Architect Charles Thomson endorses George Santayana's view that those who forget the lessons of the past are condemned to repeat them. So here's a reminder
Housing Futures, the joint "crystal ball" initiative of CABE and the RIBA, is valuable and timely, although there is a danger that the debate will pass the major players by while the housebuilding industry stumbles on regardless.

There are, by definition, no easy answers to "the housing problem". It has to be considered as part of the prevailing socio-economic reality. There is no evidence that the real increase in house prices will continue but even if it doesn't, the statistics seem to show that, in order to meet demand, we will need to build 100,000 more houses a year than we do.

And, like the proverbial oil tanker, decisions have to be taken now to avoid disasters in the future. With this in mind, the industry needs to enter the debate before the housing output of the next decades replicates the mistakes made by most housing programmes of the past 50 years.

If the ambition of any social initiative is to create sustainable communities – ones that are resilient, happy, secure, tolerant and adaptable – and if the overriding concern of a housing strategy is to provide decent accommodation that can support and sustain individuals and families within those communities, then by any measure the provision of public and private sector housing over the past 50 years has often been one of failure.

Typically, the initial investment has been too small, the accommodation inadequate and repairs have been piecemeal. The cost of attempting to remedy the defects of poorly designed, poorly built and poorly maintained housing has been huge, and the legacy of poorly designed buildings is virtually impossible to eradicate.

Most housing estates are essentially blots on the landscape. How many genuinely decent homes have been built in the past 50 years? As a percentage of the total output, not many. With a few notable exceptions (the Peabody Trust springs athletically to mind) the output of housebuilders has been appalling.

Dwellings, particularly family dwellings, are too small for their purpose. The minimum Parker Morris standards of the 20th century are rarely achieved. What has been provided are dwellings that are inflexible, that are often badly built, that contribute nothing to the quality of the environment and are rarely considered as part of a community of buildings. They are usually a collection of boxes around a ring road.

If you want a target to aim for, you do not have to look far. Housing should be as generous in proportion and adaptable as the middle class 19th-century dwellings that have served our urban areas so well

It is possible to generalise about where some strategic improvements ought to be made: living rooms in family homes ought to be larger; bedrooms big enough to accommodate chairs, a television and a computer desk; floor-to-ceiling heights ought to increase; families should have a place to congregate and a place to be apart; balconies and gardens should be generous.

If you want a target to aim for, you do not have to look far. Housing ought to be as generous in proportion and adaptable as the middle class 19th-century dwellings that have served our urban areas so well.

There is no reason why new housing shouldn't be better, but spatially it is usually worse. This becomes even more important as the density of housing increases. Instead of unit sizes decreasing, they should get larger. The precedents of the early apartment buildings, particularly in the cities of Continental Europe and America, support this view. Room sizes were usually far more generous but infrastructure was more integrated and communities were able to tolerate higher densities.

Some sectors of the population (particularly the growing number of single-person households) can happily exist in small homes, but it is dangerous to design out the potential for the home to function as a social and working space as well as simply a place to sleep.

The system that forces all the players to "design down" to a minimum standard needs to be challenged and overhauled. The argument that costs will inevitably increase if standards rise must be tackled. The housing industry is a zero-sum game. If higher design-and-build standards were made mandatory that would, in the long run, drive down land costs.

This approach would inevitably challenge attitudes to land assembly, planning approvals and development financing, but if our slowly developing housing crisis is to be tackled and if we are not simply going to build more ghettos to be rebuilt at huge expense by a future generation, then housing needs to be considered holistically and good design needs to be a paramount consideration.