This is an important problem. From my experience working within
Carpentries Workshops and workshops in general, the single most
frustrating part of workshops can be an instructor that types fast. When
this occurs, the learner is focused on trying to keep up, trying to scan
for errors, or evaluating error messages, and is NOT focused on the
logic, significance, or elegance of a piece of code. They are focused on
getting (only) the same screen outputs.
Even if an instructor takes the time to go over the line of code,
describing what the code is doing (which is great), it is likely the
slow typist is glancing from screen to keyboard during this time, and
misses the discussion. This also means the green and red "stickies"
don't work well for slow typists because the slow typist doesn't know if
their new line of code is working until they finish it. By then the
instructor may have moved on.
The problem is exacerbated if the instructor uses his own command
history to overwrite or edit new commands for example going back in an
iPython notebook and editing a previous cell. Then the original line of
code is effectively lost.
If the instructor cannot go slower (usually bad, as our lessons are
timed), then we provide a copy-paste alternative at Oklahoma State using
a "live" etherpad and a talented volunteer/helper. This lets the learner
"get caught up" with the instructor, and they can go back later to
figure out what was causing the problem.
-peter
On 07/16/2018 11:55 AM, C. Titus Brown wrote:
Great discussion!
A few comments from our experience doing copy-paste during our two week
sequence analysis workshop for the last 9 years. (Materials here:
angus.readthedocs.io/en/2018)
* retention of commands after copy-paste is poor, and people get into the habit
of expecting things to work (and you don’t get to see the instructor screw
stuff up, either).
* copy-paste makes the materials more useful when people revisit the web site
later.
* for true beginners, copy-paste gets them to the “ok I can see why this is
useful” stage faster and is thus more motivating (in my opinion).
So, we have chosen to move to a two-tier model, where we first introduce topics
through copy-paste. Then, after a few days, we start introducing challenges
where they are asked to e.g. execute the same pipeline on novel data, or adjust
plots, or whatever - basically, edit the commands appropriately.
Since I spend a lot of my time copy-pasting from stack overflow, Python
documentation, R docs, etc. and then editing, I feel like this last approach
approximates the approach to learning that I use on a ~daily basis (and that I
see most people in my lab using).
best,
—titus
On Jul 16, 2018, at 9:30 AM, Hao Ye <[email protected]> wrote:
And it's definitely the case that copy-and-paste can be faster -- especially if
you don't type so much that you're super fast at it.
Completely agree! Workshop attendees may be infrequently at a computer (perhaps
they are bench or field scientists), so keeping up with the typing speed of
instructors can be challenging. It's useful to keep this in mind for coding
demos (and to regularly remind folks about tab-autocompletion).
Best,
--
Hao Ye
[email protected]
On Mon, Jul 16, 2018 at 11:53 AM, Henry Neeman <[email protected]> wrote:
This is a great question!
And I've had similar experiences with typing
things in and then retaining them better.
(For me, it works with people's names, too --
if I type in your name, I'm vastly more
likely to remember it than if you say it to me.)
The thing to bear in mind is that we, being
in the business, have vastly more experience
with command line than most of the researchers
we teach and serve.
So what's obvious to us -- for example, that
many of us retain commands much better if we
type them out -- isn't obvious to them,
because they haven't had enough experience
with memorizing commands to realize that.
And it's definitely the case that
copy-and-paste can be faster -- especially if
you don't type so much that you're super fast
at it.
---
Henry Neeman ([email protected])
Assistant Vice President, Information Technology - Research Strategy Advisor
Director, OU Supercomputing Center for Education & Research (OSCER)
Associate Professor, Gallogly College of Engineering
Adjunct Associate Professor, School of Computer Science
The University of Oklahoma
3200 Marshall Avenue Suite 130, Norman OK 73019
405-325-5386 (office), 405-325-5486 (fax), 405-245-3823 (cell),
[email protected] (to e-mail me a text message)
http://www.oscer.ou.edu/
----------
On Mon, 16 Jul 2018, Purwanto, Wirawan wrote:
All,
I am new to Carpentry community so please
bear with me if this is a stupid question,
or not related to Carpentry. I have been
facilitating people on my campus to use HPC
and research computing resources in general.
One thing I see with many folks learning new
computing stuff is that they are lazy at
typing even simple commands. Instead, they
rely on cut and paste operation. I believe
there is something done on your brain if you
actually type or write things down, instead
of merely staring at words and do "passive"
copy-and-paste operation. I still
intentionally type commands (even if they
are somewhat long) just to get it written on
my brain. Anyone having similar observation?
If so, how will you encourage them to be
"active" in typing rather than just do the
most convenient thing?
Wirawan Purwanto
Computational Scientist, Research Computing Group
Information Technology Services
Old Dominion University
Norfolk, VA 23529
The Carpentries / discuss / see discussions +
participants + delivery options
Permalink
The Carpentries / discuss / see discussions + participants + delivery options
Permalink
------------------------------------------
The Carpentries: discuss
Permalink:
https://carpentries.topicbox.com/groups/discuss/Taf27bc5ac21f2829-M7cefd05986f022a94b55c2e6
Delivery options: https://carpentries.topicbox.com/groups
------------------------------------------
The Carpentries: discuss
Permalink:
https://carpentries.topicbox.com/groups/discuss/Taf27bc5ac21f2829-Ma1add1da34c5f956c1bc2412
Delivery options: https://carpentries.topicbox.com/groups
