Thanks, Brian. Lots of stuff to act on, here. The only thing I'd debate with you is: "lot of detail early on". I'll have to get comments from APLers, but I don't think they'll gag at seeing a lot of code right off. What *is* off-putting is to see a longish piece of code like the CALC app exposed in the article. Why not simply give instructions to download it, and then refer to it as read?--I ask myself. Of course, people like to read Vector without a computer to-hand. But having listed the Form Editor's stubs, moderate imagination will help the reader to visualise the code of calc.ijs.
The CALC app itself needs tarting up. An Equals button would do no harm. I was going to set the reader an exercise to code the remaining buttons: - , / , *. I haven't noticed focus problems on my Mac, but now you've mentioned it, I'll try and force the bug -- and test it in Windows too. On Mon, Oct 25, 2010 at 5:44 PM, Brian Schott <[email protected]> wrote: > Ian, > > I have not completed reading JinaDay, but have a few observations. > > You throw a lot of detail at the reader early on especially in the > code snippets without explaining much of that detail. That seems to me > to be off putting, but one way to reduce that would be to group that > material together under a heading or an explanation/warning that the > main early goal is to orient the reader to the user interface, not to > learning itself. > > You discuss "locale" once earlier than you give a link for locale. > > A good link to add (unless you already have and I missed it) early on > would be the following. > > http://www.jsoftware.com/jwiki/System/Installation > > I had trouble orienting myself to the calc.ijs calculator because it > has no enter or ok button, only the "+" button. But it is a good > simple app. I also was a little disoriented when I pressed numeric > keys which sometime produced digits in the number field and sometimes > not, as I recall, depending on where the focus of the cursor lies. > Some prose describing the app's behavior might be added. > > When first I was reading the draft I had my browser zoomed in and the > long code lines required 2 lines making the postfixed NB. statements > land on a second line. This will not be a problem in Vector print or > in pdf, but it was disorienting. > > I think this is a valuable work and you should get good feedback from > this group. I am especially impressed with your knowledge of what is > different in J and Apl. It might be good to corral all those > differences into an appendix. If there is a standard ubiquitous Apl > "recipe book of sample input statements > " or app, another approach would be to state that early on and then > show how it is accomplished in J. But that does not seem to be your > direction, so feel free to ignore this. > > On Sun, Oct 24, 2010 at 3:19 PM, Ian Clark <[email protected]> wrote: >> First draft of an article for eventual submission to VECTOR entitled >> "J In A Day". >> >> http://www.jsoftware.com/jwiki/JinaDay > ---------------------------------------------------------------------- > For information about J forums see http://www.jsoftware.com/forums.htm > ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
