[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: LYNX-DEV blockquote behavior

From: David Combs
Subject: Re: LYNX-DEV blockquote behavior
Date: Sun, 19 Jan 1997 14:52:08 -0800

> From address@hidden Sat Jan 18 20:49:32 1997
> Date: Sat, 18 Jan 1997 22:50:25 -0500 (EST)
> From: Foteos Macrides <address@hidden>
> Subir Grewal <address@hidden> wrote:
> >
> >I really don't think Lynx needs to handle nested blockquotes as you
> >suggest it should.  Whether the author expects nested blockquotes to be
> >acceptable structural markup is an interesting question, the DTD permits
> >it.  It might be wise to have Lynx think of all nested blockquotes as <Q>s
> >but if people are using BLOCKQUOTE where they should be using CITE or
> >simply to get more indentation, that would be counter-productive. 
>       The intent in HTML/3.0 was to use CREDIT blocks within BQ
> (a.k.a., BLOCKQUOTE) blocks, and Q tags for nested quotes within
> the overall block, all of which is supported in Lynx.

Let me say something about the text-->postscript system I use:
it is called "Scribe", was done in about 1977, as CMU phd-thesis
by brian reid (now at dec palo-alto).

Back then, EVERYONE in universities used it, because it was
(a) free and (b) (like sgml/html) had the idea of names, like
"quotation", "enumeration" (list), etc, etc, rather than
"commands" like roff, etc.

Now (and I am making this up) being a phd-candidate in cs,
and AI then being a big thing, he presumably was expert at lisp.
ie, dynamic binding, etc (old lisp).

At any way, that's how Scribe works.  Totally "recursive" --
99% of the constructs operate based on only the current "environment"
(name-value pairs), eg current left-margin, current font, etc.

This, in a nested "quotation", the inner one knows ONLY its
OWN ENVIRONMENT it was invoked in -- with the left margin
already indented over by five or whatever -- and has NO ACCESS
WHATSOEVER to the "outer" environment (the one that existed
before the outer quotation started) -- so OBVIOUSLY the inner
would indent five (more -- but of course it didn't know that
it was five "more" -- just five from the CURRENT left margin).

Now, it seems to me that that is the INTUITIVE best and expected
behavior.  It is the only thing that COULD be called "user friendly".


Now, I know little if anything about sgml (or html) -- except that
I do have two sgml books: the original one ca '87, and that "bible"
of '92 or whenever by Goldberg, I think his name is, the guy who
created sgml.  Worked at IBM, I think.

If html (and sgml itself) didn't automatically use this "environment"
idea, maybe it's only because he came from "different place" way-of-thinking.

Is very
strange to me, all this talk of whether neste

d quotations shouyld do
what seems  like the totally obvious thing.  Maybe someone can
elaborate on this a bit.  Surely I cannot be the only one
who is confused by all this.   

Sounds like when the user explicitly
sOh ays he wants nested quotations, we or html dtd or someone says,
"no you don't -- what you REALLY want, whether YOU know it or
not, is some vertical spacing but no more indentation for the
nested ones.  We know more about what is right for your writing
than you do.  Trust me."

If that isn't a (bad) application of DWIM ("Do What I Mean"), I don't
know what is...

Hope this helps (to confuse matters even more).
; To UNSUBSCRIBE:  Send a mail message to address@hidden
;                  with "unsubscribe lynx-dev" (without the
;                  quotation marks) on a line by itself.

reply via email to

[Prev in Thread] Current Thread [Next in Thread]