See various .pm files.

General:
 * Copy on write (helps for $x = -$x; cases etc)

Math::BigFloat:
 * ! bround(): some bugs may lurk in there (warnings)
 * ! bsqrt() (infinite loops for some cases (sqrt(1000))
     maybe: find a more proper ending condition
 * ! bmod() 
 * accuracy() & precision() maybe finished (bnorm() in every op, testcases)

Math::BigFloat::Constant:
 * really?

Math::BigInt:
 * remove trace()
 * objectify() is too slow and questionable (actually, it are only a few
   percent overhead...still)
 * ++,-- could be faster in Bigint via hardcoded (manual) increment
 * look at mul() and div() for more speed
 * bpow() could be faster by not using div() (pre-shift only last array elem?)
 * bround() gives wierd results for -$scale, abs($scale) > $len), and is only
   used by MBF for -$scale = -$len, anyway. POD is wrong for this, too.
 * theoretically, we could use $BASE_LEN = 7 instead of 5, to exploit 20% less
   work for huge numbers. Unfortunately, some code has hardcoded into it that
   each array elem is 5 digits wide.
 * the actualy calculation math code should be moved to Math::BigNum RSN!

Math::Big:
 * bernoulli() for $n > 20
 * tan() (depends on bernoulli)
 * more tests for cosh(), sinh(), tan() and all the others

Math::String::Charset
 * ! finish bigram support
 * different charsets at each pos (for grouping)
 * binary search in bstr() to find out base number?

Math::String:
 * finish cache string form and dec/inc

Please send me test-reports, your experiences with this and your ideas - I love
to hear about my work!

Tels <http://bloodgate.com/>
