@Christopher2222 

As I see it, geting back at least some of the things that were better in the old primes will require a new administration. But, I am afraid, much of the damage is irreversible.

@Mac Dude 

The issue is that no "application" in the Application Center is a replacement for a missing integrated (bundled) and maintained tool. It implies, following my example of the Fourier series, that no facility in shipped Maple can build upon the routines computing these series, contained in a package uploaded there, like the one you have mentioned. So, either this facility gets stuck, or its developer has to write private (internal) code for the computation of these series, as it was the case of the PDE solver facility (pdsolve). Of course, it is there but users have no access to this tool. And there might be more than one independently developed private version, who knows? A situation that makes no sense, making maintainance more complicated. And this is just one tiny example out of many.

@Mac Dude 

See e.g. here. An excerpt:

The Maple V Share Library was a wonderful facility for mathematicians comprising some 140 Maple routines, packages and worksheets written by Maple users and contributed freely to the Maple community. It used to be bundled as part of the Maple package but that practice ceased with Maple 6. Only some 25% of this vast corpus of knowledge is currently available via the Maplesoft Application Center. The remaining 75% has effectively been "lost" to mathematicians—until now!

@dohashi 

This is somewhat more clear. Now, earlier you have used the expression "controler kernel". Is what you call now "main kernel" the same? If I follow your terminology correctly, both cases are local grid (always with ordinary Maple), so distinctive names for them seem missing. Certainly, such an explanation should be included in the documentation of the Grid package, and I hope that you would not call them case 1 and case 2.

Also, the conclusion of an earlier part of this discussion about the local mode was that in a machine with N cpus/threads (N=8 in my example), there are N+1 kernels, the controler and a grid kernel in one cpu, and a grid kernel in each of the other cpus. But this was the case of the controler (main?) kernel outside the grid, apparently case 1 only. So, what about the case 2? N kernels?

@Mac Dude 

Certainly, the Application Center is somewhat organized, but not about the maintainance of the applications and their integration with the main system. So, this is a step back wrt the old Share Library. About actual developments and its market, Maplesoft claims that Maple has a O(10^6) user base. And there are currently several open source CAS projects nowadays (Maxima, Sage, Axiom(s), Reduce, etc) which have most likely a much smaller user base and somehow they progress, in cases rather well. So, I cannot believe that there could be any shortage of human resources within the Maple user community.

Note that at the same time its own developers are highly stretched, frequently involved in very sophisticated or fancy projects. But at the same time you may find critical or basic areas like symbolic integration that are almost frozen for many years (and count the number of bugs reported here). And many other areas, even basic ones, that were never developed (e.g. there is no facility for Fourier series). So, yes, I think that there are resources and market within the community. The problem is rather that Maplesoft's paternalistic policy disincentivizes such activities.

@Carl Love 

No! The StackExchange way was what killed the spirit of this site. In StackExchange way, discussions are forbidden, and there is a censorship system by which a "police" may close a thread when he considers that there is a discussion, a fully subjective criterion. In StackExchange way, everything is constrained to an artificial classification: e.g. a question or an answer, so that there is no room for posts that contains questions answers and comments all together as in any normal exchange. In StackExchange way answerers compete for votes, instead of elaborate together better knowledge. Frankly, the StackExchange way is for narrow minded persons, I do not want it.

@Christopher2222 

As defined here, a mindset is not "based" on a site as Stack Exchange, but it is a preexisting reason. A way of thinking that made Bryon to find that sorting feature as good and decide copying it, even when no user had asked it, and even when it was evidently bad for this site. In other words, I do not understand his way of thinking.

@Axel Vogt 

Not so long ago, Bryon said that he still sees a value in sorting by ratings (hence the default, I think). I cannot understand his mindset.

@Axel Vogt 

Yes, I keep observing this no update after post bug quite frequently, though not always.

@Mac Dude 

On my side, my general attitude is paying attention only to posts that show text input. Something that can be copy&pasted. Typically, I leave aside posts that show only images or only links to an uploaded worksheet as I do not have time to retype input or download worksheets to know what is the issue.

During its first epoch, basically the Gaston Gonnet time, Maple was somewhat more open towards community contributions and included the Share Library. Certainly it was not open enough, but it was something. But during the current administration, the system became more closed, sadly. The main problem with single author repositories is maintainance: in the short or long run it cannot be kept updated. The only solution is a collaborative framework as typical in opensource projects. Unlikely to be set up by Maplesoft, as I see it. And frankly, I would not rely on it, even if it were set, as it would be maintained the same as this site...

@dohashi 

Yes, indeed there is a terminology issue here (or more than one). Possibly one of the consequences of your introduction of the subject of the Grid Computing Toolbox to this thread, which was not the subject for the OP, nor probably for the rest of posters (certainly not for me). By the way, it is obvious to see that the same kind of terminology problem occurs as a consequence of the unnecessary introduction of such a subject into the Grid package documentation.

So, we have to go back through your statements, trying to extract the information relevant to running in a single machine as it is all mixed. You have described here two cases:

The paragraph is a little confusing.  The "called outside the grid, on a non-compute node, not as part of a parallel computation" is basically defining two cases, one where the kernel calling Map is outside of the Grid, or the kernel calling Map is part of the Grid.  If the kernel is outside of the grid, then Map calls Launch to create nodes to compute the Map command and the result is returned to the calling kernel.  This is the case that most users want to use.  This is also the case shown in the first few examples.  The second case is intended to be used with the Grid Toolbox (although it does work without it) where there are kernels running on multiple machines managed by a system external to Maple.  What this case does is really just for sophisticated users who are writing complex Grid code. The last example shows this case, although it may not do a very good job of illustrating the difference.

The first one is the local mode with the controler kernel outside the grid, that I have made the effort to clarify earlier in this thread. So the second mode is still the problem. If you say that "it does work without [the Grid Toolbox]", the first consequence is that this second mode works in a single machine with standard Maple, OK? Then you say that "The last example [of ?Grid,Map] shows this case". And this example shows the usage of the command Launch, isn't it?

Now, in your  subsequent message you say, as a "clarification":

The Grid package is the api for both local and distributed Grid.  However when programming for the distributed Grid, functions like Launch, Send and Receive become much more important.

The normal way of understanding those paragraphs is that the second case is about "distributed grid". So, if now you say that it is not, it is completely unclear what you meant by the second case.

@itsme 

The first point to note is that such collections are, as you may imagine, work in "unending" progress. And consequently, what would make sense is their development in a collaborative fashion. But then the problem is of context. As things stand, Maplesoft does not support (provide a framework for, or perhaps like) opensource/collaborative development around Maple. So, such development has to be carried out independently.

If you search this site, you will find several posts in the past about independent development initiatives. As far as I know, none of them went far (or actually started). Somehow, needed agreements do not seem to be reached easily among members of the Maple community.  Also, years ago, I have lobbied for a Maple wiki site, but Maplesoft did nothing. Some day, Alec set up one of his own but rather than greeting it, as I would have expected, it received hostile consideration from Maplesoft and users side, plus actual atacks.

So, agreement should be reached on an independent, stable and secure site for such development. Not so simple as you may see...

Older posts still have incorrect dates. E.g. look at this thread. Comments are wrongly dated May/June 2010, as the thread is of February 2008.

@Mac Dude 

Actually, I do not type such rules everytime I need them. That way might not be much more efficient than typing the desired result itself! Rather, I have a collection of them in a text file and just read it to the session by a command. More correctly, I have several files like that, each for a different area were obvious or needed functionality is missing in Maple.

@mehdi jafari 

As I read the OP's description, it looks more like an interface/configuration problem. May be a problem with missing fonts, or rendering them, etc. Perhaps a screenshot plus system details may be more informative than a worksheet.