Because the lack of a competent organisation for this kind of expertise, it would be necessary to create a committee gathering specialists: engineers, computer experts, researchers, clinicians, representatives of builders, etc.

     At first this committee should establish standards "ISO" and classes for various uses: search, hospital, private hospital, cabinet (medical, dentist, podologist, etc.).  Labels to be attributed concern mechanics, electronics, data processing (algorithms, conviviality, etc.).

     I can give to the committee all the tools which I used for working out my own platforms.

     For the control, it is very simple: I thought of several solutions to verify platforms by using their software:          

• 1) Three electromechanical vices are put on gauges. By modulating the three pressures by program, one can feign all the migrations of the COP. This method is the most rational but it does not put in evidence the problems of flexion of the platform sheet.
• 2) The XY table . I worked out a XY truck (figure) capable of moving on the platform a weight of several kilos with a step of 1 micron. They are two step by step engines, 400 steps each tour which activate micrometers that move 0,5 mm each tour (resolution: 1,25 µ). The whole is programmed with computer. You can execute any movement you want with the mass, for instance a circular movement. It is for example easy to execute a circle, three cms diameter, and to verify it by software (length, area, FFT, etc). It would be very instructive to verify with this tool all the platforms.

• 3) The most simple means to verify platform qualities (mechanics, electronics, algorithm, etc.) would be to put on the platform a gallows with a rotary arm by which a weight can be moved . The arm would be rotated by means of an engine with programmable speeds (one or several tours in 51sec.). Software should give circles of various radii and stabilograms would be in sine and cosine. One can verify:

a) resolutions (weight and COP),

b) the linearity,

c) algorithms (length, surface, FFT, auto- and cross correlation),

d) noise as well as problems of spectral folding, etc.

         The gallows allows to draw only circles. But that's enough. I have this apparatus.

• 4) A mechanical vibrator put on the platform, the frequency of which can be adjusted to measure the frequency response and especially to verify the problems of spectral folding. I insist on this point, I can assure you that some FFT peaks are "Ghosts".

          At the conclusion of these tests, it will be possible to establish a sort of covering of the platform putting in evidence all the defects.