Spot-on, Bob. Oh, yes. We could have done so so much more (and so so differently) with a running J process integrated with jwiki -- and not only to offer the user a working J to try things out. For example (just one of many possible opportunities) it could have handled most of the heavy work of hyperlinking -- which has been a labor-intensive and error-prone process. Another example: we could have ditched the collection of Ancillary Pages in favor of a new crop of interactive tutorials -- a 21-century follow-on of the highly successful j602 "Labs". And all those little embedded code samples would have taken on a distinctly different form.
And -- yes -- I have J verbs to launch the NuVoc page for a given primitive. Though not to recognise its valence (yet). Easy to write as a patch for locale 'jijs' in j602. And no doubt for JHS and JQT too, but I haven't got into the guts of those yet. In this day-and-age, there's no enduring technical reason for the UI of a product and the UI of its Help system to be deployed as two distinct interfaces. Only (a) tradition, and (b) the fact they're designed and implemented by two distinct bunches of people. Study Ableton Live ( www.ableton.com) for a product with a rich functionality that's seamlessly integrated with its on-line help. Or if you want to compare it with another programmer IDE (after all, writing J verbs is not really like writing music), how about Apple's Xcode? Recall too that even in the 1990s Mathematica was delivering ALL of its on-line help in the form of "notebooks" (*.NB) -- the prime medium of user output with that product. The moral equivalent in J would be to deliver the entire content of jwiki, including NuVoc, in the form of IJS scripts (...or IJT's??*!) with no diminution of users' experience. Something along these lines is technically achievable with JHS -- but Javascript isn't a serious arrow in my quiver (and I don't see why it should have to be). Frankly I'd like to see J itself as an alternative scripting language in place of Javascript (it needs its own J-like DOM interface -- as a new crop of Foreigns maybe?) -- though the security implications of wide deployment of J in that role are worrying, when you think of how Java and Flash are having to patch a never-ending stream of security exposures. But at least NuVoc as it stands can be used in conjunction with a 40 year old design of UI, like jconsole, which is still popular among important classes of J users (i.e. those of us over 70). There's always this need for bringing up the trailing edge of technological innovation. Which is what I see NuVoc as doing, if you want to be brutally clear-eyed about it. On Fri, Jun 27, 2014 at 4:26 AM, robert therriault <[email protected]> wrote: > Congratulations to both Ian and Henry for great work on Nuvoc. > > One of the things that I found particularly illuminating was the J test > for a number of reasons: > > 0) Not a surprise, but under the rules of not being able to use a running > session I did spectacularly badly on the test. I won't give my score but it > probably rhymes with hero. > 1) This made me think about how I investigate J and the fact that the > interactive session is key to my experience. This is true of labs, trying > out phrases or solving problems. > > This made me think of a few extra possibilities: > > 0) How powerful would Nuvoc become if it were provided in a window next to > a running J session? I know this can be done in JHS. > 1) Further, if a double click on a primitive in the interactive session > took you to the corresponding Nuvoc page in the second window? Probably > some javascript dazzle to do that. > > Nice work both of you. > > Cheers, bob > > On Jun 26, 2014, at 7:14 AM, Henry Rich <[email protected]> wrote: > > > The new NuVoc is twice revised from the previous version: first I added > technical details, then Ian rewrote it all to achieve that 'straightforward > style', aimed at the programmer who is trying to get a job done. It's easy > reading, achieved by careful writing: simple presentation of concepts with > examples, using typography as a framework for the flow of ideas. You also > need formatting conventions that don't perplex a novice, a page design that > both lets the beginner understand a primitive at a glance and gives the > expert full detail, and enough hyperlinks and supporting material to bring > a newcomer along. > > > > It was an eye-opening apprenticeship for me to see him do it. Writing > this kind of material is very different from writing a book. > > > > Henry Rich > > > > On 6/26/2014 8:41 AM, Simon Barker wrote: > >> I also would like to say a big thank you for what is turning out to be a > >> work of great depth and richness. Really enjoying reading it, the > >> presentation and straightforward style are just encouraging me to keep > >> reading. A true page turner! > >> > >> Kind regards, > >> > >> Simon > >> . > >> > >> -----Original Message----- > >> From: [email protected] > >> [mailto:[email protected]] On Behalf Of Ian Clark > >> Sent: 26 June 2014 12:11 > >> To: General forum > >> Subject: Re: [Jgeneral] Announcing NuVoc, Version Two > >> > >> Thanks, John. > >> > >> > >> On Wed, Jun 25, 2014 at 4:02 PM, John Baker <[email protected]> > wrote: > >> > >>> Ian, > >>> > >>> I have been following NuVoc revisions for sometime. Let me be the > >>> first to express my thanks for everyone that has worked on NuVoc. This > >>> is is an excellent explication and recasting of the J dictionary. > >>> Eveyone at all levels of J expertise will find something of value in > >>> NuVoc. Just this morning it reminded me of _: and p.. facilities I > >>> sort off knew were there J but I have not used lately. > >>> > >>> Well done everyone! > >>> > >>> > >>> On Wed, Jun 25, 2014 at 8:24 AM, Ian Clark <[email protected]> > wrote: > >>> > >>>> This post is a formal announcement of NuVoc, Version 2. > >>>> > >>>> For the details, go to > >>>> http://www.jsoftware.com/jwiki/Vocabulary/AboutNuVoc > >>>> > >>>> This saves rambling on about what is quite adequately covered there. > >>>> But one thing deserves to be repeated... > >>>> > >>>> Thanks are due to Ric Sherlock, who designed the NuVoc portal and > >>>> drafted trial layouts for the primitives-pages, to others who > >>>> contributed pages > >>> and > >>>> pioneered presentational ideas, notably Devon McCormick, David > >>>> Lambert, > >>> Bob > >>>> Therriault and Dan Bron, plus several others who commented on > >>>> material > >>> and > >>>> offered encouragement. (Apologies to any significant contributors > >>>> whose names have been omitted.) > >>>> > >>>> But most of all, my heartfelt thanks to Henry Rich, at whose > >>>> instigation this major revision was undertaken, and whose monumental > >>>> and meticulous labors over the past half-year to make it happen have > >>>> been truly > >>> Herculean! > >>>> But, as I'm sure you'll agree, by no means Sisyphean :-) > >>>> > >>>> Ian Clark > >>>> -------------------------------------------------------------------- > >>>> -- For information about J forums see > >>>> http://www.jsoftware.com/forums.htm > >>>> > >>> > >>> > >>> > >>> -- > >>> John D. Baker > >>> [email protected] > >>> ---------------------------------------------------------------------- > >>> For information about J forums see http://www.jsoftware.com/forums.htm > >>> > >> ---------------------------------------------------------------------- > >> For information about J forums see http://www.jsoftware.com/forums.htm > >> > >> > >> ---------------------------------------------------------------------- > >> For information about J forums see http://www.jsoftware.com/forums.htm > >> > > ---------------------------------------------------------------------- > > For information about J forums see http://www.jsoftware.com/forums.htm > > ---------------------------------------------------------------------- > For information about J forums see http://www.jsoftware.com/forums.htm > ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
