Perhaps you are looking for the command evala which stand for "evaluate in an algebraic number field." To get Cardano up and running, set the environment variable _EnvExplicit := true; and solve a cubic or quartic.

Indeed it is clever. Since Q(dog), Q(cat), Q(here), Q(there) and Q(everywhere) are all zero, we get an extended meaning of "zero almost everywhere."

The function Q above tells me that Euler's constant gamma, Pi^exp(1), exp(1)^Pi, Zeta(3), etc are all irrational. Probably they are, but some of these are still on the undecided list. So I do not regard Q as a true representation of the characteristic function of the rationals. It just some sort of fake thing that it trained to say that its integral over a interval is zero.

Try something like conjugate(foo); If you are using point-and-click, good luck. LaTeX has solved all these issues years ago, in a superior fashion. Would it not be nice to have synchronized Maple/LaTeX syntax? I propose $\overline{foo}$ or maybe unlatex(\overline{foo});

Yes of course you are correct, Int(f(t),t=a..x) is a function of two variables "a" and "x", not one. But let me correct my statement: F:=proc(x) Int(f(t),t=0..x) end; Diff(F(x),x); Note I replaced the lower limit by "0" instead of "a", and Maple still uses partial derivative notation instead of d/dx, so my question still stands. Reminds me of a scene in the movie "The Gods Must Be Crazy."

It seems the problem has to do with the ambiguous definition sqrt(z) for z complex. The square root function is not well-defined over the complex domain. You need a branch cut. If you replace sqrt in your eqn with -sqrt, then the two solutions are solutions. So in your original equation, it seems the two "solutions" are in fact solutions, if you appropriately select the branch cut for sqrt! Or to say this in more standard terms, solve is giving "extraneous" solutions. One way to get partial results is to replace sqrt(z) with z^(1/2) and assume real: eqn := 1/2*((b^2+a^2) +((b^2+a^2)^2-4*a^2*c^2))^(1/2)-c: S:=[solve(eqn,c) assuming real]; subs(c=S[1],eqn):radsimp(%); subs(c=S[2],eqn);radsimp(%); There are still rough edges, but this is not a bug. Rather the peculiar behavior has to do with vague assumptions on the range of values for a, b and c, and the vagueness of sqrt(z) for z complex. Try this: eqn:=sqrt(x+18) = z; solve(eqn,x); Is the "solution" x=-18+z^2 a solution? It depends on z, or what branch of sqrt your are using. Yet this is not a bug.

You might try to get rid of the do loops, and construct your matrices along these lines: d:= (a,b) -> if a=b then 1 else 0 end if; lambda1:=proc(N) proc(i,j); Matrix(N,N,(mu,nu)->d(j,mu)*d(i,nu)+d(j,nu)*d(i,mu)); end; end; lambda1(3)(2,2); lambda2:=proc(N) proc(i,j); Matrix(N,N,(mu,nu)->-I*(d(i,mu)*d(j,nu)-d(i,nu)*d(j,mu))); end; end; lambda2(3)(1,2); lambda3:=proc(N) proc(n) sqrt(2/(n^2-n))*LinearAlgebra[DiagonalMatrix]([seq(1,i=1..n-1),-(n-1),seq(0,i=N-n)]); end; end; lambda3(3)(2);

It seems that you must remove the integral "boundary value" from dsolve, then follow up by using solve (not dsolve) on the result, with the integral boundary value. For example: > restart; sys:={diff(f(x),x,x)=a,f(0)=1}; > soln:=dsolve(sys); > assign(soln); > f(x); > solve(int(f(x),x=0..1)=1,_C1); The point is that your integral equation is not a boundary value, it more like is a "global value."

Ray, We use a Windows server. I would have to ask our system administrator for any details. I will pursue this if you ask me to do so. The browser I used when I tested your question bank was Firefox. I installed your question bank at this link http://mapleta.math.uwec.edu:81/classes/HofackerSmith/ Register as a student into the class and see how fast or slow it is for you, and let me know how it goes.

I loaded your question bank and tested it out on my institution's installation of MapleTA 2.51 and I did not find it at all slow. Everything worked reasonably fast. I frequently use maple("printf(MathML:-ExportPresentation($f))"); and have never thought it made things run slow. I bet you will find the problem is with your server. Let me know if you want to test how it works from my institution, on the other side of the earth from you.

Try latex(Diff(x^2,x));

Your exact objective is not too clear, but the basic missing idea seems to be "plots[display]". You want something like this: $f=maple("proc(x) x+1+stats[random,normald]()"); $g=plotmaple("plots[display]([plot(x+1,x=0..1,thickness=2),plot($f,0..1,style=point)])"); The command "plots[display]" allows you to display different graphs together. Nice touches such as text, gridlines, symbol style, etc are much easier to deal with in Maple11 than in Maple10. Our institution has Maple11, but MapleTA 3.0 has not yet arrived. (MapleTA 3.0 is perpetually "shipping soon.") So I am temporarily/perpetually making graphs with Maple11, using many of its bells and whistles, and exporting the graphs as gif files for inclusion in MapleTA 2.51 questions. I wish the release of Maple 11 and MapleTA 3.0 had been timed better! When we finally do get 3.0, I will go back and modify the TA question so that the graphs are generated on the fly with Maple, rather than linking to a gif.

One thing I have never been clear on is the correlation between the version of MapleTA being used and the the assumed version of Maple. For example, MapleTA 2.** seems compatible with Maple10.**, but not with Maple11.**. But I cannot figure out how to verify the correlation. The plotting tasks you ask about seems more straightforward in Maple11, but it is not clear what version of MapleTA you are using. Are you using MapleTA3.0?

If you assigned n, for example n:=5; then deleting the assignment from the worksheet does not remove the assignment from memory. You need to do something like n:='n'; or restart;

For starters, there is the "dot" operator a.b which is supposed to be left associative and non-commutative. But it's just a start.