?? so ... ?? vague idea now is that what bump more or less means is ... ?? related to how joyal and street describe the rep theory of gl(n,f_q) ... in that ... the "cuspidal content" of an irrep in the joyal-street picture corresponds to the "maximal torus together with character" in bump's picture ... ?? ...

?? issue of ... ?? what bump means by "maximal torus" ... ?? extent to which intended concept can be / is / should be seen as purely "abstract group"-theoretic ... vs "alg gp"-theoretic in some sense .... ???? .... ?? seems more like the alg alternative, though in some ways that seems surprising / weird ... ??? elements contained in no maximal torus ... ??? elt of order 3 in gl(2,f_3), for example .... ?? 8 conjugacy classes here ... ??? 3 conjugacy classes in split maximal torus .... 11,21,22 ...

4, 31, 22, 211, 1111

?? 1111 splits in 2 ??? ....

?? 31 also splits in 2 ?? ...

?? what else ?? .... ?? maybe the even ones split in 2 ??? ...

01
21

21
20

20
02

02
12

12
10

10
01


1 1

2 2

3 1

4 2 ?

6 1

8 1


4 4 8

12
11

31 3 3,6

21
20

12
10


22 2 2,4 ????????

01
20

211 2 4
1111 1 1,2