Moving from the pure P2P JXTA environment last year has been a lot of fun and has laid the groundwork for easy integration with the Wave code which had not been announced at the time of the change. The fact that many CRM, ERP and Accounting systems are currently offering an XMPP option with nearly the same level of support as Email is both a welcome and growing trend. This has been good for presence and proximity based services on devices that were difficult to cater for with the JXTA toolset alone while allowing us to limit the number of different versions of protocol in the middleware stack.

The addition of an XMPP server in each middleware stack allows for the federation of communication services which provide interactive messaging, processing, reporting and search capabilities in addition to enabling federation with not only other XMPP users but to users of both public and private services such as Google Wave and others which are currently in development, including SAP. The integration of the Wave protocol is interesting as it was developed primarily as a communications tool with a focus on group collaboration. The main issues at the moment are the speed at which the protocol development within Google has been progressing and the corresponding changes to the protocol. We feel however that the introduction and use of this protocol is important for us and the additional time required for completion of the new version is a rather small price to pay when the myriad benefits are taken into consideration.

Some of the interesting benefits of the new protocol include the ability to develop Robot processes that seamlessly integrate with your workflows, providing a substantial reduction of manual intervention. Some of the immediate benefits include the ability to develop your workflows using the graphical workflow designer to develop abstract process through collaboration with other menbers of your organisation in real time and without the installation of additional software. These abstract processes are then saved and later used to instantiate executable workflows, which are then monitored by the robots using business rules supplied (via a JESS based Rules Engine) prior to the processing and publication of individual messages.

The security model is looking great as well and provides fine grain control over the visibility of individual messages and tool functionality within the workflows. It also allowed us to develop some great gadgets for making data available by simple insertion of a block of code in any HTML page. In fact, we generate pages for response from users where required and these may be consumed on almost any device capable of rendering HTML5.

We hope to publish the full specification for the new version in the next few weeks and will have other entries here and in the Wiki to describe all of the new functionality as well as to give a bit of background for all of the new technologies incorporated in this version. We are also in the process of finally producing both the YouTube/Vimeo videos and a series of Podcasts to introduce the new features and technologies from both a practical and theoretic viewpoint.

Hopefully over the course of the next few months we will post more frequently and with luck might even get a style down. For now we will just plod along with this hopefully getting better at this blog stuff over time.

Pharm2Phork actually started as a bit of research for a firm wishing to track reusable fruit bins through their supply chain. In the analysis portion of the project we encountered so many challenges that it became an extremely interesting exercise.

After completing the initial investigation a few of us sat around and thought about what, in an ideal (non-commercial) world, could be done to meet the challenges that had presented themselves. On the edges of the chain there were small farmers, pack houses, transporters and intermediate storage facilities (to name only a few) with no computers, connectivity or manpower to collect data and pass it back in a typical connected fashion.

Over the last eight years we strove to solve the issues and have come up with what we think are some excellent solutions, albeit not always using popular technologies, but what we would consider to be the appropriate technologies for each task. We were lucky as we had funding as a research project and there was no push to use technologies to pad ones CV for the next job, or any existing preferences.

We are now refactoring the project and branching out to include additional projects both IT related and non IT, using the same approach in order to find a good mix of solutions for several challenges now facing small to medium size farmers, especially in the developing world.

We hope that these pages can be of interest and possibly of some use to others seeking to solve similar issues in this and other areas of development, but also to provide for ourselves a record of the effort and the process that leads to the completion of the current set of projects. Hope that you will join us and participate as after eight years or so new sets of eyes and viewpoints will be of both interest and value.