I initially wanted to do a plaintext version with ANSI C escape codes; and HTML one; and the TeX one. The first proved very annoying after two tables; the second very annoying given special/reserved HTML characters that should be escaped, and the third almost likewise. An internal and official one would be neat as well. GNU Octave/MATLAB is a good example of REPL with help pages. Anyway..
As Henry mentioned, I'll be linking to primitives' NuVoc page, or other resources. Ian Clark may have ideas beyond the obvious above one. Visible only to pdf viewers, links will be framed by some color. Were this HTML, you could fit an encyclopedia-worth of info in title attributes, but editting and maintaining would put me off. Having a jump-off point would be useful. Regarding color, there are four colors used for the four parts of sentence, plus a gray. Those who cannot perceive yellow, will adjust accordingly, there are 3 sets of 4 highly constasting to a white background colors, to choose from if one is discontent. It bears repeating, opinions on anything, be it small or major, especially if negative, will be helpful in designing the base document. Hopefully, new tables, blocks, examples which would allow creation of new/better/other refsheets. A first draft will be coming next week probably. Aug 6, 2022, 20:50 by rauldmil...@gmail.com: > A (probably future) purely online presentation would likely benefit > from "tooltips" (or: hints which are shown on hover).. > > -- > Raul > > On Sat, Aug 6, 2022 at 10:47 AM Ian Clark <earthspo...@gmail.com> wrote: > >> >> I've been misapprehended. Maybe because I traded precision for terseness. I >> used "footnote" when I meant narrative [text] in general. As opposed to >> symbolic [text], which is the real meat of a reference card. >> >> The old 6.02 refcard has very few *footnotes* as such (i.e. narrative >> linked to its target with a superscript numeral), but it exhibits the same >> defect, with too much in-line narrative. >> >> It's a compromise: more means less, as is the case with color. Too much >> narrative and it begins to communicate noise, not information. Bundling-up >> the in-line narrative of the 6.02 refcard into actual footnotes showed this >> clearly for me. If you had to ask where to put them on the page, my answer >> had to be: in another place, i.e. on their own webpage. >> >> > how would the programmer know what's behind the link, and whether to click >> it? >> >> If you simply replaced every superscript numeral with an asterisk, that >> would alert the reader to the link's relevance. I was thinking of only one >> weblink printed on the refcard, for the benefit of the handful of >> individuals still using (laminated) paper sheets. >> >> Footnotes (with superscript numerals) are a visually poor way of linking >> narrative to its target anyway. If the refcard is going to be viewed >> primarily on-screen, clickable, then footnotes are contraindicated. >> Spider-lines are better; hypertext is better still. >> >> So what would I like to see. If the refcard is mainly to be viewed >> on-screen, then it no longer needs to bow to the conventions of printed >> paper (except for nostalgia). It's just a schematic diagram of symbols, >> even if those symbols are textual. A good design would have no narrative, >> just clickable symbols. >> >> But isn't that the heart of J? >> >> On Sat, 6 Aug 2022 at 14:59, Henry Rich <henryhr...@gmail.com> wrote: >> >> > The footnotes serve to reduce the amount of text in the table. If you >> > put all that information into the table, the table would become much >> > bigger and would have large areas of empty space. Space is going for >> > $1000/square inch. >> > >> > The footnotes are not for completeness. They give only information that >> > we think a programmer might need to have at hand to solve a problem. If >> > we have that right, putting the info into a link doesn't really serve: >> > how would the programmer know what's behind the link, and whether to >> > click it? >> > >> > BUT: it has become clear that my model for use of the refcard was >> > wrong. I thought users would do what I do: print it & laminate it. It >> > has become clear that they use it from a window on their monitor, and >> > that links to NuVoc from the refcard are important. >> > >> > Fortunately, Viktor is working on adding the links. >> > >> > Henry Rich >> > >> > >> > On 8/6/2022 9:11 AM, Ian Clark wrote: >> > > Once you start letting-in footnotes (…refugees from the Land of >> > Textbooks?) >> > > where do you stop? >> > > >> > > Isn't the proper place for all such supporting information at the far end >> > > of a weblink, which is all you need to display on the card? >> > > >> > > There you can show the J Reference Card with spider-lines to the >> > > "footnotes", in the reader's own language, if you will. The webpage can >> > be >> > > adjusted and expanded in the light of hard evidence. Maybe that's the >> > place >> > > to document language-changes since 6.02. >> > > >> > > As for multiple coloration, so pretty, and I love it. If color is >> > > available, why not flaunt it? Who cares if it's overused nowadays? >> > > >> > > But as many as 7% of the population have color-defective vision ( >> > > https://en.wikipedia.org/wiki/Color_blindness#Epidemiology). Many >> > sufferers >> > > go though life without being aware of it: evidence for its redundancy in >> > > the world of reference cards. (The hue component of color, that is.) >> > > >> > > On Fri, 5 Aug 2022 at 07:06, 'Rob Hodgkinson' via Programming < >> > > programm...@jsoftware.com> wrote: >> > > >> > >> Joining a group of ’similar responses’, I second the vote Devon (and >> > >> others) outlined below. >> > >> >> > >> My only added comment, I printed the 3 pages to compare on paper. >> > >> Although I generally use screens I always believe a new user should be >> > able >> > >> to print/use what they have as cribs. >> > >> This did not work (please test), you will find the text is “too close” >> > to >> > >> the left and right paper margins and gets cropped (does not appear) on >> > all >> > >> 4 edges. >> > >> Admittedly I am printing on A4 (standard in Australia), and perhaps it >> > >> works on US Letter (I can’t test). >> > >> >> > >> I really commend your work Viktor & Henry, this is very helpful to have >> > >> updated, thank you. >> > >> >> > >> Rob >> > >> >> > >>> On 5 Aug 2022, at 8:12 am, Devon McCormick <devon...@gmail.com> wrote: >> > >>> >> > >>> 1. c. footnotes below each table >> > >>> 2. b. about right >> > >>> 3. b. about right >> > >>> 4. c. like it >> > >>> 5. d. yes, and I would like to see names added for the adverb >> > >>> 6. b. both >> > >>> >> > >>> Have a wide landscape version as though one would print it on a >> > foldable >> > >>> card. >> > >>> >> > >>> >> > >>> On Thu, Aug 4, 2022 at 4:27 PM Jan-Pieter Jacobs < >> > >> janpieter.jac...@gmail.com> >> > >>> wrote: >> > >>> >> > >>>> Good job! >> > >>>> >> > >>>> I answered the poll in-line below: >> > >>>> >> > >>>> Op do 4 aug. 2022 om 17:02 schreef Henry Rich <henryhr...@gmail.com>: >> > >>>> >> > >>>>> 1. In each document, look at a couple of tables with a footnote, and >> > >>>>> refer to the footnotes. Which layout did you find easiest to use? >> > >>>>> c. footnotes below each table (easier for on-screen viewing, where >> > I'd >> > >>>> use >> > >>>>> it most). >> > >>>>> >> > >>>>> 2. What do you think of the size of the main text? >> > >>>>> b. about right >> > >>>>> >> > >>>>> 3. What do you think of the size of the footnote text? >> > >>>>> b. about right >> > >>>>> >> > >>>>> 4. What do you think of the coloration (indicating part of speech)? >> > >>>>> c. like it (but I'd double check with a colour-blind person whether >> > >> it's >> > >>>>> recognisable / legible) >> > >>>>> >> > >>>>> 5. Consider the two tables 'Adverbs' and 'Conjunctions'. The adverb >> > >>>>> table has a column of menmonic names, the conjunction table does not. >> > >>>>> Are the names helpful? >> > >>>>> d. yes, and I would like to see names added for the conjunctions >> > >>>>> >> > >>>>> 6. In the Foreigns table, would you prefer to see the system-defined >> > >>>>> name in addition to/instead of the m!:n value? >> > >>>>> a. m!:n only >> > >>>>> >> > >>>>> 7. Do you have suggestions for formatting? Write them in. >> > >>>>> >> > >>>>> >> > >>>> 0) +1 for landscape mode! >> > >>>> >> > >>>> 1) I find it slightly confusing that the footnotes are not in order of >> > >>>> appearance (the first one on the page is 17), and there are some >> > >> footnotes >> > >>>> that appear to be out of sync, but I bet this will be ironed out in >> > the >> > >>>> final version :). >> > >>>> >> > >>>> 2) The arguments table could benefit from a L/R column, there seems to >> > >> be >> > >>>> space available for this. >> > >>>> >> > >>>> 3) I'm still missing some verbs, but I don't know whether you're (in >> > the >> > >>>> long run) going for completeness. >> > >>>> >> > >>>> 4) The modifier train table takes up a lot of space... I recently >> > >>>> summarised it for myself as follows by bunching together like cases (I >> > >> hope >> > >>>> it comes through, best read in fixed-width font): >> > >>>> >> > >>>> Train | Type | Result >> > >>>> -------------------------- >> > >>>> [N] V N | noun | apply verb monad [dyad] >> > >>>> [NV]V V | verb | hook [fork] >> > >>>> ACV V CV | adv/conj | fork after applying AC. note: *no A V C* >> > >>>> A A V | conj | (u A) (*v* A) V ; fork if verbs produced >> > >>>> AC A [A] | adv/conj | ((u AC [v]) A) [A]; apply advs/conjs in order >> > >>>> NV C NV | any | apply C to produce any part of speech >> > >>>> NV C AC | adv/conj | NV C (u AC [v]) >> > >>>> AC C NV | adv/conj | (u AC [v]) C NV >> > >>>> AC C A | conj | (u AC [v]) C (*v* A) >> > >>>> AC C C | conj | (u AC [v]) C (u C v) >> > >>>> NV A | any | apply A to NV >> > >>>> NV C / C NV| adv | bind to C, partial application >> > >>>> A V | adv | (u A) V >> > >>>> A C | adv | (u A) C *u* adverbial hook (e.g. ]: C) >> > >>>> C0 C1 | conj | (u C0 v) (u C1 v) >> > >>>> (foot)Notes: >> > >>>> - AVCN always remain in the result in the order they were in the train >> > >>>> - A and C can take N or V-> any, so e.g. CC can be verb application >> > >> instead >> > >>>> of a hook: 2 (&[.) + is 4 >> > >>>> - AC is adv hook, opposite order of verb hook (and only 'monad') >> > >>>> - monad to adverb: (]. [.) e.g. (].[.)> is an adverb opening its left >> > >> arg >> > >>>> I highlighted in the above with ** what I consider to be less >> > intuitive >> > >>>> combinations. Maybe it's of use if you'd like to reduce the size of >> > the >> > >>>> modifier train table. >> > >>>> >> > >>>> 5) In the adverb table, the dyadic case of x m&u y -> m&u^:x y is >> > >> missing, >> > >>>> and could easily be added as an optional part (as for e.g. ^:). Also, >> > I >> > >>>> think that u@n should be rendered as u@(n"_); u n would seem to imply >> > >> that >> > >>>> the verb is already executed. >> > >>>> >> > >>>> 6) Links to NuVoC for each primitive would also be nice for a final >> > >>>> version, so one can click through for a longer description if needed >> > >> (when >> > >>>> used on a computer). >> > >>>> >> > >>>> All that said, great job, it looks very pretty. >> > >>>> >> > >>>> Keep up the good work, >> > >>>> Jan-Pieter >> > >>>> ---------------------------------------------------------------------- >> > >>>> For information about J forums see >> > http://www.jsoftware.com/forums.htm >> > >>>> >> > >>> >> > >>> -- >> > >>> >> > >>> Devon McCormick, CFA >> > >>> >> > >>> Quantitative Consultant >> > >>> ---------------------------------------------------------------------- >> > >>> 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 >> > >> > >> > -- >> > This email has been checked for viruses by AVG. >> > https://www.avg.com >> > >> > ---------------------------------------------------------------------- >> > 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
