Which spline spaces for design?

We recently determined the largest class of spaces of sufficient regularity that are suitable for design. How can we connect different such spaces, possibly with the help of connection matrices, to produce the largest class of splines usable for design? We present the answer to this question, along with some of the major difficulties encountered to establish it. We would like to stress that the results we announce are far from being a straightforward generalisation of previous work on piecewise Chebyshevian splines.

RésuméNous avons récemment déterminé la plus grande classe d'espaces (de fonctions suffisamment régulières) bons pour le design. Comment connecter de tels espaces pour produire la plus grande classe de « bons » espaces de splines ? Nous donnons la réponse à cette question en pointant certaines des difficultés majeures rencontrées pour l'établir.