One idea I've been using is to show explicitly what's going on in the background when you're using defaults by instantiating all the default settings:
http://www.loria.fr/~rougier/teaching/matplotlib/#using-defaults versus http://www.loria.fr/~rougier/teaching/matplotlib/#instantiating-defaults Nicolas On Mar 25, 2013, at 18:43 , Damon McDougall wrote: > On Mon, Mar 25, 2013 at 12:17 PM, Thomas A Caswell > <tcasw...@uchicago.edu> wrote: >> I think there is something to be said for not starting from pylab. >> Answering questions on SO, a good chunk of them (by volume) can be >> traced back to not understanding the magic that pylab is doing for you >> in the background or not even knowing magic is being done for you. >> Starting from pylab makes easy stuff trivial, but slightly more >> complicated things a much bigger lift to figure out how to do (as >> compared to the conceptual difference in how hard they are). >> >> A tutorial that starts from the POV of building the figure out of >> parts sounds like a good idea to me. At a minimum, a key with the >> different parts of the figure labeled with what family of classes >> control them would be great (or if something like that already exists >> make it easier to find;)) >> >> Tom >> >> On Mon, Mar 25, 2013 at 12:03 PM, Benjamin Root <ben.r...@ou.edu> wrote: >>> >>> On Mon, Mar 25, 2013 at 12:46 PM, Phil Elson <pelson....@gmail.com> wrote: >>>> >>>>> I am putting together a beginners tutorial proposal that I will submit >>>>> soon >>>> >>>> That's great to hear! Are you planning on making the tutorial material >>>> part of mpl's docs or using the content that is already out there? >>>> >>>> >>> >>> It is all new stuff, but I have been taking inspirations from other >>> tutorials I have seen and said to myself "You are all teaching it wrong!" >>> :-P >>> >>> I am ignoring pylab (risky, I know), starting with a *very* basic NumPy >>> primer, and then moving on to teach matplotlib from the perspective of "here >>> are what the parts of a plot are called and what they are for, and see what >>> happens when we put those parts together". It is an ingredients approach, >>> essentially. >>> >>> Hopefully, aspects of it will be useful for the docs when it is finished. I >>> am also hoping that having a ipython notebook version of it will help others >>> to improve it for future conferences (there should always be an intro to >>> matplotlib tutorial at SciPy). >>> >>> Ben Root > > That seems like a good approach to me. Thanks for doing this. I just > submitted a tutorial, but it assumes people know how to make a line > plot already. Perhaps I should learn from this assumption and > communicate better on this list and garner interest about what people > would like to see a priori. > > Thanks for putting this together, Ben. Out of interest, are you > diving straight into the pyplot state machine, or are you taking the > more object oriented approach of setting up the canvas and figure > object explicitly? I use the OO approach all the time, but I only use > the non-interactive backends like Agg and PDF. > > On a not-too-orthogonal note, I'd personally like to see a tutorial on > hooking in mpl into other GUI-like applications. Paraview seems to do > this a little, but I'd like to see someone do a soup-to-nuts > walkthrough for it, just because I have no experience doing this; I'm > a terminal hermit. > > Best wishes, > Damon > > -- > Damon McDougall > http://www.damon-is-a-geek.com > Institute for Computational Engineering Sciences > 201 E. 24th St. > Stop C0200 > The University of Texas at Austin > Austin, TX 78712-1229 > > ------------------------------------------------------------------------------ > Everyone hates slow websites. So do we. > Make your web apps faster with AppDynamics > Download AppDynamics Lite for free today: > http://p.sf.net/sfu/appdyn_d2d_mar > _______________________________________________ > Matplotlib-devel mailing list > Matplotlib-devel@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/matplotlib-devel ------------------------------------------------------------------------------ Everyone hates slow websites. So do we. Make your web apps faster with AppDynamics Download AppDynamics Lite for free today: http://p.sf.net/sfu/appdyn_d2d_mar _______________________________________________ Matplotlib-devel mailing list Matplotlib-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/matplotlib-devel
