Are there plans to try to reduce the inner inconsistencies in sage?

[Georgi Guninski]

In sage 10.4, I can prove $-1 = -oo$ and I am concerned about citing
CAS with this inconsistency.

Are there plans to try to reduce the inner inconsistencies in sage?

sage: Kx.<x>=QQ[];f=Kx(0);f.degree()
-1
sage: gp(f).poldegree()
-oo

[Dima Pasechnik]

All you see here is gp taking a less standard choice of deg(0).

It's hardly an inner Sage's inconsistency.


>

[William Stein]

Georgi,

Also, you can browse all of the currently 5006 open "plans" for Sage here:

https://github.com/sagemath/sage/issues

 -- William

--
You received this message because you are subscribed to the Google Groups "sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sage-devel+...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/sage-devel/3CD656F5-63F3-4650-8967-CC792EDD8F7B%40gmail.com.

--

William (http://wstein.org)

[kcrisman]

>
>Are there plans to try to reduce the inner inconsistencies in sage?
>
>sage: Kx.<x>=QQ[];f=Kx(0);f.degree()
>-1
>sage: gp(f).poldegree()
>-oo


All you see here is gp taking a less standard choice of deg(0). 

To expand slightly, 

sage: Kx.<x>=QQ[];f=Kx(0)

sage: type(gp(f))

yields

<class 'sage.interfaces.gp.GpElement'>

(see https://sagecell.sagemath.org/?z=eJzzrtCzqbCzDQyMjrVOs_Wu0DDQ5OUqqSxI1Ugv0EjT1AQAndAJVQ==&lang=sage&interacts=eJyLjgUAARUAuQ==)

which isn't a Sage element.  Unfortunately, to alert the user to this every time would be very noisy; it probably suffices to remind oneself that using maxima() or gp() or the like does not give one Sage elements, per se, and sometimes behavior will be different.

That doesn't mean we don't have inner inconsistencies, and likely some are in the list of open tickets William pointed to!  We welcome work on them.  Just that this particular thing isn't an example thereof.