1: \begin{abstract} 

2: Finding a zero of a sum of maximally monotone operators is a fundamental problem in modern optimization and nonsmooth analysis. 

3: Assuming that the resolvents of the operators are available, this problem

4: can be tackled with the Douglas-Rachford algorithm.

5: However, when dealing with three or more operators, one must 

6: work in a product space with as many factors as there are operators.

7: In groundbreaking recent work by Ryu and by Malitsky and Tam, it was 

8: shown that the number of factors can be reduced by one. 

9: A similar reduction was achieved recently by Campoy through a clever 

10: reformulation originally proposed by Kruger. 

11: All three splitting methods guarantee

12: \emph{weak} convergence to \emph{some} solution

13: of the underlying sum problem; strong convergence holds in

14: the presence of uniform monotonicity.

15: 

16: In this paper, we provide a case study when the operators involved

17: are normal cone operators of subspaces and the solution set is thus

18: the intersection of the subspaces. Even though these operators 

19: lack strict convexity, we show that striking  

20: conclusions are available in this case: \emph{strong} (instead of weak) 

21: convergence and

22: the solution obtained is (not arbitrary but) the \emph{projection onto the intersection}.

23: To illustrate our results, we also perform

24: numerical experiments. 

25: \end{abstract}

26: