[sage-devel] Re: complex and unsigned infinity

2015-06-19 Thread Ralf Stephan
> > If not, then two tickets will have to be revised before they get included, > and a class for it created. > Also, this then would be wrong: sage: maxima('infinity').sage() Infinity sage: type(_) because >From Maxima manual Constant: inf inf represents real positive infinity. Constant:

[sage-devel] Re: complex and unsigned infinity

2015-06-19 Thread kcrisman
> > > Is Sage's unsigned_infinity intended to model complex infinity? > What else would it be? I think one of the reasons for it was precisely to provide something for Maxima's `infinity` to become in Sage. Though note sage: Infinity +Infinity which is annoying. But we already have sage: -

[sage-devel] Re: complex and unsigned infinity

2015-06-19 Thread Samuel Lelievre
Is Sage's unsigned_infinity intended to model complex infinity? >> > > What else would it be? > Well, you can compactify a line - into a segment by adding two points (-oo, +oo), - into a circle by adding one point (oo). Think of slopes of lines through the origin in the plane RR^2; the vertica

[sage-devel] Re: complex and unsigned infinity

2015-06-19 Thread kcrisman
> > > Is Sage's unsigned_infinity intended to model complex infinity? >>> >> >> What else would it be? >> > > Well, you can compactify a line > - into a segment by adding two points (-oo, +oo), > - into a circle by adding one point (oo). > > To me, unsigned_infinity could refer to that, > and co

[sage-devel] Re: complex and unsigned infinity

2015-06-19 Thread Travis Scrimshaw
> > sage: -oo < 0 > True > sage: 0 < unsigned_infinity > True > sage: -oo < unsigned_infinity > False > > as documented, it should be okay. I mean, what does comparison even MEAN > with complex infinity in the mix? I suppose you could make it incomparable > to anything but itself. Really I t

[sage-devel] Re: complex and unsigned infinity

2015-06-20 Thread rjf
You can certainly have an unsigned infinity that is not complex. If you have signed zeros, then 1/(-0) is negative infinity 1/(+0) is positive infinity 1/(0) is unsigned. Or as previously stated you can think of the unsigned infinity as a completion under division of the real line in some projecti

[sage-devel] Re: complex and unsigned infinity

2015-06-22 Thread kcrisman
> > Or your could point to Maxima, not explain anything and say , uh, "What he > said". > In some sense, I think that's why we have it in the first place, though Sympy also has something like this ("zoo"). I'd be very happy to either disallow any comparison or follow what Maxima and Sympy do

[sage-devel] Re: complex and unsigned infinity

2015-06-22 Thread Ralf Stephan
On Monday, June 22, 2015 at 2:40:53 PM UTC+2, kcrisman wrote: > > Or your could point to Maxima, not explain anything and say , uh, "What he >> said". >> > > In some sense, I think that's why we have it in the first place, though > Sympy also has something like this ("zoo"). I'd be very happy to