On the idea of a survey in which we target using the census, we could do a post-censal survey similar to the survey on the vitality of official-language minorities. We could use techniques such as oversampling—you're right that we can go to a certain level—but we have other capacities as well. We can use a modelling approach to model at the small area level. As a scientific organization, we have used these techniques on many occasions. This is something with potential. We will need to test it and look at it, but it is something that is feasible.
On top of that, there were my suggestions about administrative data. Administrative data allows you to cover 100% of the population as well. This is another avenue worth looking at.