Testing Theories

In our advocacy for titling of public lands a Cadastral Map is essential. This is a map of lot parcels containing such information as the Lot Number and lot boundaries, and associated data about the disposition of each parcel.  This map lets us know which lots can still be covered by RA 10023, the Residential Free Patent Act, and which lots are not covered because they may have been titled already, or there is an existing claim, or there is a dispute over the property. 

The problem is that this map is not readily available with Municipal Governments.  DENR holds this data, mostly on paper, and mostly in the form of Lot Data Computations - a table of numbers which can be interpreted to form the polygons of lot parcels. We were challenged to come up with a method to convert these DENR data into maps that can be used both by DENR and LGU personnel.

We already knew that officials of DENR VII and of several LGUs in Cebu Province wanted to do something about this. DENR was willing to share their data. LGUs were willing to assign personnel to the project. So we posited a theory: LGUs would be able to build their Cadastral maps if DENR shared the Lot Data Computation with them, and if the LGUs would have the resources to get this work done. 

We needed several things to build on their interest and to help their officials and employees to actually create digital Cadastral Maps. First, we needed To do that, we needed several things. First, we needed an affordable software. Second, we needed an expert user of the software who could teach LGU personnel to use it. Third, we needed a training venue. We had a limited budget so we had to look for the least expensive sources of these three. 

We also had theories about these three components. Our first theory was that a Free and Open Source Software called Quantum GIS would be good enough to convert Lot Data Computations into parcel polygons. Quantum GIS is not only affordable; it can be downloaded for free from the Internet. It is being maintained and improved by an army of volunteer programmers - the latest version, Lisboa is 1.8.0 - meaning there have been at least 7 versions before. Hopefully, there will be more versions in the future.

We already knew that there were at least two expert QGIS users. We theorized that we could ask them to train officials and employees from interested LGUs in Cebu Province. Having worked with LGUs at some point in their careers, they would be able to understand the technical and political context of a GIS operator’s work in a local government.

Our third theory was that we could find a local computer facility with sufficient desktop units and learning venue, whose management and staff also want to learn to use Quantum GIS.  We believed we could do some sort of ex-deal: they provide the venue and computer units while we arrange for the trainor.

Nice theories, and they would remain that way without testing. So we tested them - in the simplest and most straightforward way we knew how (remember our budget was tight).

First, we got sample Lot Computation Data from our friends in DENR and sent these to three Quantum GIS users. We asked each one of them to convert the tabular data into polygon files. All three successfully created the polgyon files. Conclusion: the theory was correct, Quantum GIS can be used to create parcel polygons from Lot Data Computations.

To test Theory Two, we tried to contact the two QGIS experts we knew to ask them to handle the training. One could not be reached. The second was willing, but only if his Mayor would allow him to do the training on official time. We asked his Mayor, formally. We did not get a formal response but our trainor got his authority to travel.

Testing Theory Three was straightforward: we identified local computer schools and sent feelers to owners and managers until we got an indication of interest. We got a response from a College, so we talked with the President and owner. We proposed our deal and we got it: he would provide the venue, the computers, the LCD projector, the Internet connection; we would provide the trainor; his technical staff would be included in the training.

Having secured all three components, we sent out invitations to 10 LGUs. 6 responded, 6 sent personnel for the training. A month later, only 2 were able to present their pilot Cadastral Maps. So was our big theory correct? Well, yes - it was correct for 2 LGUs. There were factors in the 4 other LGUs that prevented them from completing their pilot cadasters - and to address those factors we’d have to come up with more theories.

In recap, we had one big theory - that LGUs, with data from DENR and sufficient resources, could produce their own Cadastral Maps. To test this theory we came across three other challenges - and theorized solutions to these challenges. But the more important point is that we acted on these theories to test them. Fortunately our theories were proved correct. If they had been proven to be wrong we would have had to come up with other theories - and test them too.

What theories can you test about your own work? What is the simplest and most straightforward way for you to test these theories?