendor neutral means that no one vendor is supposed to CONTROL the roadmap or the contributions or anything about the project. [3] 6) How complete should we make the solution? This is not an issue of ambition, it is an issue of focus and of generating the conditions for a Virtuous Cycle - whereby "volunteers" (often paid by companies) contribute code not because it is Charity and a nice thing, but because they see a business imperative to stay on top of the latest release... so choosing what is the Commons is a critical thing. On this topic, a lot has been discussed over the years. One way to look at this is that the software is 90% ready to go for a small org, but less for a Bank with more features (e.g. card issuance, payment integration, advanced KYC, advanced underwriting and scoring, identity management, Helpdesk, front end components, etc). Making it easier to layer on top of Fineract is what drives some of the design thinking, especially around the restful API concepts. 7) Since there is a mountain of things to do, looking up from our desks to ponder the bigger vision is often hard. At the beginning of the year, I set these goals [4] In 2025, I'd also like to propose the following: * Readme and asciidoc documentation is improved and versioned with each release, with wiki content streamlined to project management. * Builds improvements as discussed in this dev thread https://lists.apache.org/thread/9l46b7p6bxt7c475o94wrg3k2wc54zdv * Progress on each of the FSIPs as outlined at https://cwiki.apache.org/confluence/display/FINERACT/Fineract+Significant+Improvement+Proposals In a virtuous cycle of innovation, we continue to need the "roadmap" of items like those on the FSIPs as well as a set of "anti-roadmap" items, aka much later priority roadmap items. The roadmap is defined by the community not by any one Vendor, but contributions drive that roadmap. We have to ask “what is being planned?” Or “what would you like to add?” And discuss. Thanks [1] https://cwiki.apache.org/confluence/display/FINERACT/Fineract+PMC+Reports+to+Apache+Software+Foundation+%28ASF%29+Board [2] https://www.apache.org/dev/pmc.html [3] https://community.apache.org/projectIndependence.html [4] https://lists.apache.org/thread/1799o3xx9n6f9nmxfqkryjcf2cdqslb7 On Mon, Sep 8, 2025 at 7:17 PM Kapil Panchal < [email protected]> wrote: > We are here to achieve Market Monopoly! Apache Fineract aims to serve the > world's three billion underbank or unbanked which is statistically > significant. This also aligns with the vision and mission of World > Governments. This essentially suggests that we are all here to help the > world alleviate poverty and this by itself is no trivial task. We are > surrounded by governments, governing agencies, reserve banks, financial > institutions, regulatory bodies, security agencies etc. > > Apache Fineract is in direct competition with similar product offerings > from a plethora of companies, one of the products that I came across > googling is from Oracle themselves and which also happens to be a > commercially paid product. So one might be tempted to ask what is the key > differentiator between Apache Fineract and Oracle? One is free to use and > the other is paid, so by default Apache Fineract wins? That would be at > best naive thinking, the customer is sane and has all the power and > required resources to choose from the market without affecting the end > customers. So Apache Fineract has no significant benefit as compared to any > other product when it comes to product adaption from the end customers. > > So what is the biggest motivation for a customer to choose a product? > Having worked for commercial companies myself, the biggest differentiator > comes from how the genetic structure is of the organization. Most of the > commercial companies biggest moving force is its balance sheet. Less P/L > and so kill the product etc, and most of the people are lost to sloshing > management positions. Whereas Apache Fineract stands out in this respect. > Apache Fineract is meritocracy, meaning it credits merit and its aim is to > retain bright sparks just like ourselves, look at the life history of some > of the creators of Apache Software Foundation. But what use is having a > good brain without management? It is again a disaster waiting to happen. > > Management is required, so that people are well aligned and are performing > to their fullest. Management has to be either horizontal or vertical or > both, but certainly not criss-cross or sloshing management. > > What is volunteering without accountability? I depend on some of my peers > for good code and inessence my peers depend on my deliverables. Calling on > dependable people is what the management should facilitate and that is what > makes Apache Fineract, Apache Fineract. > > DataStructures: Just yesterday I was working on this piece of code where > Collections are used, this comes to me as a surprise since Collections are > type safe but they are not structure safe. Meaning Collections can process > Set, Lists, Queues, Stacks etc. so if a Collection of type set is used in a > command operation then the database will have unique rows, but if the same > Collection uses type as list, though it will have unique primary keys but > duplicate database rows. This is a very big bug as the next programmer may > not know this nuance while coding, as he is only working with a subset of > the code as mentioned in his Jira ticket. And if there is no unique > constraint to the database column then we also have data integrity issues. > > Algorithm: Correct usage of algorithms is paramount to achieve code > quality and effectiveness, one of the algorithms that I saw was used versus > what would have been used. Let's say there are colourless buckets and the > user has to put colored balls in the bucket such that all the coloured > balls are sorted, but the user cannot see inside the bucket, so user has to > remember the previous step (this also makes the data to be available as > sorted and so this algorithm will break if the original data is not sorted > and sorting cannot be done when there are many Ids with complex joins) as > to what bucket he placed his last ball so that if it is same it will be > placed inside the same bucket or next but the caveat here is that the user > only remembers his last ball and blind to the bucket, verses in a map (in a > map the data can be in any order and the key can collect a List or a Map) > there are coloured buckets and the user has to simply place the same > coloured balls into the same bucket this is irrespective of the flat data > sorting order? > > Conclusion: > > - > > *Monthly Contributor Induction* – Onboarding sessions for new > developers to understand coding standards, architecture, and Fineract’s > data model. > - > > *Regular Technical Syncs* – Focused discussions on design choices, > algorithm selection, and performance optimization. > - > > *Dedicated Requirements & Steering Separation* – Developers focus on > clean implementation, while steering handles roadmap direction. > - > > *Milestone Recognition* – Celebrating major contributions (new > features, performance boosts, architectural refactors) to encourage > long-term commitment. > - > > > - > > Thanks, > - > > Kapil > > > On Tue, Sep 9, 2025 at 3:44 AM James Dailey <[email protected]> wrote: > >> Kapil - in the future, please be aware that you are making a REQUEST. >> We are volunteers here, not some service company that offers to keep the >> build working such that "we developers can work". No such thing exists. >> You also could propose to fix things and create a PR. "Fix it yourself" >> would have been a completely appropriate response from the listserv and in >> many other open source projects is exactly what you will get back. I'm >> glad no one said that. >> >> I think you just joined the project, and there is a steep learning >> curve. It would be helpful if you could help the next person by >> documenting what you did that solved your problem. Maybe suggest some >> documentation with a PR to our asciidocs, or open a Jira ticket with >> details if you find something specific that is broken. >> >> Is this you? inkedin.com/in/kapil-panchal >> <https://www.linkedin.com/in/kapil-panchal> >> >> I'm wondering what your project is about. Are you involved in the DPI >> efforts? >> >> James >> PMC Chair >> >> >> On Sun, Sep 7, 2025 at 10:28 PM Kapil Panchal < >> [email protected]> wrote: >> >>> Gentleman, Ladies, >>> >>> The latest development branch is failing builds, it seems there is an >>> error with e2e tests. Please fix it so that we developers can work. >>> >>>
