links in *Apropos* buffer

links in *Apropos* buffer

[Drew Adams]

There is duplication in the links of an *Apropos* buffer: the
command/function/macro/variable/plist/face name has a link that is identical
to one of the links below it (Command, Function, etc.). Apparently, it is
the first such link that is duplicated. And most apropos entries have only
one link below them (they are only a command or only a variable etc.), so
this is nearly 100% duplication.

A user will click both the command name and the word "Command" to see what
they point to, and will discover that they are the same thing.

I propose that we do away with this duplication. Here are some possibilities
(alternatives):

1. Just remove the link on the command etc. name.

2. Have the link on the command etc. name take you to its treatment in an
appropriate manual (for a command, it would do
`Info-goto-emacs-command-node', for instance). Since there might be several
such entries, this might actually access the relevant index entries or it
could do as `Info-goto-emacs-command-node' does (put other entries in
`Info-history').

3. Have the link on the command etc. name take you to the source code that
defines it. As now, whenever there are multiple entries for the same name
(e.g. function, variable, face), the definition of the first of these would
be used. It could be argued that a link on the entry name is too prominent
to send you to source code, especially C code.


I vote for #2. If that's not feasible, then #1.

Re: links in *Apropos* buffer

[Richard Stallman]

    1. Just remove the link on the command etc. name.

    2. Have the link on the command etc. name take you to its treatment in an
    appropriate manual (for a command, it would do
    `Info-goto-emacs-command-node', for instance). Since there might be several
    such entries, this might actually access the relevant index entries or it
    could do as `Info-goto-emacs-command-node' does (put other entries in
    `Info-history').

Those both seem good to me.
Thanks for suggesting it.