[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Refcard
From: |
Jake |
Subject: |
Re: Refcard |
Date: |
Thu, 1 Jul 2010 14:40:50 -0400 |
I agree on all points Rik; that's a nice summary. The only thing I can add to
that is mentioning xlsread as well as csvread, since there is a decent need to
use that function introductory MATLAB classes. I might add that I'm very glad
that someone took the initiative to make a real xlsread function instead of the
tab delimited file reader it was a year ago.
On Thu, 01 Jul 2010 11:03:13 -0700
Rik <address@hidden> wrote:
>
> > Subject:
> > Refcard
> > From:
> > Jake <address@hidden>
> > Date:
> > Thu, 1 Jul 2010 08:24:54 -0400
> > To:
> > address@hidden
> >
> > To:
> > address@hidden
> >
> >
> >
> > Reference: https://savannah.gnu.org/bugs/?30260
> >
> > The bug started with just some spelling corrections, but there has been
> > some discussion about missing functions and the implications of
> > adding/subtracting content (messes up alignment, etc). I was thinking that
> > the very basic function fzero() should be added somewhere to the refcard.
> > The refcard is included with the source. Please discuss.
> >
> 6/1/10
>
> Jake,
>
> I agree with most of the discussion on the bug list. I find that I never
> use Info for reading the documentation, preferring either the pdf or HTML
> versions instead. I believe we could reduce what amounts to a tutorial on
> Info, and replace it with more information on Octave itself.
>
> The sections I see that could be removed are [Motion in Info, Node
> Selection in Info, Searching in Info]. Instead, I might just list the
> command to call up help on Info once they have launched it.
>
> I also think the command-line editing options are less useful than more
> documentation about Octave itself. If the user is a beginner then they can
> get by with using cursor keys, etc. On the other hand, if the user is an
> expert, they probably already know, and won't forget commands, like
> transposing characters and yanking words. Another issue is that key
> bindings are not stable in the same way that the calling parameters to fft
> are stable. I use VI bindings so all of this information is just plain
> wrong. Other systems may have customized certain key combinations which
> means users will be perplexed when they try a suggested command and it
> doesn't work. The sections that I see could be eliminated are
> [Command-Line Cursor Motion, Inserting or Changing Text, Killing and
> Yanking, and perhaps parts of Command Completion and History].
>
> In [Index Expressions] there should be a mention of the special variable
> 'end' as in 1:end.
>
> In [Linear Algebra] there should be mention of 'eigs', used to find just a
> few eigenvalues of a sparse matrix.
>
> I'm not sure of the exact section to place these in, but the new
> optimization functions fzero, fminbnd, and fminunc should be added.
>
> I think the section [C-style Input and Output] could be shortened. We
> should be encouraging the easiest way to get data into Octave which is
> often just load which works for simple columns of data. We also need to
> mention the newly added csvread and csvwrite.
>
> In [Selected Built-In Functions], I would mention that 1i, 1j are both
> equivalent to the sqrt(-1). I would also put a reference to e, the base of
> natural logarithms.
>
> The section [Cells and Structures] should mention structfun. Somewhere,
> maybe here, there should be a reference to arrayfun.
>
> [Plotting Functions] should have some mention of the new graphics handles.
> Or maybe this should be its own section with some of the other commands
> like figure, clf, gca, gcf. I would also add surf to the list of plotting
> commands.
>
> --Rik
>
--
Jake <address@hidden>