After sleeping on this... It's worth pointing out that the https://www.jsoftware.com/help/dictionary/partsofspeech.htm page does list the parts of speech for each primitive, and does link to dictb.htm for verbs.
That's easy for newcomers to overlook and easy for experts to have forgotten about, so it is a bit weak in that sense, which led to our current discussion. That said: NuVoc suffers similarly, and its sprawling layout seems to me to be more difficult to scan (and, for example, seems to claim that 5!:0 is a verb). The original dictionary treatment -- with key data at the top and examples further down is much easier to cope with than the NuVoc approach where key reference information is frequently lower on the page than examples. (And, I do not know if a complete rebuild of NuVoc to restore some dictionary structures would be acceptable or would be considered too disruptive to the current efforts). Thanks, -- Raul On Tue, Jun 28, 2022 at 1:53 AM Henry Rich <henryhr...@gmail.com> wrote: > > You are right. There are several places where the Dictionary is incorrect > or incomplete. To preserve the speed and style of the Dictionary, we > decided to leave it as is. NuVoc Is now the official definition of the > language. > > Henry Rich > > > > On Mon, Jun 27, 2022, 11:31 PM Elijah Stone <elro...@elronnd.net> wrote: > > > I don't know if there is anything to be gained by rehashing this argument, > > as I failed in the past, but I will try one more time: > > > > The dictionary states that x u@v y may be substituted for u x v y. In > > fact, x u@v y may not be freely substituted for u x v y, because there > > are cases where such a substitution would change semantics. > > > > That the dictionary states something incorrect is a bug in it. > > > > On Mon, 27 Jun 2022, Raul Miller wrote: > > > > > On Mon, Jun 27, 2022 at 3:30 PM 'Pascal Jasmin' via Programming > > > <programm...@jsoftware.com> wrote: > > >> You have brought up a bug in dicitionary, that has survived to J6.02 > > > > > > Is a characteristic a bug if it's by design? > > > > > > The dictionary writeup on verbs -- > > > https://www.jsoftware.com/help/dictionary/dictb.htm -- describes the > > > features of verbs which are common to all verbs. This is information > > > which would need to be repeated on every dictionary page for the > > > dictionary to completely describe the characteristics of those verbs > > > on each page. (And, venturing partially into this territory is why > > > NuVoc pages tend to be rather long.) > > > > > > Anyways, this writeup on verbs includes this text: > > > > > > "Finally, each verb has three intrinsic ranks: monadic, left, and > > > right. The definition of any verb need specify only its behaviour on > > > cells of the intrinsic ranks, and the extension to arguments of higher > > > rank occurs systematically. The ranks of a verb merely place upper > > > limits on the ranks of the cells to which it applies; its domain may > > > include arguments of lower rank." > > > > > > That said, I think that each of the dictionary's verb pages should > > > explicitly state that the verb is a verb, and that statement should > > > link back to this dictb.htm which details what a verb is. > > > > > > In other words, where > > > https://www.jsoftware.com/help/dictionary/d620.htm currently has: > > > > > > <table width=100%><tr> > > > <td align=left width=33%><b><font size=+2>Atop</font></b></td> > > > <td align=center width=50%><font face="Courier New" size=+3>u@v > > > mv lv rv</font></td> > > > <td align=right width=17%><b><font size=+2></font></b> > > > </tr></table> > > > > > > I think it should have something like: > > > > > > <table width=100%><tr> > > > <td align=left width=23%><b><font size=+2>Atop</font></b></td> > > > <td align=left width=10%><a href="dictb.htm">verb</a></td> > > > <td align=center width=50%><font face="Courier New" size=+3>u@v > > > mv lv rv</font></td> > > > <td align=right width=17%><b><font size=+2></font></b> > > > </tr></table> > > > > > > ... and all of the other dictionary pages describing primitives should > > > have a similar reference link. > > > > > > And, that said, note that I am focusing here on utility rather than > > > appearance -- someone with more of an eye towards appearance may be > > > able to take this general approach and come up with something more > > > appealing which: > > > > > > (*) Explicitly states, in the definition, the type of word being > > defined, and > > > (*) Links that type statement to the corresponding page which spells > > > out what this means in the context of the language definition. > > > > > > And... that said... the omission of this kind of detail could be > > > considered a bug (though there's probably a better way of describing > > > this kind of problem). > > > > > > Thanks, > > > > > > -- > > > Raul > > > ---------------------------------------------------------------------- > > > 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
