Open Letter to South African BEE / SMME Vendors

Dated 2010-10-14

Dear partners, in one way or another we have all been involved in contracting with government. This is not an official SITA communique at all but I know many of you are involved in the BEE and / or SMME space and this is an area where SITA and government are committed to grow the local economy. I am in my own capacity an advocate within SITA not only for this area of business but also more specifically for growing the use of free and open source software.

I do this because I passionately believe there is a good business model for it in South Africa, more specifically in the area of government business. My reasons for believing this are:

  • The South African Cabinet has committed to giving equal preference to open source (yes it may not yet be implemented the way it should have been – but we have to all help make it work). This means that government has to have had at least evaluated suitable FOSS alternatives before procuring proprietary solutions and SITA is already starting to query this in requests we receive. Recently in Parliament these queries have also been raised about certain large contracts that have been awarded. Closer to home a government department did an objective evaluation (using a proper evaluation methodology based on TCO, Support, Maturity, Functionality, etc) for a new content management system and compared 5 proprietary products versus 5 FOSS products and a FOSS product won hands down in all categories! This will go to market now for a service provider to implement and support it.
  • SA Cabinet has directed that open standards will be used for document exchange within government in support of the MIOS. This means that in practice FOSS products such as OpenOffice (and more recently LibreOffice) “should” be able to compete side by side with the likes of MS Office and similar (if the documents have to be exchanged between Departments in ODF format). The open standards endorsed by SA government are not just local, they are SABS as well as international ISO standards.
  • FOSS enables any company (large or small, metro or rural) to have full and equal access to the source code in order to understand how the software operates and to enable them to make certain customisations so that a client can have own branded software.
  • Although the software is free, and this is the key part, government is not in the position to support it themselves and as an enterprise organisation they actually have no wish to. Government needs to be able to pay for services in order to implement and support software as a service (whether it be proprietary or FOSS). Yes some government entities still think FOSS is “free” software but various FOSS advocates are working hard to change this perception. It is about the services to support the free software.
  • Government is obligated to invest in local companies for products and services. FOSS actually maximises this investment in local companies as 100% of the cost to government can go to local companies’ services.
  • Government is the ideal “client” to invest and initiate this this type of business model as they require services, they are obligated to invest in local businesses, they are obligated to create jobs which indirectly uplifts communities.  Government itself is not in the business of making profits, but rather they are in the business of service delivery to citizens.
  • FOSS as a service provides a low barrier of entry to SMME’s as the only investment is training. The first step is achieved through obtaining the software itself and the code for free, and the following step is free online community support through a world-wide community.  An obvious goal would be to get some form of accreditation or certification, and some FOSS projects do actually provide this.
  • As the FOSS model is not about box dropping, it requires a supplier to be familiar with the vertical industry (be that ECM, help desks, telephony, network management, ERP, etc).  The upside of this though also is that a service should be more sustainable going forward than ad-hoc box dropping.
  • One obvious way of tackling the issue is (like a well known Cape Town vendor has already done) is to pick an alternative FOSS product where the vendor already supports a proprietary product in the same vertical market, and to tender both of those solutions to government. Another vendor has taken the free Asterisk PABX software and written their own easy to use GUI front end for configuring Asterisk. This they installed on a server and it was bought by PGWC locally as a service. The GUI front end is pay-for-use though and belongs to that vendor.
  • A vendor who can kick start a FOSS based service like this with government, could grow it to support commercial companies not only locally, but even internationally. It also offers the ideal opportunity for expanding such support services within rural areas of South Africa.

This is not yet an ideal or mature model at all and many issues need to still be tackled and debated. The fact is though also that FOSS adoption is growing rapidly throughout the world. Some of the more obvious obstacles still though are:

  • Government going to tender for a specific proprietary product – SITA is opposed to a single product tender and one of the gates to be passed is asking what objective and transparent evaluation took place to limit the tender to a single product. SITA’s view is that the market should be tested against an open specification and the best product selected. Another issue here is that the Cabinet policy states that FOSS products need to have been considered – SITA should be asking for evidence of this. Both proprietary and FOSS solutions can be submitted by industry as bids on any SITA tenders.
  • Marketing – obviously proprietary vendors spend much of their money on advertising, gifts, lobbying, etc. Some are better experts at this than actually producing quality software. The only counter I can suggest right now is FOSS lobbyists (such as myself and many others) and the FOSS Policy that tries to level the playing fields somewhat.  We also have government’s FOSS Programme Office that is doing much to promote the use of FOSS and they have a magazine that is distributed to government. There is also the FOSS government website at (where the FOSS newsletters can be obtained). The CSIR, Dept Science and Technology, etc are all strong proponents of FOSS. Many local vendors themselves already have good relationships with government and after all government is not buying the software, they are buying the confidence and support that comes from the relationship and track record of the vendor.
  • Free – this was a problem as government used to perceive they get the software for free and then they struggle to get support, training, etc. This is changing as they realise they must pay for support and this will provide the training, maintenance, etc that they require. There are training houses in SA that give Ubuntu, Red Hat, OpenOffice, etc training and these opportunities will grow.
  • Security and Quality – these are really no longer issues that are raised about FOSS as most more mature FOSS solutions have come a long way already and are world renown in terms of their reliability, support, etc.

So where can you begin….? Well have a look so long at my own list of FOSS software that I have compiled and refined. It is not by any means officially endorsed by SITA or government but I have tried to list only software that is relatively mature and that have already received worldwide recognition. In some instances I have also indicated which government agencies or international corporations are using the software. The software is free to download and use and best of all – in most instances the software will run on Windows, Mac and Linux desktops (interoperability is another key goal of SA government). Have a look at

SITA is relooking at FOSS on transversal contracts as it can be appreciated that the procurement model for FOSS is slightly different. One thing I am pushing for is that a contract for say GIS, includes both proprietary as well as FOSS solutions on the same contract. The challenge is, who will bid from the vendors side to have the FOSS solution included? This would be the ideal but in the absence of vendors we may have to certify some FOSS solutions ourselves and put them on the contracts.  In the interim though contract 570 for services is being used. Contract 570 can be used to procure services (no product, hardware or licensing) to implement and support a FOSS solution. Other transversal contracts can be used for the purchase or any hardware that is required. We also have the option of open quotations up to R500,000 inc VAT or open tender above R500,000 to procure more complete solutions.

I hope this gives a small glimpse into what could be possible in the near future and I look forward to discussing the issues further with any of you.

Danie van der Merwe