@ecterrab 

Yes, this is the Bryon's cat effect: a superposition of above and below states :)

@josephap83 

Good that Charles has addressed already your question. The administrators of this site have removed all the subscriptions to threads previous to the reform of early June 2010. So, I have not received any email notification about your comment...

Certainly, the standard way to make these vector/tensor calculations in Euclidean flat space is using an orthonormal basis, and this is what formulas in the literature show. For cartesian coordinates, as in the example of April 2010, the coordinate basis is also orthonormal, so there was no need to go into this distinction. But for curvilinear coordinates, they are different, so care has to be taken...

Note also that, since 2010, some further developments in the area of tensor calculus have occured in Maple.

It is this site's developers who follow the layout of the stackoverflow site, not "we" the users of this site. I rather avoid it.

This recent question about a three-years old thread, where I have written some comments, prompted me rereading it. So, I have found that one of my comments was broken: a typeset equation was converted to ordinary text and a picture was missing. Thinking about repairing it I have looked at my local archived copy, an html file (in primes1 html file archiving was fine, but currently is broken). Opened it, and with surprise, I saw the equation nicely typeset. I have looked at the source code of this html file and found an URL to this gif file in the maplenet area.

Sounds like that, after over a year, there are still more missing files hidden out there that could be recovered from the disaster...

@Will

I can tell the files weren't re-uploaded because files in the new primes, would have a URL like http://www.mapleprimes.com/view.aspx?sf=.... However, Jakubi's files have this url http://www.mapleprimes.com/files/143_Surfplot.lib, which is from the old Primes.

Certainly, the format of the links imply that those files were uploaded in primes 1, but you cannot conclude from that fact that those files were not re-uploaded. In fact they were re-uploaded. And from the date you should know whether it was primes 1 or 2 at that moment (May 30). I am almost sure that it was primes 1. Indeed, the first threads that I have saved locally from primes 2 are dated June 3.

And clearly the format of the article does not suggest anyway that "it was a copy/paste from an earlier post", hence "it wasn't created by re-uploading the files". For sure, it was created by re-uploading the files. I would have expected that you base your statements by looking at the log of my activity in the forum that day.

@Christopher2222 

There is a Spanish saying: "La tercera es la vencida" (in English). I wonder whether it will be true. So it goes:

surfplot.zip

This zip file contains: Surfplot.ind, Surfplot.lib and 8933_data2.mw.

By the way, I have got an error message when trying to upload Surfplot.lib:

You cannot upload a .lib file.

Try again

It seems strange to me that this extension is not valid in a Maple forum, even when it corresponds to an old version of a library file.

@Christopher2222 

Yes, I have checked after upload that those links worked. It would be very interesting to know an explanation on how they were also broken. I wonder whether it makes sense to upload those files a third time. Or perhaps it is better sending them to you by private mail.

Christopher2222

A month ago, or so, Robert Israel noted the same problem with those links in a post to another thread. Late at night I have uploaded again those files and posted the new links in a comment, but most likely I have not archived that thread. And sadly, primes 2 does is not providing a list of my comments nor a list of my uploaded files. And a useful search facility by date is missing. So, those links are somewhere, but who knows where?

May be that Robert has a clue. If not, may be that you are not lucky today...

Not only that. The logical structure of this thread is completely upset. This was its branching "skeleton" after my local backup version, where tabs stand for former indenting:

Arrays Vs Matrices
Comment on 2007-06-29 10:50 From dcasimir

        A question of dimensionality
        Comment on 2007-06-29 11:31 From John Fredsted

        rtables
        Comment on 2007-06-29 11:43 From acer

                Thanks to both of you.
                Comment on 2007-06-30 18:38 From dcasimir

Did you mean array vs matrix, or Array vs Matrix?
Comment on 2007-06-29 11:13 From acer

        What's The Difference ?
        Comment on 2007-06-30 18:20 From dcasimir

                arrary vs Array, matrix vs Matrix
                Comment on 2007-06-30 19:30 From acer

Why not permute directly with the Array constructor?
Comment on 2007-06-29 11:30 From acer

        Interesting, but seemingly problematic
        Comment on 2007-06-29 11:37 From John Fredsted

                permute the index ranges as well
                Comment on 2007-06-29 11:48 From acer

                        Impressive, again
                        Comment on 2007-06-29 11:55 From John Fredsted

                                ArrayTools
                                Comment on 2007-06-29 12:36 From acer

        Oh boy, I'm sleeping
        Comment on 2007-06-29 11:39 From John Fredsted

        Impressive
        Comment on 2007-06-29 11:48 From John Fredsted

Observe also their submission times. Apparently, in the current version, comments are ordered after their timestamps, irrespective of their position in the tree. As the parent-child relationship is missing in many cases, I find that reconstructing the logic of this thread by reading this version is very hard, if not impossible. In this sense, I think that many legacy threads like this one were spoiled.

Fortunately, I have just observed that a spare copy is available in the web archive (though the broken link seems also broken there)

This question was made in the context of acer's post, in whose example kernelopts(Evaluator) is bound to a procedure:

kernelopts(Evaluator=proc(x::uneval) x=eval(x); end proc):
kernelopts(Evaluator);

    kernelopts(Evaluator) = (proc(x::uneval) x = eval(x) end proc)

Clearly, NULL cannot be its default value:

restart;
kernelopts(Evaluator=NULL);
a:=1;

                                a :=

restart:#recovering default
a:=1;
                                a := 1

It sounds to me like the default value should be a procedure (or something that evaluates to that). Clearly, any disruption could be done by binding nonsense:

restart;
kernelopts(Evaluator=4);
a:=1;

                                a := 4
kernelopts(Evaluator);
                                  4

In Maple 13.02, with gfun 3.52, I get:

myHeunD := diffeqtoproc(deq[Heun], y(z)) ;
  myHeunD := proc(path, prec := Digits)
local disk, center, arg, rad, local_exp, res;
    if prec <= 0 and type(path, 'complex'('numeric')) then
        arg := convert(path, 'rational', 'exact');
        for disk in [] do
            center, rad, local_exp := op(disk);
            if signum(abs(arg - center) - rad) < 0 then
                userinfo(4, 'gfun', sprintf(
                    "using precomputed series at %a", center))
                    ;
                res := eval(local_exp, z = arg);
                return nthterm:-makeitfloat(res, prec)
            end if
        end do
    end if;
    userinfo(2, 'gfun',
        "using multiprecision analytic continuation");
    ancont:-analytic_continuation({diff(diff(y(z), z), z) -
        (-2*z^5 + 4*z^3 + z^4 - 2*z - 1)*diff(y(z), z)/(
        (z - 1)^3*(z + 1)^3)
         - (-1/3*z^2 - 5/2*z - 3)*y(z)/((z - 1)^3*(z + 1)^3),
        y(0) = 1, D(y)(0) = 0}, y(z), path, prec)
end proc

A reason to upgrade?

I see that Maple 13.02 version of the PolynomialTools package exports FromCoefficientList, while the Maple 12.02 version does not. Actually, in Maple 12 I have installed gfun 3.19 which does not include NumGfun, but just because I have not tried updating it.

As I see it, the problem arises from int:

restart:interface(version); Digits:=14:
  Classic Worksheet Interface, Maple 14.00, Windows, Apr 5 2010 Build ID 479326

f:= x -> (x+1)^(x+1):
F0:=x->evalf(Int(f(t),t=0..x)):
F:=unapply(evalf(int(f(x),x)),x):
F0(0)=F(0);

                               0. = 0.

F0(1)=F(1);

                      2.050446235 = 2.333333333

It should be traced to spot the origin of this problem.

Happy to have helped a bit with the recovery.

@bthur 

Were old polls pages been removed? I have just been loking for the current page of an old poll "Where should Maplesoft direct its technical focus?", whose page I have archived as development-focus.htm, but I have not found it in search yet. It contains interesting comments.