Present divide and conquer strategy more abstractly

[defeo]

Page 14, line -2 and Page 17 line -11. Generally it is better to write double-and-add schemes as an associative operation in a monoid and let the implementor choses the addition chain to use. In this case give the law '' on tuples T_j=(zeta^{-j},sigma^j,xi_j,thetaj) such that T{j}T{k} = T{j+k}

[defeo]

Seriously?! This request is absurd!

[schost]

Actually, I kind of agree.. If I understand correctly what s/he's saying, we already have something like that p. 5, and I like that there's only one formula.

[defeo]

@schost, would you mind handling this one?

[schost]

Done (w/o monoids). Hopefully the formulas aren't too wrong. I tried to harmonize the notation a tiny bit.

[defeo]

At the beginning of p15, there is

so that we want to compute θ_a,r = θ_a

and, shortly thereafter,

Then θ_a = θ_a,r

Isn't this weird?

[defeo]

In the proof of Prop. 8, a letter b is used for a generic polynomial. However in the whole section b is something else (a divisor of h). Maybe we should pick a different letter?

[defeo]

I double checked. I believe formulas are correct. Thx!

[schost]

That's weird; I'll fix that tonight.

On 16 December 2017 at 11:59, Luca De Feo notifications@github.com wrote:

At the beginning of p15, there is

so that we want to compute θ_a,r = θ_a

and, shortly thereafter,

Then θ_a = θ_a,r

Isn't this weird?

— You are receiving this because you modified the open/close state. Reply to this email directly, view it on GitHub https://github.com/defeo/ffisom/issues/15#issuecomment-352195638, or mute the thread https://github.com/notifications/unsubscribe-auth/AD-ry8A6WK0BXEffg1y21tBRH-zhixMjks5tA_cAgaJpZM4P9aq- .