how about making GF(16) default to GF(16,'conway') no symbols injected, no conflicts...
On Wednesday, 20 January 2016 21:28:21 UTC, Nils Bruin wrote: > > On Wednesday, January 20, 2016 at 1:04:49 PM UTC-8, John Cremona wrote: > >> >> > Is there any reason to not make GF(16) behave as GF(16,'x') does, >> > unless specified otherwise? I do not see the point of requesting the >> > user to provide a character if (s)he does not care. >> > >> >> +1 for having a default, though I would choose 'a'. >> >> > There is a potential confusion there: > > sage: ZX.<x>=ZZ[] > sage: ZY.<y>=ZZ[] > sage: F=GF(16,'x') > sage: R=F['x,u,v'] > sage: S=F['y,u,v'] > sage: S(x).degree() > 0 > sage: R(x).degree() > 1 > sage: S(y).degree() > 1 > sage: R(y).degree() > 0 > > This is actually a bit of a mess: generator names are supposed to be used > in figuring out coercion maps, but currently this doesn't seem to happen > too much with GF(p^r) . [Really, F(y) should be an error, because it's > totally not clear where y should go]. So currently, having a default name > in GF doesn't seem to be too much of an issue because GF violates normal > coercion rules. However, if this were fixed then a default generator name > (especially when it's something usual such as x) would lead to odd > interactions with coercions from various structures. > -- 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 [email protected]. To post to this group, send email to [email protected]. Visit this group at https://groups.google.com/group/sage-devel. For more options, visit https://groups.google.com/d/optout.
