Avik We gave everyone a full month to bring up relevant arguments on this thread prior to a VOTE. We also previously asked for input in May and June 2023 and decided then to deprecate the project. We previously, in December 2022 and January/February asked someone to step forward (for the upteenth time) to create a FIRST official release of Fineract-CN. I don't recall you raising any objections previously.
In any case, I’ll briefly re-open this discussion thread with your comments about the problems of the current Fineract vis a vis the architecture of Fineract-CN. We will continue to finalize the vote, but we should hash your concerns out and at least acknowledge them. Please note however that unless there is sufficient interest in actually doing a release, the community has seemingly voted with their feet. Your comments - provided on VOTE thread: Avik wrote on 07-Jan-2024… All repos should not be archived (Ex:- accounting, template, provisioner). > Regression across modules, flaky integration tests, difficult to refactor > code are consequences of the architecture and current architecture is not a > long term sustainable solution. > New modules like overdraft/buy now pay later or scalable > savings/current/checking accounts should not be built in Fineract just > because of maturity of reporting module, UI components and other similar > reasoning. My response: The Fineract code has undergone significant changes over the past two years and is now heading to release 1.9. The Fineract-CN, by contrast, has had very little activity, and has NEVER, in six years, managed to get a release. SIX YEARS! Regardless of "should" or "should not", the functionality you mentioned is now in fineract. That decision was made. Is the Fineract-CN architecture superior to Fineract? Maybe. Lots of arguments to make but really it comes down to the fact that Fineract-CN is not an active project. So, it doesn't matter if it is superior. VHS won. Is Fineract-CN the only way to be cloud native? The only way to create a modular approach? No. Much of the work recently in Fineract makes it "cloud native", under most definitions. Progress on modularity is happening. Is the fineract architecture sustainable long term? The general consensus I have seen is that while there is a LOT of code debt, and hundreds of stale tickets by the way, starting over isn't always the best approach. We have a number of efforts underway to create modularity within the code base and to effectively refactor much of it. This is incremental. Is it easier to start afresh with new code and "clean" well-separated concerns, in a new architecture? Yes. Almost always. But, unless you can sustain an effort to get to functional equivalency, which is a far higher bar, then it's a false trail. What is done is done. Fineract-CN needs to be archived. We have heard substantial comments that it has created confusion for people coming to the project, and muddies the water in other ways. But, if it was a viable project of the Fineract community, then we would be obliged to explain those differences and make the onramps clear. But, it is decidedly NOT a viable project, so let's archive the repos. On Thu, Jan 4, 2024 at 1:00 PM James Dailey <jdai...@apache.org> wrote: > This post generated no response. I’ll take that as a clue that removing > the repos isn’t a problem. > > Sent from Gmail Mobile > > > On Mon, Dec 4, 2023 at 3:56 PM James Dailey <jdai...@apache.org> wrote: > >> TL;DR - Archive the Fineract-CN (cloud native) repos? >> This is a call for discussion, not for a vote. I will call for a binding >> VOTE in one week. >> >> HISTORY >> The Apache Fineract(R) Community accepted a completely new architecture >> about five years ago. That new concept was intended to revise and refactor >> Fineract into new microservices. The PMC voted to accept the code and our >> wiki pages and communication shifted to gather devs around the new >> cloud-native microservice concept. >> >> Unfortunately, until a project - or sub-project in this case - >> can produce a release, it does not get acceptance. There was never a >> release of the software. In the Apache process, this is a key requirement. >> >> In the meantime, the Apache Fineract(R) main github was significantly >> updated and continues to be the focus of work and refactoring. One of the >> main things achieved in recent months is to make Fineract "cloud native" in >> ways that matter. >> >> DECISIONS >> We took the decision in May 2023 to deprecate the content and we have >> done so. There may still be some outside links pointing to our site for >> this sub-project, but effectively, we have made it clear that this is not >> the focus on the Fineract Community. >> >> DISCUSS >> We may now want to archive the github repositories. There are 40 (forty) >> repos. There was an attempt in Jan 2023 to get a release out but it was >> not completed. The problem is that there are FORTY repos, compared to ONE >> for our main dev repo. >> >> We think that when we archive them, they remain available but are marked >> as "archive", and won't allow for write access. >> >> If you disagree, now is the time to make your opinion known. We will have >> a vote in a week. We are looking for any objections. If there is an >> objection, we can hold off for another 6 months or a year. >> >> >> >>
