[omaha] Python Workshops @ DoSpace Group

[Jeff Hinrichs - DM&T]

Burch,
I think you are hitting the nail on the head. And it is good that you are
pointing out that the goal might appear to be too geared to a programming
as a profession.  I am glad that you said something, because I am looking
to approach this the other way round.  That is to inspire programming in
other professions.   With the current push on in the sciences for
repeatable results in data analysis, I would think that within the next
decade that anyone who works with data for a profession (not programmer's
but analysts, scientists, etc.) would be ill equipped if they can't wrangle
data.

To that end, many just like you and myself (when I attended college, you
only received 1 credit hour for a language class.  They were graduating
Computer Scientists not programmers.) pick up a language or other tool,
based on our ability and the reference material that we have at hand (or
online).  And that first exposure sits like concrete, as you stated. The
example you sited about file handling is very insightful and I think lends
credence to my goal of making the first contact for those new to
programming as positive as we can.

My goal is to have the examples that we show to the class be as good as we
can make them.  Be written in a way that lends itself to good convention
and best current practices.  So that when we show students example code --
it is of good quality, not a questionable example that was thrown together
to demonstrate one thing while completely ignoring all other good
practices.  I believe the examples we pick for use in these still fictional
classes should show not only what we want to demonstrate on the micro level
but also be a good example on the macro level.  So that if a
student/attendee used one of our examples as a starting point for something
bigger, they would be on good footing.   When people first start
programming we don't write code as much as borrow it from other places.
*wink*   My goal is to create the best quality code examples, that they can
borrow for other things.

Also, for beginners a blank file is too much.  The examples I've come up
with so far require that they fill in the missing pieces, concatenate some
strings here, finish out the function body there, but the rest of the code
should be stellar because we wrote it for them.  Doc strings where they
belong, conventions followed for naming variables (no more i, j, k which
dominate in Fortran or x and y like it is a math problem)with descriptive
names.  So that when the time comes for a student to create a variable out
of thin air they will be more likely to be descriptive too.  But even if
they don't use descriptive names students won't be berated, what matters is
that code does what is intended.

As for testing, I am not going to try and sell you on it.  (Like backup and
restore, most are not concerned until the ugly happens.) I am not even
planning on talking about it in these opening workshops.  However, that
doesn't mean I am going to favor code examples that are not testable.  Far
from it.  Everyone of my current example exercises has associated tests.
The students won't run the tests but my agent, sam will.  When they finish
an exercise I'll know, as sam will let me know their code is running and
passing the tests.  That also requires that the creator of the specific lab
write in a way that is testable.  However, none of this is a burden to the
student as they are not going to be running the tests or even thinking
about creating testable code but they'll be learning it subliminally.

And now that I've used the word, subliminal -- I just realized that was my
goal all the time.  In each example, given on a slide or as an exercise
there is the part that you are talking about discretely.  Focusing on that
topic.  However, the surrounding and supporting code is also teaching and
informing the student.  That is the part that needs more love in just about
every example I've seen on the internet.

If none of this made sense or you still don't agree (However, I think that
I'm tripping over myself agreeing with you.) please say so.  The point of
this thread is to have a conversation where we can flesh out some
reasonable goals and then see if we can meet them.  So if you are reading
this and have an opinion, a point, a question then please, please chime in.




On Sun, Jan 3, 2016 at 8:36 PM, Burch Kealey via Omaha <omaha at python.org>
wrote:

>
> I will make an observation - this is not meant to be critical but I want
> you (collectively) to chew on it.  If I want to learn to be a programmer I
> would look for a well structured learning path.  If I want to learn to do
> things to make my life easier I am not sure I want to take the same path as
> if I wanted to learn to be a programmer.
>
> This note is sent in response to both Jeff's and Wes's posting.  Please I
> am not trying to be critical but you guys are programmers by training and
> vocation I am not.  I don't have the time to learn to be a programmer.
> Unit Tests, what the heck are those?  Why do I care?
>
> I ask a question on SO and a bunch of idiots jump up and are critical for
> the fact that my code is not PEP 8 compliant - okay it makes them feel good
> but it does not help me.  For example, for the longest time ( say 5 years)
> when I opened and read a file I would do this
>
> mydata = open('some file path','r').readlines()
>
> That is not PEP 8 compliant (I don't think), but when I first Googled to
> learn how to open a file and read the contents into a list - I think this
> is the webpage I found
>
> http://openbookproject.net/pybiblio/tips/wilson/loadingfile.php
>
> It was only this year that I took the time to learn the right way
>
> with open('some file path','r') as f_handle:
>     data = f_handle.readlines()
>
> Why did I not bother learning the right way before this year, I was too
> busy learning how to do other things and it was working for me.  Why did I
> even bother learning the right way, because I got sick of the way the
> conversation would get channeled if I had a question or comment about
> something and the pros would say well you are not opening the file right.
> My response is who gives a drat, that is not the problem, but they are
> stuck on that issue.
>
> I am probably not explaining this well - certainly you want to teach
> people the right way but I would be careful about how far you take it.  Our
> software is used by over 5,000 people and I have never written a unit test
> (have no clue).  I know that is not a big number but it is the amazing
> thing about Python.
>
> Don't cringe, drink some OJ!
>
> Just an observation.
>
> Happy New Year
>
> Burch
> ________________________________________
> From: Omaha <omaha-bounces+bkealey=mail.unomaha.edu at python.org> on behalf
> of Wes Turner via Omaha <omaha at python.org>
> Sent: Sunday, January 3, 2016 8:12 PM
> To: Omaha Python Users Group
> Cc: Wes Turner
> Subject: Re: [omaha] Python Workshops @ DoSpace Group
>
> https://wrdrd.com/docs/tools/#python
> * http://learnxinyminutes.com/docs/python/ # ./git
>
> "[omaha] For Python beginners"
> https://mail.python.org/pipermail//omaha/2015-May/001816.html
>
> * http://docs.python.org/tutorial
> * https://scipy-lectures.github.io/intro/language/python_language.html
> * https://github.com/audreyr/cookiecutter-
> <https://github.com/audreyr/cookiecutter-pypackage>pypackage
> <https://github.com/audreyr/cookiecutter-pypackage>
>
>   *Create and test a package w/ git[hub] and CI"
>
> * Rosalind [Bioinformatics] exercises are outstanding
>   1. they're already set up for online grading/points
>   2. you get to learn about the domain
>   * | Homepage: http://rosalind.info/
>   * https://wrdrd.com/docs/consulting/data-science#rosalind
>
> * "How to create and *test* a package for your Rosalind solutions"
>
>    * "Testing as a core scientific process #126"
> https://github.com/scipy-lectures/scipy-lecture-notes/issues/126
>
>    * TDD first!
>       [null] hypothesis, code, test!, [commit]
>
>       *
> https://westurner.org/wiki/awesome-python-testing.html#workflow-evolution
>
>          *
> https://wrdrd.com/docs/consulting/education-technology#jupyter-and-tdd
>           *
> https://wrdrd.com/docs/consulting/education-technology#nbgrader
>
> [omaha] November 18 Meeting - Celebrate and Plan!
> https://mail.python.org/pipermail/omaha/2015-November/001943.html
>
> > The software carpentry lessons are great (and designed for group
> instruction IIUC):
> > * https://software-carpentry.org/lessons.html
>
> ... https://westurner.org/opengov/us/ne/#cs-skills-for-stem-fields :
>
> * https://wrdrd.com/docs/consulting/software-development#computer-science
> To all of who offered to help in any form with giving Python workshops @
> DoSpace,
>
> I've been hammering out some ideas for a class targeted at those new to
> programming in particular.
>
> We should probably start a conversation about it and see what we can put
> together.
>
> Here are some goals, because we need to have something to measure against:
>
> * should allow for students to do something with Python as soon as possible
> - don't want them bored and inattentive.
>
> * The See/Do affect of examples should have the following qualities:
>
>   * be pep8 compliant (only deviating when we are teaching a specific
> topic.)
>
>   * be easy to test.  No example has print or input in the middle of a
> function other than main or one designed to deal with user interaction.
> Too many examples mix I/O with processing leaving students with a knack for
> writing hard to test code.  (All example code and problems should have an
> associated test suite [py.test] to accompany them.)
>
>   * Should emphasize DRY and only deviate when showing them undry
> situations and how to correct.
>
>   * Early coding problems are often quite boring, so there should be some
> sort of setup for them, a backstory, to get the students invested in their
> solution.  For Example - learning comparison operators -- "Sam is a young
> AI with untrained neural nets and we need to write a module to help train
> him had to organize two things so he can put them on a shelf in order."
>  Sounds a lot more interesting than write a function that compares 2 things
> and output them in order from smallest to largest. *yawn*
>
>
> So what are your ideas?  What are you willing to help out with?  We'll need
> material reviewers, testers and trainers and TAs for actual classes.
>
>
> I believe someone has offered to be our DoSpace ambassador. Sorry, but I
> didn't write down your name at the meeting.  Would you speak up once again.
>
>
> --
> Best,
>
> Jeff Hinrichs
> 402.218.1473
> _______________________________________________
> Omaha Python Users Group mailing list
> Omaha at python.org
> https://mail.python.org/mailman/listinfo/omaha
> http://www.OmahaPython.org
> _______________________________________________
> Omaha Python Users Group mailing list
> Omaha at python.org
> https://mail.python.org/mailman/listinfo/omaha
> http://www.OmahaPython.org
> _______________________________________________
> Omaha Python Users Group mailing list
> Omaha at python.org
> https://mail.python.org/mailman/listinfo/omaha
> http://www.OmahaPython.org
>



-- 
Best,

Jeff Hinrichs
402.218.1473