Hi,

We are planning on releasing of our e-commerce system as opensource software. I am writing to you with some doubts concerning the idea and the release process itself. Our project is entirely based on CDI, Weld and Seam 3 and you are the community which I know the best. We have finished development and we have a working implementation of pharmacy store in Poland, so it is a complete and working software. 

Main (worth mentioning) features of our system: 

* multistore functionality based on single product catalogue and orders panel. It means that you can have more than one frontend store at the same time for the same catalogue.

* fully functional backend system for the store in terms of product catalogue, categories, stock quantities, sku-s, order processing flow and user authentication and authorisation

* promotions & loyalty programme functionality

* html template mechanism for frontend stores based on Wicket

* integrated full text search engine through Hibernate Search

* affiliate system with fully functional store ready to put into html layer on affilate site (we have a dedicated promotional page for this where you can read more about it and see a live demo presented directly on your site such as that one: http://demo.minikoszyk.pl/demo/omega/www.seamframework.org/)

* integrated email communication to and from customers - all communication is accessible via single backend panel

* integrated reporting system based on Pentaho

* user documentation and promotional material templates

This is a brief description of what functionalities we have already. Do you think that there is a need for opensourcing such a system? We know that there are some opensource ecommerce systems released but they are of poor quality and lack  necessary features. One that stands out is http://www.broadleafcommerce.org/ but it is based on Spring so our system would be in technological opposition to that.

We don't have vast experience in turning project into opensource but as we use other opensource projects a lot, we believe that good OS project is built on documentation, source code and automatic test system. Based on that knowledge we have divided this proccess into a few smaller stages:

Prerelease stage:

Provide documentation of the whole system in terms of overall architecture, particular module descriptions and business functionality.

Realese stage:

Our plan is to opensource one module at a time. For each module we plan to provide:

* technical documentation of module functionality and its features

* code refactoring, tests and release into the public available repo

* release of developers documentation containg some tips about usage of the module, possibility of extending and known issues

Post realease stage:

Provide installation instructions for building complete system from scrach.

What are your thoughts about this plan? Do you have any best practices about such process? Could you suggest improvements?

Last thing we are not sure about is the english name of the project. Currently it is "miniKOSZYK" which can be translated from polish as "mini shopping cart". Polish name is catchy and when we discuss it with other people it automatically stays in your mind but for global market this might be too difficult.  Should we change that name prior to global release or leave the current name as it is?

regards,

Marek