My favourite journalism
Update me by mail...
The case for open source B2B
Firstly this Wikipedia quote talks about B2B and the industry supporting it:
Transaction charges run to 15%, plus membership fees, and usually tax is paid on all barter trade as if dollars had been received. Brokers often abuse their privilege, spending beyond what can ever be redeemed, and taking the first pick of trades as they become available. All of this is preventing the industry from thriving at a moment when money is being sucked out of the economy and barter should be coming into its own.
At the heart of all this incompatibility and competition is the software. Each provider maintains its own proprietary software, which is either expensive, or decrepit, or both. Most of these applications are from the first generation of web software which was about monolithic applications. These are characterised by being hard to change, especially as they get larger and more complex. The rigidity of these applications is constraining innovation and experimentation. The law, and the web, and user expectations are changing faster than ever and these small software houses can't keep up. The industry is not only carving itself up into silos, but crippling itself trying to maintain so many different software packages to do the same job.
The modern approach to software, (since around 2005) is to use open source frameworks. The framework provides a basic web application with basic tools for display, user management, content management, data manipulation and suchlike, and can be used for a very wide range of applications. This means a large community makes itself responsible for maintaining the framework. Then for more specialised purposes, this same community produces and maintains plugins which allow rapid development of commonly used features. The plugins work together because they adhere to the standards and conventions of the framework. This means, in the case of barter, that 80-90% of the needed code is already written and being used for other purposes. Then typically, a web shop is hired to analyse what is really needed and produce any new modules, to be open sourced and finally a layer of configuration and perhaps a unique graphic design to the site. This approach to software development, using 'layers' of sharing enables a drastic reduction in cost, while as much code as possible is shared.
The barter industry needs to be doing this instead of competing on software. The various providers need degrees of customisation, but underneath, what they want is the same. If software budgets were pooled, each providor would quickly find that they benefitted from the investments of others, they would have a more attractive package to offer their members, and it would be much, much easier to trade accross the network. Also monitoring and regulation would be simpler with a single reporting format. With standard data formats, members would find it easier to migrate to different networks. Exchanges could easily plug-in to a global marketplace. Providors would be able to complete on the efficacy of their matchmaking.
Let me put it another way. The proprietary software model is not working for the barter industry. Simple game theory indicates that once open source is introduced the others will soon not be able to compete. The leading adopter() of an open source framework will have a significant head start in regenerating the industry. And those who don't catch the train will be left in the cold.