Prev: FindMinimum inside another FindMinimum
Next: Compile with Boolean Arrays
From: Sjoerd C. de Vries on 13 Aug 2010 06:55
On Aug 12, 11:31 am, Themis Matsoukas <tmatsou...(a)me.com> wrote:
> --A simple(r) way to change the working directory for input/output operat=
ions from the default (root) to the directory of the current notebook. The =
only way I know how is
>
> SetDirectory[ToFileName[
> Extract[
> "FileName" /. NotebookInformation[EvaluationNotebook[]], {1},
> FrontEnd`FileName]]]
>
> Being incapable of memorizing it, I save this monstrosity in a safe vault=
and recycle it from notebook to notebook.
>
> Themis

SetDirectory[NotebookDirectory[]] seems to do the trick...

Cheers -- Sjoerd

From: Hans Michel on 13 Aug 2010 06:56
If WRI Mathematica Notebook expressions were represented as a DTD then
The current representation would be
<!ELEMENT Notebook (Cell+)*>

I am just proposing:
<!ELEMENT Notebook (Compartment+|Cell+)*>
<!ELEMENT Compartment (Cell+)*>

Or the like (may not have the DTD declarations correct, just a hint). So
the introduction of this new expression would not disturb older files that
are a sequence of "Cell".

I don't think the problem is with the frontend interface handling too much,
but with market forces. Whatever happened to Display PostScript?

WRI solutions (Mathematica, FE, Workbench, webMathematica, Publicon,
Packlets, W|A,...) have been many things to many people, all of whom are
trying to learn, get work done, etc... What I put forward is not going to
cut the noise. For the lack of a good analogy to use I'll go with the
following:
In a complexly boisterously crowded but productive room with an open door
where the noise is overflowing into another room, a simple solution may be
to close the door. Closing the door is an additional action to something
that is productive just noisy.

What is the frontend missing I don't know (upgrade the Quiet function). We
were asked to suggest things that we would like to see in the frontend,
otherwise more noise and complexity to an already complex system. I have
many ideas on reducing the noise. Nevertheless, I was hoping to possibly
enhance composition from a batch to an interactive process by putting
forward a concept of compartments.

Hopefully WRI is working toward elegant solutions to the complexities of
Mathematica "System of ..."

Hans

-----Original Message-----
From: AES [mailto:siegman(a)stanford.edu]
Sent: Thursday, August 12, 2010 4:26 AM
Subject: Re: Frontend suggestions

In article <i3to0s$sao$1(a)smc.vnet.net>, "Hans Michel" <hmichel(a)cox.net>
wrote:

> The FE and the language itself needs a new expression between Notebook and
> Cell (above CellGroup which is still a Cell) we need Compartment.
> Compartments can exist side by side in the screen environment and content
> can overflow into the next linked Compartment.
>
> With Compartment you can now do interactive pagination.
>
> The FE still feels as if one is writing in Galley mode. Pagination is or
> feels batched. Is it a notebook or a scroll?
>
> So in version 8 or 9 or earlier I would like to see:
> Mathematica notebooks consist of lists of Compartments objects.
> New in 8 or 9 or earlier

It's hard to tell how this would work out in practice, but at least it's
a new, novel, and very clever and interesting suggestion, together with
some perceptive comments.

But at the same time, Mathematica's core and essentially unresolveable
problem is that it has already gone way, way beyond the size and level
of complexity that can be compressed into a single interface and learned
and managed and managed and used by the large number of individuals who
just want to use it to get work done in some much more limited area.

And no adding of additional capabilities and complexities is going to
resolve that.


From: Simon on 13 Aug 2010 06:57
On Aug 12, 7:26 pm, "S. B. Gray" <stev...(a)ROADRUNNER.COM> wrote:
> On 8/10/2010 12:55 AM, David Bailey wrote:
>
> > Since the subject of possible enhancements to the FE is being discussed=
,
> > I wonder if we could determine which feature would be most popular. I
> > have two requests:
>
> > 1) Infinite undo!
>
> > 2) Better documentation of functions such as Import that perform a
> > range of operations. So many details seem to slip between the cracks.
> > For example, nowhere does there seem to be a statement that .mid files
> > can be exported but not imported!
>
> > 3) A total ban on the word "typical" - it doesn't belong in
> > documentation that is supposed to be definitive!
>
> > Please add suggestions in replies and we can see which is the most popu=
lar.
>
> > David Bailey
>
> >http://www.dbaileyconsultancy.co.uk
>
> I vote for infinite undo, AND -
>
> 1. MUCH more rational behavior of text handling in general. Some of the
> things it does are either gross bugs or totally bizarre decisions. If
> anyone else cares I will list the ones that bug me the most. Given all
> the dumb things that it does I would never consider it for serious
> writing tasks.
>
> 2. Selection of noncontiguous text like MS Word (finally) does, so you
> can apply changes to a bunch of separated text all at once.
>
> 3. Better visibility of cell properties for example non-evaluatable,
> text cells, etc.
>
> 4. Better visibility of found strings during searches. I waste time
> trying to find the things it just found. Add backwards searches and
> better notification of what it's doing.
>
> 5. An easy way to have Print work more like it does in C (a language
> that I hate). Specifically a way to print pieces of lines while Mathemati=
ca is
> calculating rather than Print working only in whole lines.
>
> Steve Gray

1) Infinite undo might be difficult in a notebook -- too many ways to
change things, but some level of multiple undo would be handy.
(Although I find teaching students in computing labs, the single step
undo is handy -- I can show them something then delete it and they
can't recover it with undo)
Also, a convenient method of version control / back-ups would be
useful.

2) I'm not sure what you mean by this...

3) I'm happy with the cell brackets - but it could be changed.

4) I agree. Also you should be able to use the Mathematica (string)pattern
matching in their search...

5) I think that there's good reasons that Print doesn't work like in
other languages. I normally only use print when I'm debugging. But I
had a go at writing a function that does what you want: PrintOn[expr]
will append expr to the previous Print cell. It could do with some
cleaning and hardening though....

PrintOn[expr__]:=Module[{nb=EvaluationNotebook[],ebox=Riffle[ToBoxes/
@{expr},""],data,old},
NotebookFind[nb,"Print",Previous,CellStyle];
old=NotebookRead[nb];data=old[[1]]//.RowBox|BoxData|InterpretationBox-
>(#1&);
NotebookWrite[nb,ReplacePart[old,1-
>BoxData[RowBox[Flatten[{data,"",ebox}]]]]]]

Simon

From: Helen Read on 13 Aug 2010 06:57
On 8/12/2010 5:31 AM, Themis Matsoukas wrote:
>
> --A simple(r) way to change the working directory for input/output
> operations from the default (root) to the directory of the current
> notebook. The only way I know how is
>
> SetDirectory[ToFileName[
> Extract[
> "FileName" /. NotebookInformation[EvaluationNotebook[]], {1},
> FrontEnd`FileName]]]

SetDirectory[NotebookDirectory[]]

--
Helen Read
University of Vermont


From: David Park on 13 Aug 2010 06:58
I realize that many people still believe in printed documents, or are forced
to use them by requirements imposed by others. So for them pagination is
important. Nevertheless, it is an ugly artifact imposed by the physical
medium that has nothing to do with the information content.

In electronic media reasonable Sectioning, governed by the subject matter,
with moderate scrolling is a far better way to organize and present
information.

Electronic media have some of their own bad forms and one of these is the
jump-link that jumps to a far distance part of the document to obtain an
ancillary piece of information. An example might be a reference, or a
distant equation that will be used at the reading point. Rather than jumping
to another part of the document (jerking the reader around) the ancillary
information should be brought to the point the reader is at.


David Park
djmpark(a)comcast.net
http://home.comcast.net/~djmpark/


From: Vince Virgilio [mailto:blueschi(a)gmail.com]

>
> With Compartment you can now do interactive pagination.

SNIP

No need for Neologisms.

Why not simply call it a Page?

As you imply, Pagination is a crucial concept.

Vince Virgilio



First  |  Prev  |  Next  |  Last
Pages: 1 2 3 4 5
Prev: FindMinimum inside another FindMinimum
Next: Compile with Boolean Arrays