I normally don't recommend building the plot data structure from scratch, for a number of good reasons. However, in this particular case, it is best to bypass the pointplot3d code and to send the structure directly to the GUI. The library code does various type checks and conversions that are unnecessary if you are absolutely sure your input is correct. For a large plot like this one, use > PLOT3D(POINTS(A), SYMBOL(POINT), AXES(BOXED)); We are hoping, in the future, to offer library routines or options that will allow you to avoid such difficulties with very large plots. Paulina

I normally don't recommend building the plot data structure from scratch, for a number of good reasons. However, in this particular case, it is best to bypass the pointplot3d code and to send the structure directly to the GUI. The library code does various type checks and conversions that are unnecessary if you are absolutely sure your input is correct. For a large plot like this one, use > PLOT3D(POINTS(A), SYMBOL(POINT), AXES(BOXED)); We are hoping, in the future, to offer library routines or options that will allow you to avoid such difficulties with very large plots. Paulina

Yes, the parsed input is saved in the worksheet but it's safer to use the Typesetting[Settings] command. If the user of the worksheet re-executes it, then the user's default settings are applied and the original parsed meaning is lost. Paulina

Acer, if you are creating a worksheet and using a non-default parsing rule, then I would suggest adding the rule to the worksheet with Typesetting[Settings]. For this particular case, you'd add the command Typesetting[Settings](numberfunctions=true). Paulina

Sorry, I didn't realize you were using the Statistics package and thought ErrorPlot was your own command. My message referred to the results you'd get simply using the plot() command. Anyhow, this looks like a problem in ErrorPlot (it needs to be updated to accept strings) and we'll try to get this fixed for a future release. Paulina

I'm a little bit confused myself. The only error I get with Maple 11 is with color='Blue', and this is expected. The plot routine does not ignore the color option; instead, it gives the error that 'Blue' is an invalid color. For many versions of Maple, we had 16 available colours, and these were specified as names, not strings. In Maple 11, when we added a large number of new colours, we were concerned that name conflicts would be a problem. Users don't always remember to quote names and colour names like 'Peru' or 'Seashell' might be used in other contexts. Thus, we decided to move to strings for the new names and added string equivalents for the old names as well. To avoid confusion, it's best, in the future, to just stick to using the strings shown on the plot/colornames help page. Of course, the old names will still be accepted for backward compatibility. I regret that we've not had a chance to update all our examples to use the new names, but we are working on that. Paulina

The internal representation of typeset expressions is not documented in Maple. Working with typesetting from the command-line is not a feature we currently support, though we may one day. Note that we're continuing to improve the typesetting system from release to release, so there's a chance the underlying representation will change. Having said that, there are some aspects of the typesetting system that you can control programmatically. These are documented on the Typesetting help page. For those who enjoy using the mouse more than Jacques does, these same features are available through the Typesetting Assistant (see "Typesetting Rules" in the View menu). Paulina

I've verified that the first problem is a bug and have entered this in our database; we'll try to get it fixed in a future release. I couldn't reproduce the second problem but I've sent this over to our GUI people for investigation. Paulina

About your request, acer, to have the atomic identifier information more widely referenced in the help pages .... To be frank, it's a lot easier to add a few thoughts to the blog than to update the help pages, as the latter requires careful editing, testing, etc. and involves more than just the developer's time. However, we recognize that users sometimes have trouble navigating the help system. It is my hope that all the blog content will eventually make it into the help pages. Actually, the content is already there, but we can do better in making it well organized and easy to find. So these blog entries are not meant to be a replacement for help pages but rather a step towards making the help system better in the long run. Paulina

Your question about how to avoid evaluation of expressions in 2-D captions seemed like a good topic for the next blog entry. See my answer here. Paulina

acer writes:

ps. The documentation makes it appear that only symmetric/hermitian floating-point sparse systems can be solved with iterative methods. Yet trying it with real nonsymmetric float[8] sparse Matrices shows via userinfo that there are also distinct nonsymmetric solvers. All the sparse floating-point methods show NAG function names through userinfo, ie. with infolevel[LinearAlgebra] > 0 .

Yes, there is a conjugate-gradient-squared (CGS) method for nonsymmetric floating-point sparse linear systems. Unfortunately, mention of this algorithm somehow got left out of the help pages but we'll try to remedy this in an upcoming release. Paulina Chin Maplesoft

Evgeni writes:

I do not know still what should I enter into LSSolve to get the vector of residuals.

LSSolve will not return the residuals directly, but you can compute them using the code suggested by Robert Israel. The Statistics commands will return the residuals if you use the 'output' option. For examples of how to do linear fitting with the Statistics package, see the Statistics[PolynomialFit] and Statistics[LinearFit] help pages. These pages include a couple of examples that use the 'output' option. We are working on providing more examples for our users. You can also try looking at the Task Templates and example worksheets for other ideas. Paulina Chin Maplesoft

While you can use Optimization[LSSolve] directly, consider also using the commands in the Statstics package. For example, calling Statistics[LinearFit] with the output=solutionmodule option gives you access to a variety of information, including the residuals. See the Statistics/Regression/Solution help page for more details. The Statistics package also includes a number of useful visualization commands. (See the Statistics/Visualization help page.) The ScatterPlot command allows you to fit a curve to a set of points and easily plot the points and curve together. Paulina Chin Maplesoft

In Maple 10.03, you can also use command completion to insert a name with an arrow over it. (Issue ?commandcompletion to get help on command completion.) In 2-D math input mode, enter Vec and invoke command completion. Among the "Vector" entries is a template that consists of a name with an arrow over it. After you insert this into your worksheet or document, you can replace the name with one of your choice. Paulina Chin Maplesoft

In Maple 10.03, you can also use command completion to insert a name with an arrow over it. (Issue ?commandcompletion to get help on command completion.) In 2-D math input mode, enter Vec and invoke command completion. Among the "Vector" entries is a template that consists of a name with an arrow over it. After you insert this into your worksheet or document, you can replace the name with one of your choice. Paulina Chin Maplesoft