Strange results from assume (maybe a bug?) ad asksign

Hi, I am developing a software running within Maxima and I need to define 
angles within the four region of the Cartesion plane (0<alpha<%pi/2, %
pi/2<alpha<%pi,...). In example, if the angle alpha belongs to the region 
0<alpha<%pi/2, I write:

(%i1) assume(alpha>0,alpha<%pi/2);

but...

(%i2) is
(alpha>%pi);
(%o2)          unknown

when it should be false.  Also, if one 
defines a partially redundant rule like this:

(%i3) assume(a>1,a>2);
(%
o3)                           [a > 1, a > 2]
(%i4) is(a>1.5);
(%
o4)                            true

when it should be unknown, if the rules 
are linked by a boolean and (or isn't so?).
Also, with the above assumption 
about alpha:

(%i5) asksign(sin(alpha));
Is  sin(alpha)  positive, negative, or 
zero?

Maxima should easily understand that the sine of an angle within 0 and %
pi/2 is positive...

In the Maxima manual I read, in the page explaining 
assume:

"Maxima?s deduction mechanism is not very strong; there are many 
obvious consequences
which cannot be determined by is. This is a known 
weakness."

Ok, but there are some things quite banal to be evaluated, but not 
less important... In my (for now) little software there are many trigonometric 
functions, and I am continuously asked about the sign of sine and cosine. I 
have solved the problem by declaring:

assume(sin(alpha)>0,cos(alpha)>0);

that 
is the same thing of saying 0<alpha<%pi/2, but it's not efficient and it is a 
quite serious issue for my code.
Are these bugs or lacks of functionality in 
Maxima :-(? Or am I wrong about something?

Stefano