BTOB – Mission to Costa Rica


The bad weather in NYC ruined my travel plans to San Jose.  I left home around 11am but the flight only took off at 5:45pm. I missed the connecting flight in Miami which departed at 8pm. Had to spent the night in of of the many hotels that surround Miami’s airport.


Finally arrived in San José at 11:30am (this time of the year CR is two hours behind EDT ). After lunch, we had a meeting with the team that is preparing the report on FOSS in the public sector which is part of the larger FOSS and SME project which we are funding from the ICTD TTF. The report was one of the recommendations that we agreed with the CO and the national counterpart last December. What is really interesting about way the report has been developed is the methodology they have used to capture data on public sector investment on ICTs and software licenses. It turns out that different institutions use different classification categories to record the purchase of software licenses.

In addition, in most cases such purchases are lumped together with service and support contracts as well as with customization of applications. This is entirely an institutional matter as current administrative procedures allow for this. From  a procurement point of view, it is clear that buying a software license is pretty much like buying a piece of hardware (one pays a fixed price per unit purchased) whereas service contracts and professional IT services are different and entail even more competitive processes. Being that as it may, we do not really know how much CR is spending in software licenses. And changing current administrative procedures can be rather complicated. The team did try to estimate such spending but it was clear to them that such number was subjective and could no be shared publicly.

The report will be ready in a couple of weeks. And I have asked UNDP CR to request the team to fully document the methodology so we can analyze, systematize and share with other countries. We also need to carefully position this report in the current national context.

The real purpose of my mission is to advise the Costa Rican Social Security Agency (CCSS, which runs Oracle on a Linux server) on how to strategically use FOSS for some of its core operations. CCSS is one of the larger institutions of the country which provide health services to all the population, including the poor and marginalized (for a brief history of CCSS see this document:, in Spanish).

Like many other countries around the globe, Costa Rica is also facing a fiscal crisis. And one way to address this is to reduce public expenditures. In this context, CCSS is under the microscope amid allegations of bureaucracy, inefficiency and even corruption. While some are calling for full privatization, most are seeking alternatives to keep it running but becoming much more efficient and transparent. A national group of “notables” have also made a series of recommendations on CCSS. And one of them relates to the use of FOSS as a means to reducing expenditures.

The issues at CCSS are certainly much larger than FOSS. I shared this during my initial meeting with the UNDP RR/UN RC. And this fact entails that we have to be both very strategic and careful on how we approach this. UNDP is well aware of this and wants to thread carefully. But on the other hand supporting CCSS and helping it “modernize” (so to speak) can be a bit thing for UNDP, specially if this is done in partnership with other agencies that are already working with the agency.

My initial ideas include: preparing a diagnostic of the ICT situation in CCSS; use this to identity gaps and niches where FOSS can be useful; ask the CCSS Board of Directors to issue a FOSS policy which officially allows for its internal use (without having to replace all proprietary software; have an internal implementation strategy which will support FOSS deployment and bottom-up use; assess user demand and user satisfaction with current service provided (online and off line); explore the interoperability between CCSS divisions (they have six large divisions and multiple software and hardware platforms) and explore creation of internal one-stop  windows; assess internal capacity to support FOSS and innovation; and explore partnerships with private companies that support FOSS in the country.


We spent most of the morning at a meeting with the Infrastructure and Technology division (ITD) of CCSS discussing some of the issues on the possible migration to FOSS.  The Deputy Director of the division, who is not a technologist at all, made a very good presentation on CCSS. We learned that the institution is one of the largest in Central America with 50 thousand employees using over 20 thousand machines with a national network of over 970 local community centers, 103 sub-regional health centers and 29 national hospitals. CCSS does reach almost each and every of the 4 million or so inhabitants of the country. On the governance side, the agency is led by a President who is appointed by the President of the country and has a board of directors composes of thee government representatives, four from the private sector and two from unions and cooperatives. CCSS has six large divisions each with a director. Apparently,coordination across divisions is an issue (a feature common to most large organizations.

On the ICT side, ITD has over 500 programmers who in one way or another support both the central and the local health services. The health network is decentralized which means that many of the local nodes of the network are allocated financial resources for ICT expenditures but can in fact purchase whatever they want without following any ICT standards, etc. In fact it seems ITD is not entirely sure of what is going on in most of the network nodes and cannot seem to be able to properly manage the situation.

CCSS is essentially a Microsoft and Oracle shop. While some of the servers do use Linux, most office and business applications run on those platforms. For the last couple few years the agency has switched programming from visual basic, aspx, etc to Java which in a way is good news. But on the other hand we know that Java was purchased by Oracle a couple of years back so this is just a way for Oracle to gain market share from Microsoft – although Java is supposedly Open Source. The agency does not use PHP or Python for example, nor does it have any expertise on most open source programming languages -at least in the center; I would not be surprised if we find FOSS use in some of the local nodes of the network.

From the conversations I quickly learned that ITD knows little about FOSS. Not surprisingly, this creates lots of resistance, specially from ITD technical staff who feel their jobs will be on the line if FOSS is deployed on a large scale. So part of the job was to sort this out and instead try to open doors for the upcoming change process. To our surprise, ITD has prepared a report on the potential migration to FOSS. The report (which was given to us only on paper) has to core parts: the first maps the wide variety of systems that CCSS is using, ranging from old IBM mainframes using COBOL to the current plan  to deploy tables for some of the local service nodes. The second part was focused on FOSS and the potential benefits and risks in adopting it. While the quality of the first part of the report is excellent, the same cannot be said about the second one. There is thus clear need to increase the level of FOSS awareness in ITD personnel (for example the technical staff have never heard of Drupal).

At the meeting, I explained UNDP’s approach to FOSS and put on the table a couple of examples from other countries where FOSS has been successfully deployed. Certainly, the starting point for any FOSS policy and project is not the “forced” migration of desktops to Open Office. Experience tell us that this usually lead to failure as resistance from end users is very high and can end up backfiring against using FOSS at any level. Instead, the initial focus should instead be on critical applications and systems that the institution develops and manages as this does not directly affect end users and provides technical staff the opportunity to learn by doing. This view brought a sense of relief to ITD staff at the meeting and immediately lowered the levels of resistance.

At the end of the meeting 3 hours later, the Deputy Director told us that she was now clear on what needed to be done and informed us that the division will be a key partner in deploying FOSS within the Agency.

After lunch, we met with the President of CCSS along with the UNDP RR, a member of the Board of Directors, and the Director and Deputy Director of ITD. In short, the meeting was a rehash of the discussions we had in the morning. At the end of the meeting, the President of the Agency informed us that she is fully on board in the process. She also requested ITD to take lead and work closely with UNDP to develop a strategy to make this happen. At the meeting, the Director of ITD raised some issues on FOSS which I believe were not welcome by either the President of CCSS and the board member.


Early morning meeting with CAMTIC (ICT Chamber of Costa Rica, to explore ways in which the chamber can provide support to the planned process of change within CCSS. The President of the Chamber (who is also the head of ICT at the Central Bank of Costa Rica), the Executive Secretary and 5 to 6 members of the FOSS chapter of CAMTIC were present at the meeting which took place in UNDP’s offices. CAMTIC has international ICT companies as members so it has to thread carefully around the issue of Open Source.  FOSS companies at the meeting were sceptical about the potential of getting large competitive contracts from CCSS.   Some even mentioned that the risk of supporting this project was very high to them as they are still very small and would have to drop all other projects to keep up. If say CCSS decides to suddenly stop the project  then they will have to close shop.

Two good ideas emerged at the end of the meeting: one is to start marketing the services of FOSS companies in the country. The issue is that CAMTIC apparently has no resources to do this. The second one was to work with UNDP and approach national institutions that support local SME development to get some resources to beef up FOSS SME capacity, marketing, etc. The idea of having the 30 or so FOSS SMEs work as a a consortium was also discussed but the issue here is the actual implementation. SMEs that are already doing well and have already some market share might have no incentive to do this.

A working meeting with CCSS ITD followed. We basically discussed concrete next steps for the FOSS work CCSS will be undertaking with support from UNDP. We agreed to support both the elaboration of an overall diagnostic of CCSS ICTs (including HR, management and governance issues) and based on this prepare a draft strategy with short, medium and long terms goals and objectives. CCSS also agreed to start  a FOSS pilot by using one of the already planned and approved new initiatives.

In the afternoon we visited the Universidad de Costa Rica (UCR, A new UCR Rector was recently appointed who apparently is friendly to FOSS. As a result, the  new head of the computer center is a well-known FOSS supporter. During my visit to the computer center, I was amazed on how much UCR has already done with and for FOSS and how little people outside the University know about it. UCR has not only many  FOSS courses and training sessions (according to UCR they have already trained thousands of young prople on FOSS!) but has also developed some applications and recently organized a couple of meetings on FOSS, one involving countries from Central America and Mexico. I suggested to the computer center director to launch a PR campaign so the country knows about all this. I also asked if UCR will be interested in supporting the work we will be doing with CCSS and the answer was a resounding yes.

The Deparpment of Science and Technology from Brazil’s Ministry of Health ( has approached UNDP Costa Rica to explore ways in which the can assist the country in that area of work. The initial contact was established during Rio+20 where UNDP’s CR RR was present. In this light, the proposal to be developed will need to have three legs: health, sustainable development and new ICTs. The last meeting of my mission focused on this and included the former deputy Minister of Health, the UNDP focal point for environment in addition to the UNDP and the governance focal point. The deputy Minister provided 3 concrete areas of work while I pushed for using ICTs more strategically in the sector. A draft proposal will be share by the environment focal point next week before it is shared with the Brazilians the week after.


Left the hotel at 5am to catch the 7am to Miami. The Miami-NY leg was a nightmare as somehow the wrong “navigation software” was loaded into the airplane’s computer (according to the pilot). It took 3 hours to solve the problem which was only detected after we boarded the plane. To make matter worse, the flight did not offer Wi-Fi acccess to I could not even work while flying back home.

Cheers, Raúl


Print Friendly, PDF & Email