Thoughts on a co-operative address database-1By
The following thoughts come from Mr. Joe Lubenow, a member of GADA’s Steering Committee:
I do think there are technical and business model difficulties that would be obstacles to success.
As an example of the technical issues, consider the case of Zambia, for which UPU S42 template approval was just gained today, with indispensable help from Pierre. They have no database and also no national addressing standard, though the approval of the template could be a starting point for developing a national addressing standard based on the common S42/CEN vocabulary. As we worked on this in recent weeks, the position of certain address elements was reappraised, some less frequently occurring address types were overlooked until late in the process, and it was a learning experience for all involved. Moreover, it is clear that consistency in existing file layouts could not possibly be expected for some time until a national standard is published and those who seek to follow it have time to adjust.
Because of this, there would be a technical difficulty in determining what is the same address and what is not. Absent standardization the “proper” form of the address could change back and forth in the manner of a controversial Wikipedia entry. Not all names of persons are distinctive and so whether two records with differing addresses represent the same person or two different persons is to some degree indeterminate. If the name is distinctive, this would help in recognizing a COA if instances were dated and over time there were multiple recent records from different sources all pointing to the same conclusion. If many people have similar names this becomes much harder to decipher. But who would decide the exact criteria to be used?
That raises the business model issue. Wouldn’t there have to be a company or a group of companies taking responsibility with the others primarily in the role of beneficiaries? Few potential beneficiaries would subscribe until the system had sufficient volume and proven reliability. Perhaps the responsible companies could have a favorable rate for accessing the system to compensate for the additional resources contributed. In principle this could alleviate some of the concerns raised by Martin, but then there would have to be negotiations, and not on a once for all basis, but with periodical readjustments. This resembles the terminal dues issue that consumes vast amounts of time at the UPU. The statistics reflecting use of the system would probably have to be transparent, which would raise competitive concerns.
All in all, this would be a real challenge to pull off, without even considering privacy and security issues. But it is certainly a stimulating idea.