Re: lynx-dev dev.16 patch 5 corrected

[Klaus Weide]

On Mon, 13 Dec 1999, Henry Nelson wrote:

> > Just to make my standpoint clear, not to start an argument: "better
> > safe than sorry" could cut both ways, depending on who was right on
> > the technical points, and in my opinion is an argument for making
> > --enable-internal-links the default.  But I will leave it at that and
> 
> Perhaps, then, emphasis of "prevent" is in order.  (I debated, but left
> it not; see below.)

It's ok as it is, IMO.

> > >         to interpret URL-references as suggested by RFC 2396, and to 
> > > prevent
> 
> You or Tom might emphasize that "prevent," e.g., "*prevent*," or word it
> more strongly: "not allow" or "block."  I do hope you will agree with
> "attempts to" over categorical expressions, however.

Fine with me, and actually preferred as you wrote it.

All I can say is that the code "attempts", I can't make any guarantees that
it succeeds.  There is the possibility of implementation errors.  It is
also possible that there are questionable "design choices", that would be
discovered, discussed, and possibly corrected if anyone but me ever *looked
at* this stuff.  (For example, my interpretation is that with the NOCACHE
('x') key the user explicitly asks for reloading, and so resubmitting of
POST content, without further prompting, is allowed in some situations when
'x' is used.  Ditto even for 'd'ownload.  Review by others - which to my
knowledge has never happened, except for Fote's wholesale & undetailed
rejection - might well show that these assumptions on my part are wrong.)

> > A nit: should "predicts" be "has predicted"?  I don't really care though.
> 
> My main purpose in making the suggestion was to depersonify the subjects,
> but keep the active voice when possible.  Using "predict" in the present
> tense means that the "alternate opinion" has not yet been [unequivocally]
> refuted (proven wrong), and also hints to the reader why the default is the
> way it is.  Admittedly my biased opinion since Fote never gave us an
> example that we [=users] could empirically test to see his point.

Ok, so let's leave it like that.

I recently gave
   <http://enteract.com/%7Ekweide/test/addition.cgi>
and
   <http://enteract.com/%7Ekweide/test/test-postlink1.html> as demos, in
response to Leonid.  They are designed to show the differences between
--with and --without-internal-links, to a visitor willing to experiment
a bit.  We don't know what examples Fote would have chosen to
demonstrate his point, so this is the best I can offer (until/unless
there are more specific questions).  You (everybody) are invited to test
/ explore / critique the behavior of lynx with those, or make up better
tests.  Feedback still appreciated.

   Klaus