roman_pearce

Mr. Roman Pearce

1688 Reputation

19 Badges

20 years, 82 days
CECM/SFU
Research Associate
Abbotsford, British Columbia, Canada

I am a research associate at Simon Fraser University and a member of the Computer Algebra Group at the CECM.

MaplePrimes Activity


These are replies submitted by roman_pearce

There actually is functionality available in the software running MaplePrimes to being posts in from other websites. The idea of bringing in the content of of the Usenet and Yahoo groups is a good one, and we will consider it. I'm going to cast a big "no" vote for this idea, at least for pulling in threads from Usenet. The goals of this site are fundamentally incompatible with an unmoderated newsgroup, and you will expose yourself to the censorship issue if you try to merge the two. That, and you will make this site a target of trolls.
I've noticed the relatively light traffic on this site as well. The bottom line is that we need more Maple content. This is what I would do: - Publish a blog item for every 5 submissions to the Maple application center, with a one sentence summary of each application and a link. - Publish non-marketing articles from the Maple reporter newsletter. Try not to annoy people with shameless self-promotion :) - Publish papers from last year's Maple conference. Maplesoft owns the rights to these papers, and by my estimation you could do one every week. Contact the authors and ask if they would be willing to contribute a blog entry, perhaps with a link to Maple code. This would also give them a chance to elaborate on how the work has developed, without the tedium of writing another paper. I for one would find this very interesting, and I guarantee you could lure in a great deal of the serious Maple community with free papers. Do it every Friday so people have something interesting to read when they aren't overly busy. - Once you've built up traffic, ask people who are submitting a paper or a poster for this year's Maple conference to post a blog entry discussing their work. This is a great way to get advance word out and increase interest. At the very least, any student will bite. It's too early to do this now. Do it in the 3 months or so leading up to the conference. The fact is that Maplesoft already has a lot of strong resources. The Application center gets over 10 submissions a month, and the Maple conference published a 500 page proceedings last year (poor Ilias!) You need to leverage these resources to make this site a success.
Well I guess that explains all those bugs in my programs :)
Well I guess that explains all those bugs in my programs :)
People are going to laugh, but I code everything in pico (or GNU nano), one file per package. There are ways to split the code up into a main file which loads other files (look for the mload command) however that has the disadvantage that you can't update the code while in Maple (say when you are tracking down a bug). Emacs is the way to go if you want syntax highlighting. As for OOP, the Maple object you want to use is a module. Modules can contain both data and procedures, either local or exported. It is the standard format for Maple packages (see the help page ?module). You can do a lot of neat things with modules, however if speed is a concern then you should use a flat datatype like a table. Consider this example: suppose I want to represent a graph with vertices connected by edges. In C++ a natural way to do this would be to make a vertex class which contains a linked list of neighbours. Then each vertex in the graph is an instance of the class. You could do this with Maple modules as well, but for the purposes of programming algorithms in the Maple language it probably won't be fast (you might say "OOPs!"). There is overhead to OOP which is now (thanks to good compilers and fast hardware) quite negligible in a compiled language, however it will still be felt in an interpreted language like Maple.
I wonder if the problem has something to do with tabbed browsing. When I visit MaplePrimes I usually go straight to "new content" and open everything new in its own tab.
When I wanted to simulate Maple input and output in my thesis, I used the following: \def\MapleInput#1{\noindent{{\small $>$ {\tt #1} }}} \def\MapleOutput#1{{\begin{center} #1\end{center}}} \def\MapleWarning#1{\noindent{{\small {\tt #1}}}} Here is an example: \MapleInput{f := x\char94 2 + 1;} \MapleOutput{$f := x^2+1$} The main irritation was having to use \char94 for ^ and \langle and \rangle for left and right angled brackets. Very weird when you are used to Maple notation. I would like to see the latex package extended to translate Maple input. I'm sure it already does somehow. That would be useful.
To answer 3), I had no problems installing Maple 10 on my Gentoo system which is NPTL, so there should be a way to get it to work. Unfortunately I can't give you any concrete suggestions, but I can tell you what I know. - Maple 10 uses Java 1.4, but it can be made to run with Java 1.5 if you install the Apache Crimson component. See this thread for details: http://beta.mapleprimes.com/forum/problem-with-maple-under-java-1-5-on-osx I would expect Maple to work if you can get it installed. - There used to be an option to install Maple from the command line. Does the installer has a --help option ? Can anyone from Maplesoft comment on this ? I think it would be the best way to go.
I'm currently using Safari 1.3 on OS X 10.4, but I've had the same thing happen on Konqueror 3.5 on Linux, Firefox 1.5 on OS X, and Firefox 1.5 on Windows. Thanks for looking into this, by the way.
By default using Maple isn't going to break it or slow it down permanently in any way. It is possible (with some effort) to modify the Maple library, and if any files are damaged (for whatever reason) then you should reinstall Maple. Otherwise, defragmenting your disk and making sure you have enough memory will give you the best performance. There is one thing which does affect Maple's performance over time, and that is garbage collection. Basically, it means that as you use more memory Maple starts to run slower. If you do big calculations for a long period of time and Maple starts to run slower, this is why. Unfortunately Maple won't free the memory that it uses unless you use the "restart" command - which kills all of your variables. One way to keep Maple running fast is to make garbage collection happen less often. By default on a 32-bit machine it happens every time Maple uses 4MB of memory. This keeps overall memory usage low. If you have memory to burn (ie: more than 512 MB) and you want more speed, you increase the amount of time between garbage collections. You do this by setting kernelopts(gcfreq=...); The default is 10^6. I timed solving a 64 x 64 linear system with 1200 digit numbers. This is what I got:
gcfreq    time    mem use
10^6     127.889    40MB
8*10^6   113.658    70MB
64*10^6  111.052   244MB
As you can see, a moderate increase in memory usage (for today's systems) gives you a good increase in time, but if you increase gcfreq too much then you will just end up wasting memory. Ultimately, the best setting depends on the nature of your problem. For some problems you will find that performance will keep increasing even with gcfreq settings above 64*10^6. For other problems the setting of gcfreq won't seem to matter. Best of luck, anyway.
I'm not sure what you mean. You can update the software by going to "Tools"->"Check for updates", the current version is 10.02. Is there a bug that needs to be fixed ?
You can select your text/math and use View->Expand Document Block to see what Maple has created. View->Collapse Document Block undoes it.
You can represent the root using RootOf(p, x). For example: p := x^8 + x^6 + x^3 + x^2 + 1; alpha := RootOf(p, x); Factor(x^2+x+1, alpha) mod 2; The second argument to Factor is to indicate an algebraic extension, so the factorization is computed in Z[2](alpha)[x]. If you want Maple to display alphas instead of RootOf(...) you should use the alias command instead of assigning to alpha. For example: p := x^8 + x^6 + x^3 + x^2 + 1; alias(alpha=RootOf(p, x)); Factor(x^2+x+1, alpha) mod 2;
The standard print dialog on the Mac has a "Save As PDF" button at the bottom - allowing every application to print to a pdf file instead of a printer. I copy the plot to a new worksheet or document (Document mode is nice for this) and then use File->Print.
It would be nice, but the proper way to support it is to have a pdf printer driver for the operating system. That way you don't have to build pdf support into every application. The Mac and Linux already have this, and there are lots of free drivers for Windows, like PDFCreator (gpl). See this page for a more comprehensive list.
First 33 34 35 36 37 38 39 Page 35 of 39