On February 3, newspapers reported that the Ministry of Land had signed a contract with an Indian company to digitise land records of several districts initially at a cost of $10m-20m (which presumably will total over $100m to roll out to all of Bangladesh).
While modernising land records is undoubtedly a critical undertaking for Bangladesh, two big questions arise.
Firstly, is the technology platform which will be used for these land records free/open-source and guaranteed to be available to Bangladesh in perpetuity with no license fee? Secondly, can’t any of this work be done by local companies?
The nature of land records is such that once they are digitised, they will need to be used forever. So the initial cost, however high, is not as big a factor as the maintenance cost in perpetuity.
Over the last 20 years, free/open-source software (like the Firefox web browser and Linux operating system which forms the basis of Google Android) has established itself in all fields of information technology as having the lowest cost, while simultaneously maintaining quality and security on par with any proprietary technology.
None of the media reports made any mention of open-source software use in the land records project. This critical aspect of the deal may have been overlooked.
In fact, there is an open-source technology which has revolutionised land records and GIS (Geographic Information Systems) around the world.
It is called PostGIS, and is an extension of the free/open-source PostgreSQL database which specifically handles the data requirements of digital land records. PostGIS is the system underlying www.openstreetmap.org (a Wikipedia-like equivalent to Google Maps), and is proven by use all over the world since 2001.
If the government has had the foresight to specify that PostGIS be the basis of the system it will buy, well and good. However, if they have not, then there will likely be many problems and even larger costs in the future.
First of all, a proprietary database (like Oracle) will require many additional developments to handle geographical data. Certainly, foreign companies which have previously developed digital land records for Indian and other governments have already made such proprietary customisations.
But in that case, will the Bangladesh government be committing to pay these foreign companies a fee to access our own land records forever?
Shouldn’t Bangladesh control access to its own land records? Furthermore, what is the justification of paying for those proprietary customisations if open-source systems like PostGIS provide similar features for free?
Undoubtedly the foreign company selected has experience and expertise in digital land records which local Bangladeshi companies do not, as Bangladeshi land records have not yet been digitised.
That doesn’t mean that the whole work should be given to a foreign company to be done in a proprietary system owned by them. Rather, what makes sense is to specify that foreign consultants should leverage their experience to prototype and do a pilot project of an open-source land records solution based on PostGIS to digitise existing paper records.
This approach would prevent future problems of being “locked in” to a single foreign vendor forever at high cost, as the solution produced would be based on open-source PostGIS and not a foreign proprietary system; an open-source system could be taken over by local companies with PostGIS experience after the initial development and piloted by the foreign experts.
Adoption of the above open-source strategy for digital land records would benefit Bangladesh on multiple fronts.
Firstly, Bangladesh would get the involvement of foreign experts. Second, it would not be dependent on the foreign company as the final system and product would be an open-source system and not dependent on a single foreign vendor.
Finally, using an open-source system like PostGIS would open up the field to local companies as well; once the foreign experts designed the initial prototypes and completed pilot projects in a few districts, local companies with experience in PostGIS could take over the system for further roll-out country-wide.
Such an approach would be a boon for the local software industry as well as a big savings in cost and foreign currency for the Bangladesh government.
In the long term, the fact is that all data of the Bangladesh government should be stored in an open-source format which it can access through free software like PostGIS without dependence on a foreign software company and without any exorbitant fees.
There should be an open-source software procurement policy written by the government to this effect.