| Re: iTools/interactive Analysis > [message #48219] |
Thu, 06 April 2006 06:55 |
Phony Account
Messages: 14
Registered: May 2005
|
Junior Member |
|
|
I love iTools.
I use them to capture post-processed data from simulations and generate
presentation quality plots. I like the fact that I can open up the
saved file, and all the data is there, and I can tweak the look of the
plot.
What is missing are two things:
Capability to add notes to an iTool. I don't mean text that shows on
the screen, but internal notes. I would use that facility to remind
myself how I got the data, and stuff like that.
The second thing that is missing is actually a tall task: ideally, an
iTool would capture the whole data analysis session from start to
finish. Let me explain:
I work in a corporate environment, and everybody around me uses Excel
for data analyisis. Now, spreadsheets may not match to such highly
performing packages such as IDL and Matlab, _but_ in an excel
spreadsheet file, you have the data, the operations on the data, and
the plots. (btw, the plots are a pain to work with and look awful, at
least as compared to iTools). So, once you save your spreadsheet, you
can open it months later, and everything will be there.
There is plenty bad with data analysis in spreadsheets:
- It is hard to see the flow of the data analysis - at a glance it is
hard to see what depends on what.
- You see all the data whether you need to or not - I see spreadsheets
with columns and columns of data.
One can work around those problems, but the format of the spreadsheet
interface drives you to poor "aesthetic" practice.
Several times in the past I have considered writing an spread-sheet
type data-analyzer in IDL. All the data-processing facilities would
have to be encapsulated in objects with a uniform interface. But
unfortunately, I just never get fired or laid of, so I never have the
time to pursue that project.
On the other hand, if you thought iTools were slow, this one would be
much, much slower :-)
On the other other hand, at the company where I work (Tokyo Electron
Limited) we are striving to provide semiconductor manufacturers with
tools to make THz chips.
Mirko
|
|
|
|
| Re: iTools/interactive Analysis > [message #48244 is a reply to message #48219] |
Tue, 04 April 2006 15:35  |
David Fanning
Messages: 11724
Registered: August 2001
|
Senior Member |
|
|
catwithhat writes:
> When I first heard about iTools I thought it was going to be a great
> addition to IDL. A way to combine command-line-based analysis with,
> click and point modification of graphics (IDL's answer to Matlab).
> After how many years now (?) iTools still seems like a slow, difficult
> to use and ugly! behemoth that seems too unwieldy for interactive
> analysis. Am I the only who feels this way? Are we too small of a
> market? Is RSI just not hearing from us? Maybe I am alone?
Welcome to the gulag. You are going to like it here. :-)
Cheers,
David
--
David Fanning, Ph.D.
Fanning Software Consulting, Inc.
Coyote's Guide to IDL Programming: http://www.dfanning.com/
|
|
|
|
| Re: iTools/interactive Analysis > [message #48248 is a reply to message #48244] |
Tue, 04 April 2006 09:47 |
Antonio Santiago
Messages: 201
Registered: February 2004
|
Senior Member |
|
|
> After how many years now (?) iTools still seems like a slow, difficult
> to use and ugly! behemoth that seems too unwieldy for interactive
> analysis. Am I the only who feels this way? Are we too small of a
> market? Is RSI just not hearing from us? Maybe I am alone?
>
> Thanks
> Axel
>
In my case, after a couple of years without paying atention to iTools
and developed some object oriented IDL application, I was try to create
my own development framework: clases to send messages among objects,
wrap any IDL data into an object, ... and grouping this classes in sets
or packages (something similar to java) depending on its function:
classes to visualize data, to manipulate data, ...
But, oh, that's the same thing iTools do. I know they aren't the fast
way to analize any type of data, but is an existent framework (the
opposite of my inexistent previous described framework) with a lot of
good ideas (perhaps with some possible bugs, and perhaps not satisfied
all people).
Depending on the application you need to develop it can be a good tool
on your hands. Next are a set of good and bad situation, i have thinked,
to use iTools:
* Create an program to read a file and write some things into another:
BAD idea use iTools.
* Read a file and visualize its data as an image, plot, contour:
possibly, BAD idea.
* Read a file, visualiza data and interact with it calculating and
showing some results: mmm... possibly BAD idea but mmm...
* Read some files, represent data overlapping images, plots , points,
polygons, ... and interact with it as they are independent objects and
need to change its properties (color, thick, palette, ...): I think this
is a possible GOOD idea to use iTools.
* You have a belly pain (i don't know if this expression is correct) and
for some days you haven't been going to bath, and you need to modify the
source code of iTools to do something different: YES it is a very GOOD
idea. In less that one minute you are on the bath :) Sorry this is a
joke based on my own expreience. But even that bad experience I continue
thinking iTools is a great set of classes to made some types of
applications.
PS: It can be useful write a list (like above) of GOOD/BAD situationsto
use iTools.
--
-----------------------------------------------------
Antonio Santiago P�rez
( email: santiago<<at>>grahi.upc.edu )
( www: http://www.grahi.upc.edu/santiago )
( www: http://asantiago.blogsite.org )
-----------------------------------------------------
GRAHI - Grup de Recerca Aplicada en Hidrometeorologia
Universitat Polit�cnica de Catalunya
-----------------------------------------------------
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
