What I am thinking we want is 5 minutes that leaves the user thinking,
"Wow. That was cool. They did all that and it was so fast and short.
I want to learn how they did that." The demo should make extensive use
of libraries and packages to get a job done. It needs to be a job that
takes some coding - something that would take pages of C - but one where
the J program is short. You don't have to write the J program in 5
minutes! We can have the program already written, but you should be
able to type it, run it, and see flashy results in that time.
Henry Rich
On 2/15/2014 11:44 AM, Murray Eisenberg wrote:
The issue is what, exactly, one wishes to convey by a short, introductory, video.
One possibility is an advertising blurb that tells who uses the language and for
what wonderful purposes. Another starts out with some fundamental notions of the
language ("you form an array like this, assign it to a name like this,
transpose it like that using a one-argument function, etc. And yet another -- the
kind in the videos I cited, just gives a feel for what using the language is like by
actually coding some simple yet moderately useful example.
Yes, a demo like Cliff Hastings makes things seem deceptively easy. But: (1)
the help system for this particular programming system is so incredibly rich
that one could discover how to do all this oneself; and (2) this demo is
actually part 8 of an 8-part series of videos. (I just couldn't lay my hands on
the free-standing, self-contained, similar demo that Hastings had posted
elsewhere.)
[Note. There is NO error in the first plot of the fitted line: the origin is not
at (0,0) or (1,0) but at (5,0) -- because that's what the system selected as best
showing the line's plot. In the second plot, of the original, quadratic, data
together with the linear fit, the system again selects a suitable origin, in this
case (0,0). And if you don't want to let this sophisticated system make such
choices for you, you can always override its choices, in this instance by
specifying an option AxesOrigin -> {0, 0} .]
On 14 Feb 2014 20:02:42 -0700, Don Guinn <dongu...@gmail.com> wrote:
Just viewed the videos by Cliff Hastings for Wolfram. Surprised to see that
there looked like an error in the second video on making a first order fit
showing the line going above the origin when x=0. Later it showed it
correctly. Sent him a note about that.
But what really bothers me about demos like this is that they look so easy
when they do it, but if I were to try to do it I wouldn't know where to
start. He implied that one could do it without knowing much of anything of
their system. I really get tired of videos like this where they type really
fast and it looks so easy if one just knew their system well, but I usually
don't. If I was presented that screen and wanted to do what he did I
wouldn't have a clue what to do.
We need to present similar videos on J, but somehow we need to make it
obvious and logical as to what to do. His video was neat, but could I do it
as quickly and easily as he did it without putting in hours, possibly days
learning their system? I doubt it.
On Fri, Feb 14, 2014 at 7:17 PM, Murray Eisenberg <mur...@math.umass.edu>wrote:
If you'd like to see what a good quick demo looks like, done by one guy
with no fancy production values -- and of a language/system having a
state-of-the-art user interface, take a look at either of the following:
http://www.wolfram.com/broadcast/search.php?Search=app%20minute&x=-879&y=-139&video=728
http://www.wolfram.com/broadcast/video.php?channel=86&video=869
On 14 Feb 2014 19:00:45 -0500, Henry Rich <henryhr...@nc.rr.com> wrote:
As Ian [Clark] observed, a newcomer's first 5 minutes with J will be
decisive in
establishing their attitude towards the language. As things stand, it
takes a serious geek to take a shine to J in 5 minutes. Just between us
geeks, I wish there were more of us, but that's not the way to bet.
No, we need a snappy demo: an application that everyone can relate to,
showing how we can code something meaningful and get a pretty display in
under 5 minutes. Ideally it should be a YouTube video, with an
accompanying Lab so the interested user can reproduce the results
——
Murray Eisenberg mur...@math.umass.edu
Mathematics & Statistics Dept.
Lederle Graduate Research Tower phone 240 246-7240 (H)
University of Massachusetts
710 North Pleasant Street
Amherst, MA 01003-9305
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
