On Sun, Mar 29, 2009 at 5:15 PM, kcrisman <kcris...@gmail.com> wrote:
>
>> So is the following *hypothetical* behavior not possible (or not
>> desirable)?
>>
>> sage: preparse( 'differentiate(y^3, y)' )
>> '_ = var("y"); differentiate(y**Integer(3), y)'
>>
>> If such a thing were indeed possible, I think it would make simple
>> uses of Sage for calculus a whole lot easier for students.
>
> +1. And in fact it probably should define y as a variable even if you
> just do differentiate(y^3).
For the record, that will never happen by default in Sage. That goes
along with choosing Python as the user language of Sage. Some
reasons:
1. Except for ^-->** and int/float wrapping, I think the Sage
preparser _only_ transforms expressions that would be syntax errors in
Python into valid expressions. Since differentiate(y^3) is valid
Python, the preparser shouldn't touch it.
2. If you make differentiate(y^3) automatically define y, then in any
code anywhere, anytime one makes a typo in a function name or variable
name, that variable will magically spring into existence and the code
may continue to run producing wrong answers. This goes against the
"explicit is better than implicit" maxim of Python.
>
> As for the rest:
> <mild-mannered rant>
> Following Mma is not always a virtue. One of the many reasons I have
> never really used Mma and rarely used Maple is because the syntax was
> simply too convoluted to pick up from the ground level unless I made a
> big capital investment. Things have gotten better since those days,
> but still are not ideal.
Are you saying the syntax of Mathematica and Maple have *changed*
significantly since those days? What "things have gotten better"? I
have the impression that the syntax of both Mathematica and Maple's
languages are nearly identical to what they were 15 years ago.
> For Sage to compete with Ms, it should have
> at least some things that are easier to do for entry-level users in
> unambiguous situations.
>
> This is the problem a lot of OSS has had getting used outside certain
> contexts - I told a graphic designer here about GIMP and that it
> pretty much does what Photoshop does, but for free and with a much
> worse user interface, and I think that sort of said it all; her
> background is software, but none of her staff's is, so enough said.
> I'm not saying we need a Java, clickable GUI in order to make easy
> things stay easy (nor am I capable of implementing this), but letting
> easy things stay easy should remain a goal ("insane ease of use" was
> David Joyner's phrase long ago in Sage time).
>
> In this case, unlike the multivariate context, there is very little
> ambiguity. Why someone should get an error after doing
> sage:
> super_huge_function_with_lots_of_xs_I_need_to_type_accurately_with_lots_of_parentheses_etc.integrate
> ()
> is unclear to me. Unlike the
> sage: (x+1)(x+2)
> x+3
> problem, which I can understand as being interpreted as a problem, it
> seems like this proposal is solving a problem that doesn't exist. The
> only ambiguity here is for someone doing
> sage: (y^3).integrate()
> 3*y^2
> who wanted
> 0
> because the variable was x, but such a user would presumably know that
> the variable of choice makes a difference anyway and would have done
> sage: (y^3).integrate(y)
> anyway. Or am I missing something? Maybe something in Pynac would
> break otherwise? I would be very open to seeing a technical argument
> for this.
The above is language design, and in Sage development we do not have
to do language design, since Python is the language. (Except for a few
small conveniences without which mathematics becomes way too painful,
e.g., if 1/3 evaluates to 0.)
> And if there is a technical reason it is simpler to require the
> variable, but no technical reason using the default variable can't be
> implemented, of course I would be very open to doing some of the work
> to make this possible.
>
> Perhaps this is a moot point for me personally if
> sage: integrate(x^3)
> remains functional. I really think it would be silly to require
> sage: integrate(x^3,x)
> regardless of what Wolfram or anybody else says.
> </mild-mannered rant>
>
> Notwithstanding all this, a fantastic thank you to Burcin et al. for
> making
> sage: f(x)=x^2
> sage: f(3)
> 9
> not take 30 seconds!
>
> - kcrisman
--~--~---------~--~----~------------~-------~--~----~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to
sage-devel-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://www.sagemath.org
-~----------~----~----~----~------~----~------~--~---
