I'm on the hunt for a good quality commercially licensed graphing /
plotting library and wondered if anyone here had any recomendations.
The work to be done is less scientific, more presentational, (I'm not
going to be dealing with heatmaps / vectors etc.., just the usual
bar / line / bubble / radar / iceberg type graphs) so a good level of
control over how the final output looks would be a key point.
I'd be grateful for any suggestions / pointers to something useful,
thanks,
Matt.
Gnuplot is ugly. I'm using it because I don't care if it's ugly but it
clearly lacks of look & feel for presentations, as requested by the OP.
You have
http://matplotlib.sourceforge.net/
which is free and looks better than gnuplot. I'm not sure it's well
suited for presentation though.
JM
Hi,
The requirement for a commercial license comes down to being
restricted to not using any open source code. If it's an open source
license it can't be used in our context.
Until now I have actually been using matplotlib, but now that has to
change.
Nothing hides that unpleasant, inconvenient data better than adding a
lot of colors, drop-shadows, and of course the unneeded "3d" look
complete with a weird perspective.
> You have
> http://matplotlib.sourceforge.net/
>
> which is free and looks better than gnuplot. I'm not sure it's well
> suited for presentation though.
Agreed. It's almost as bad at data-obfuscation as Gnuplot.
--
Grant Edwards grant.b.edwards Yow! Used staples are good
at with SOY SAUCE!
gmail.com
You may be misunderstanding this issue, I think you are equating "open
source" with "GPL", which is the open source license that requires
applications that use it to also open their source. There are many
other open source licenses, such as Berkeley, MIT, and LGPL, that are
more permissive in what they allow, up to and in some cases including
full inclusion within a closed-source commercial product. You might
also contact the supplier of the open source code you are interested,
and perhaps pay a modest fee to obtain a commercial license.
-- Paul
> --
> http://mail.python.org/mailman/listinfo/python-list
>From The Matplotlib documentation: "Matplotlib only uses BSD compatible code, and its license is based on the PSF license."
BSD and PSF both allow commercial use. There is no "copyleft" restriction.
-EdK
Ed Keith
e_...@yahoo.com
Blog: edkeith.blogspot.com
stupid spell-checker...
Seriously, most of the graphs I've seen in "presentations" would make
Ed Tufte spin in his grave.
If it's 2D data, you don't need to use a 3D graph.
--
Grant Edwards grant.b.edwards Yow! Is it 1974? What's
at for SUPPER? Can I spend
gmail.com my COLLEGE FUND in one
wild afternoon??
Just out of curiosity, where does this requirement come from?
Matplotlib (like Python itself) is offered under a license that
basically says "here's the source code. Do whatever you want with it".
Any policy that prevented you from using Matplotlib would prevent you
from using Python too.
didn't know he died.
Python itself and all its standard libraries are open source, under the
PSF license, if I recall. Doesn't look like you can use python at all.
You simply cannot write a python program without some open source code
being used.
I don't understand the propensity of companies and individuals to think
of open source code as different than proprietary. It's not at all.
All code that you use that was not written by you or owned by you has to
be used under license from the copyright holder.
> Seriously, most of the graphs I've seen in "presentations" would make
> Ed Tufte spin in his grave.
http://flowingdata.com/2010/03/20/powerpoint-and-dying-kittens/
The company I am with is unsure, and while they will pursue the
possible use of matplotlib with the PSF license, they would prefer I
continued to look into any commercially licensed libraries that could
be used.
...and as it goes, the first thing I fired back at them when I was
told 'no open source can be used' was basically 'What on earth are we
going to use to code it in...' and that the 'no open source'
restriciton as given was overly broad when you think about it.
He hasn't.
--
Robert Kern
"I have come to believe that the whole world is an enigma, a harmless enigma
that is made terrible by our own mad attempt to interpret it as though it had
an underlying truth."
-- Umberto Eco
If'd seen some of those graphs he would have.
--
Grant Edwards grant.b.edwards Yow! I'm having an
at emotional outburst!!
gmail.com
:)
Years ago I was walking past a marketing guy's cube, and on his cube
wall he had Minard's famous "Napolean's March" graph that Tufte so
admired. I asked him if he had read Tufte's book "The Visual Display
of Quantitative Information" (in which the graph appears). He replied
that no he hadn't read Tufte's book -- he'd gotten that graph from a
power-point-driven lecture in some MBA class or other.
--
Grant Edwards grant.b.edwards Yow! I'm rated PG-34!!
at
gmail.com
is unobsuscated an obsuscated version of unobfuscated?
defniately.
--
Grant
> I'd be grateful for any suggestions / pointers to something useful,
Ignoring the commercial vs. open source discussion, although it was a
few years ago, I found Chart Director (http://www.advsofteng.com/) to
work very well, with plenty of platform and language support,
including Python.
-- David
> If it's 2D data, you don't need to use a 3D graph.
if it's tabular data, you don't need an uber-histogram
--
giampippetto, coso, come si chiama? ah si` "MMAX" ha scritto:
> Tra il trascendente e l'interpretazione prevalente del dato come
> assioma ne passa...