Can you make a million from Open Source Applications ?

The first version of the Siyaka BPM center of excellence was born as an initial idea between the Department of Industrial Engineering, Computer Science and the Department of Science and Technology’s Tshumisano Trust (an initiative which I can never spell or remember how to spell correctly), based on my involvement with the department of trade and industry where we created a wholesale fund which was intended to fund small entrepreneurs (www.samaf.co.za). As part of my exposure to the LSM 1-4 market, I experienced the lack of funding and expertise in this market place, and also maybe how technology could/can contributed to their upliftment.

I was so inspired that I even got articles published on subjects such as “the MBA for the poor”, or “An Economic Development Model” etc. Going around industry trying to find solutions that were going to support this business case even proved harder to be. Coupled this with dynamics of Government, I eventually fled to where an entrepreneur would go – how to do with the least amount of money! Just before this I lost a couple of bar in an IT venture, so development of an IT solution was totally off – financially, physically and emotionally.

This only led to one path – the holy grail of IT : Open Source . With bits and pieces we wired a few servers together, went to the internet and downloaded from ERP to Business Intelligence to Workflow to Banking to whatever looked like it was going to provide a cheap alternative for creating a business application environment (you know; give Bill Gates and al those people a go!).

The hardest lesson learnt – the average Industrial Engineer doesn’t posses a degree in Computer Sciences; so it become extremely frustrating and lonely trying to get IT people to open source their time to implement these solutions (thus, no quick double click on an .exe file and there you go scenario). So in all honesty, the effort to make this work makes it extremely hard for open source applications to take off, except if its complexities become hidden in cloud computing and the user can use it for free or start paying if it takes off.

But not all in vain, in the past three years we have managed to get a number of OS applications to work, with the top scorer our OS Stack Architecture which we use to process our process intelligence queries. At the heart of it lies a reconfigured MySQL database which proved hard to be beaten from a speed and reliability perspective (and its free!). The worst experience – OpenBravo as an open source ERP application; after weeks I could only stare at a green screen with definitely not a user-friendly interface. Our content management system, Alfresco is cool, the Business Intelligence suite Pentaho powerful, but to create data cubes; back- breaking!

Simulation modelling? Yes I can show you OS that does this – but if you remember SIMAN/PASCAL programming under the wings of Prof Paul Kruger then you don’t want to return there, or go and live in Sweden to become a housebound programmer. So instead, we have started to find cloud computing applications to be rather helpful and easy to use, although you pay small amounts for certain functionalities. Here, I refer to something like www.gooddata.com which amazingly processes your data into a cube for analytics, although we have moved to the next level of simple to use applications such as www.sisense.com and www.processmodel.com – not expensive, but it does the job for an industrial engineer who wants to balance man, machine and money.