Thursday, March 6, 2008

Openbravo ERP 2.40 Update

These days, the Openbravo Development team is working very hard to finalize the 2.40 release and I think it is time to start sharing some news with our Community.

Openbravo ERP 2.40 will be a landmark release from many points of view.

In terms of functionality it introduces many cool capabilities that our users have been asking for. Just to mention some highlights:
  • Industry Template: this feature will provide the ability to package additional modules (functional extensions) and pre-defined configuration in order to dramatically reduce the implementation costs and time by providing a one-button delivery of everything that is needed to operate Openbravo ERP in a given industry.
  • Keyboard Operations: this feature will increase users' productivity by allowing them to operate all the essential functions of the product with the keyboard.
  • Revised Security: this project streamlines the security setup and makes it easier and faster to configure the system in organizations with demanding data segregation requirements.
  • Requisitions: with this feature, Openbravo completes the support for the procurement processes by cleanly separating the roles of a requestor of goods and services from the role of the purchasing manager who creates the POs.
  • Multi-schema accounting: ability to account transactions according to multiple accounting rules and currencies.
  • Additional reports
  • Much more
For the first time, with this release, you can follow our progress and have a sneak preview of what we are doing. Each project comes with a functional specification that describes what it is about and the progress of the whole release is summarized in a status page, publicly available on the Wiki.
From that page, you can:
  • View the status of each project.
  • Access the project documentation.
  • See if the project has been completed and merged in the trunk, in which case you can start using the feature by connecting to our SVN server and building from sources.
In 2.40 we will also put a lot of emphasis on quality and on cleaning up the defect backlog that we have accumulated over the years (you know, all those small defects... many of them are not that critical and you are often tempted to defer them, but when you put them all together...).
To give you visibility into this process, we have started publishing bug projections where, on a weekly basis, we estimate the inflow and outflow of bugs and we compare the actuals with estimates.
Just a disclaimer on this (I owe it to the team who very bravely published these projections): our estimates might not be very reliable at the beginning but we are confident that they will improve over time; we hope you will appreciate our efforts to continue giving you our honest best guess of what the future backlog will be like; please do not criticize us too much if things do not evolve as planned.

As this point, you might be asking the tough question... when will 2.40 be available? Well... if you are a developer you can follow up as we go and get hold of our daily build now from our SVN servers.
If you prefer to wait for the official installers, here is the schedule as we see it today:
  • April 4th 2008: Internal code freeze - all features are completed and tested in their own branch and all branches have been merged with the trunk.
  • April 11th 2008: Installer available and beginning of Acceptance Testing. The purpose of Acceptance Testing is to validate that the release installs successfully in Windows and Linux (our two most popular platforms), works properly against Oracle and PostgreSQL, and that none of the major flows is dead on arrival. We intend to involve the Community in this test but we would like to keep it small and we will probably follow the same process that we used for 2.35 and 2.35 MP1.
  • April 19th 2008: Acceptance Testing complete and 2.40 Alpha publicly available for download on SourceForge. The purpose of the Alpha cycle is to let the Community evaluate the release and give us feedback on quality and overall readiness. As such, 2.40 at that point we will not recommend 2.40 for production usage and we will continue to position 2.35 as our latest production release.
    As you can see from our bug projections, we plan to be very busy fixing defects during the Alpha cycle and we hope that we will be able to eliminate most of our backlog by then.
  • May: end of the Alpha cycle and begin of the Beta cycle. During this phase, we will aim at deploying 2.40 in production at a limited number of customers with the purpose of validating that the release is successful in live environments. Again, during this time, 2.35 will remain our latest production release.
  • June: 2.40 will finally become production and replace 2.35. Big celebration.
That's our plan for the next four months. We will keep you updated if anything changes.

1 comments:

Raphaël Valyi said...

Hi, Paolo,

Thanks for sharing about v2.40, this is much appreciated. We, at http://www.smile.fr, are actually waiting for that version. While we see a lot of potential in OpenBravo and signed a prtnership for that reason, we consider indeed 2.35 is a bit tough considering PostgreSQL bugs and current customization productivity limits compared to say TinyERP.

But with what is planned for 2.40 and by judging on the SVN repository all the heavy work you are daily investing in 2.40, we think 2.40 will be pretty production ready at a competitive implementation cost.

We also consider OpenBravo will have definitely taken over Compiere in the oss field once 2.40 is released. BTW, Compiere isn't transparently updating its SVN like OB, so even if they improve things, no one can know and anticipate, it's barely open source. So the oss commitment of OpenBravo is much appreciated and makes the difference, it's not wasted efforts.

The platform modularity you are talking about is much wanted. Your template stuff is definitely a great step forward. Still I guess all this works well for pl/SQL customization. But we are also expecting clear guidelines on how to inject custom dev into OpenBravo without investing into non maintainable SQL code but rather into highly object abstracted and debuggable Java platform based code. So while we totally understand you can't migrate existing legacy pl/SQL code for 2.40, we are waiting for platform solutions for Java level customization. Platform level SOA like Ofbiz or TinyERP platforms would also add a lot of value to OB.

Will that be already in 2.40 or will we have to wait for the Green platform? In our opinion this has much value than functional code. Based on our open source experience, we even consider that such platform enhancements are a prerequisite for a good community development catalysis.

Keep it up. Regards,

Raphaël Valyi.
http://www.smile.fr