Both answers seem to be almost saying that it's a no-brainer that 3 tier is superior solution. perhaps
Rockford lhotka seems to be arguing that you should go with 2-tier unless the cost benefit
analysis of your particular situation comes down in favour of 3 tier.
he says :
As a good architect you should be dragged kicking and screaming into adding tiers to your system.
Security he suggests is the only area in which a 3 tier solution is clearly superior.
And he argues that
Worse, boundaries add raw complexity to software design, network infrastructure, manageability and overall maintainability of a system. In short, the more tiers in an application, the more complexity there is to deal with – which directly increases the cost to build and maintain the application.
Finally, as regards scalability, I'd like to know if it is true that if you use disconnected recordsets in ado for data access that you have connection pooling by default and hence a high degree of scalability anyway?