@Preben Alsholm 

I think that there has been no notification from the administrators of an eventual change of default. Missing notification, I keep thinking that this absence of email notifications is just a bug, the same as so many times in the past.

@Mac Dude 

The main problem, in my opinion, is that Maple does plot Dirac as it were a function from reals to reals, e.g. try:

plot(Dirac(x),x=-1..1);

As I see it, such an input should return an error message. But the status of distributions/generalized functions is unclear in Maple...

@Carl Love 

If Maplesoft had a reason for not doing it that way (less likely), and that reason and the consequent behavior were not documented, it would be a documentation bug. And if there were no such reason (more likely), it would be a code bug.

@ecterrab 

A semantic issue... Note that I have used quotes. How do you say we should call these two types of updates for distinguishing them? General and R&D?

@Mac Dude 

The reason is that these tilded names are local:

> indets(sol,'`local`');
                                     {_Z2~}

So, for the case of a single assumed name as here, this form of the two-argument eval statement is more robust:

> eval(sol,op(indets(sol,'`local`'))=3);
                                          c exp(- a/b)
                          b LambertW(3, - ------------) + a
                                               b
                    exp(- ---------------------------------)
                                          b

becoming independent of the details of the sorting algorithm for sets, which may change with versions, as Markiyan has shown.

Now, the deeper reason is the implementation of the assume facility, which is actually a patch on the original design of the Maple system. This implementation using local names has a large number of drawbacks, and has been the headache of users for many years. But it is better than nothing. A "local optimum" as Jacques has described it time ago.

@oldstudent 

Menu Tools > Check downloads standard dot updates when they are "officially" released, as I have described before. The R&D updates that Edgardo has described here are a different thing, more experimental if you whish, and have to be downloaded and installed "by hand". I hope that in the future, Maple includes a package manager like facility so that rolling mla libraries could be installed more systematically (actually meaning that more parts of the system become updated in similarly agile fashion). 

@oldstudent 

Maple 18.01 was released in May. I guess that the next dot release, if it exists, will come after the northern summer.

@ecterrab 

Yes, not what I have asked, but your comments are most interesting. Indeed, I have had diverse problems with current definitions, so that standardization after DLMF will be very welcome.

One of the keys (a new one) in the table MathematicalFunctions:-Topic (or one of the names in the output of FunctionAdvisor(topics)) is "periodicity". However, I get, e.g.:

> FunctionAdvisor(periodicity,sin(x));
The location of the "periodicity" for sin(x) is unknown to the FunctionAdvisor

What is the status of this topic?

@Kitonum 

The difference is that the procedure `<,>` is executed only when V2 is constructed:

> trace(`<,>`):
> V1:=Vector([a, b]);  V2:=<a, b>;  # Identical results
                                         [a]
                                   V1 := [ ]
                                         [b]
execute <,>, args = a, b
                                         [a]
                                   V2 := [ ]
                                         [b]
> A:=subs({a=[1, 2],b=[3, 4]}, V1); B:=subs({a=[1, 2],b=[3, 4]}, V2); # Also identical results
                                      [[1, 2]]
                                 A := [      ]
                                      [[3, 4]]
                                      [[1, 2]]
                                 B := [      ]
                                      [[3, 4]]

@oldstudent 

It is very simple, for instance using the two-argument form of eval:

> s:=solve({x >= 0, 2*sin(2*x) = 1, x <= 2*Pi}, AllSolutions, Explicit);
                        Pi         13 Pi        5 Pi        17 Pi
             s := {x = ----}, {x = -----}, {x = ----}, {x = -----}
                        12          12           12          12
> eval(s,Pi=180);
                    {x = 15}, {x = 195}, {x = 75}, {x = 255}

(this particular syntax works from Maple 13 on). Menus provide a very restricted interface to the functionality of the system.

@oldstudent 

Actually, the problem in Maple 17 and 18 occurs when using "degree" unit, hence in the case of the statement as shown by Kitonium. However, no such problem occurs when using "radian" unit, hence solving 2*sin(2 x)=1 for the interval 0<=x<=2*Pi:

> solve({x >= 0, 2*sin(2*x) = 1, x <= 2*Pi}, AllSolutions, Explicit);
                      Pi         13 Pi        5 Pi        17 Pi
                {x = ----}, {x = -----}, {x = ----}, {x = -----}
                      12          12           12          12

Tracing shows that the computation path is quite different in these two cases. About bug fixing, the time frame, in my experience, is very dependent on the problem. May be anything from days to years, depending on a variety of factors like whether there is a developer that happens to be working on the subject at the time of the report, company priorities, etc.

@oldstudent 

Kitonium result occurs in Maple 16.02 and some earlier versions. The error that you get is observable already in Maple 17.02, where a change in algorithms was introduced, apparently causing this regression bug.

@acer 

My experience with Möbius Apps including components like sliders (e.g. your link above "Vertex of a Parabola") is also that they do not work. Typically, they end up in the browser (Firefox) becoming badly non-responsive. I have reported on some of the problems that I have observed in this thread.

Edgardo, look at the linked file name for download (version Maple 18). It looks wrong at this moment.