Thinking about this a bit further, there are several "models" that we could choose from, for adding navigation to historical pages, including:
(*) The "errata" model, where the links are supplemental, perhaps with a single link added to the navigation of all relevant pages or perhaps simply linked from a landing page for the dictionary. (*) The "footnote" model, where the links are appended to the bottom of the page, perhaps with a heading such as "See Also" or "Updates" or even "Footnotes". (*) The "advertising" model, where the links are inserted into the page, perhaps at a fixed position or perhaps using a rollover or slide-in mechanism. (This one requires javascript and/or css -- but if we only concern ourselves with popular recent browsers we might even be able to throw everything into jdoc.css. For example.) Other approaches are also possible but probably less desirable (such as updating the "top jump" and/or "bottom jump" nav). That said, ... right now this is all hypothetical, because we do not have adequate documentation for obsolete syntax in nuvoc. And, the structure and quality of that documentation should inform our approach for tying it to the dictionary. Thanks, -- Raul On Tue, Sep 21, 2021 at 5:56 PM Raul Miller <[email protected]> wrote: > > Once these hypothetical Old Syntax pages exist, and are in good shape, > adding navigation from relevant old pages should be fine. > > Thanks, > > -- > Raul > > On Tue, Sep 21, 2021 at 5:15 PM Henry Rich <[email protected]> wrote: > > > > You make good sense, but no one will read the Old Syntax pages unless > > told to. The obsolete articles need to have a pointer to the Old Syntax > > page. > > > > Henry Rich > > > > On 9/21/2021 5:09 PM, Raul Miller wrote: > > > We can learn a lot from history, though sometimes finding the relevant > > > bits can be discouraging. > > > > > > Here, probably what we need is a NuVoc treatment of > > > https://www.jsoftware.com/help/dictionary/dicte.htm > > > > > > Currently, it seems to me, we have stuff like > > > https://code.jsoftware.com/wiki/Guides/Parsing and > > > https://code.jsoftware.com/wiki/Vocabulary/GerundsAndAtomicRepresentation > > > and, of course, https://code.jsoftware.com/wiki/Vocabulary/semico but > > > we do not have anything analogous to the parsing and execution page of > > > the dictionary. > > > > > > Structurally, that part of the system probably has at least three > > > layers. There's the lexical layer (word formation) represented by ;: > > > but we sort of have another layer now used in the context of explicit > > > definitions, which handles the nested {{ }} stuff. And, on top of > > > that, there's the grammatical layer which deals with sentence > > > formation (described concisely in the dicte.htm page). And it's the > > > history of sentence formation which we are grappling with here. > > > > > > Once upon a time, we had something like a fork rule that went: edge > > > cavn cavn cavn and this handled conjunctions in a variety of ways that > > > are no longer supported (I think because of their lack of utility and > > > the esoteric pressure they created). And, for this issue, perhaps we > > > could use an "Obsolete Syntax" page which documents examples of each > > > of those cases, and viable alternatives which are still supported. > > > > > > Am I making sense here? Or am I talking crazy? > > > > > > Thanks, > > > > > > > > > -- > > 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
