lynx-dev
[Top][All Lists]
Advanced
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: lynx-dev LYNX: more pleas for the L-page addrs

From: Bela Lubkin
Subject: Re: lynx-dev LYNX: more pleas for the L-page addrs
Date: Tue, 19 Jan 1999 14:58:10 -0800 
David Combs wrote:

> > If your "user mode" is "advanced", you can see the URLs on the status
> > line of the L-page -- one at a time, by moving the cursor to each link.
> > At some point in this discussion I seem to remember your stating that
> > this was not true.  It is for me, and I find it sufficient.
> > 
> Well, that sure isn't sufficient for me.  If that's what you are
> going to do on the L-page, look at the addresses one at a time
> by cursoring to the TITLE of the link (once you have visited that
> site, by someone's DWIM (actually DWHM do what HE/HER meant, back then)
> code, the address disappears and the title comes up in its place) --
> well, why have the L-page at all?
> 
> I mean, you can do that quite nicely on the MAIN page -- cursoring
> to a TITLE, then looking at the status-line.  If that is enough to
> keep you happy, why EVER go to the L-page?

You are still confused about what "title" means, although you seem to
have gotten it by the next message or so.  Nevertheless, let me
reiterate in my own words.

There are actually three things by which we might refer to a page.
There are also three pages involved.

First is the _referring_ page.  This might be somebody's home page,
index.html.  In it we find code that looks like:

  <a href="something_else.html">My k00l links</a>

On your Lynx display, this shows up as a visitable anchor labeled "My
k00l links".  It seems that this is what you were thinking of as the
title.  It isn't; it's just some text marked up with a reference.

Second is the page itself: something_else.html.  When we visit that, we
*may* find that it has a <title>This is the title</title> markup in its
<head> section.  *That* is the title.  Lynx can only know it after the
page has been visited, since it is embedded inside the page itself.  To
be even clearer: the "title" is WHAT THE PAGE CALLS *ITSELF*.

Third is the "L" page.  When we go to the "L" page of index.html, Lynx
shows us titles *if* it knows them; otherwise it falls back to URLs.  I
suspect this behavior is because the original author of the "L" page
really wanted the titles, but since they were not always available,
chose the URL as a sensible fallback.

Why ever go to the "L" page?  For me, it is mainly a way of seeing links
without all the clutter and badly conceived, graphics-oriented layout of
the original page.  I typically use it when I've already absorbed as
much information as I wanted from the page itself, but want to see if
there are any interesting links that I've missed.  It's easy to miss
them in all the clutter.  For this purpose, then, the magic
transformation of URL -> title is actually quite nice, because it tells
me at a glance that I've already visited that link.  I don't disagree
that this could also be presented some other way -- in some cases I
would be happier if it listed only URLs and tagged "*" the ones that I'd
visited.  On the other hand, some sites use totally useless URL names
and benefit a lot from conversion to titles.

> > always the URL.  Then, specially on the L-page, the status line would
> 
> Again, this status-line stuff.  If I want to get an address via the
> status-line, I'll just do that on the MAIN page!  I can see NO reason
> to go to the L-page to do what I perfectly well do on the MAIN page!

I don't have working mouse support and don't like links-are-numbered, so
it can be quite painful getting around on a cluttered page.  Also, I
frequently use Lynx over connections which are both slow and
high-latency.  On a slow connection, it's painful to page through a page
that has lots of text and few links, looking for those links.  On a
high-latency connection, moving the cursor around to the desired link is
painful because the feedback cycle is so slow.

> > show the other choice: that is, either always the URL, or title, falling
> > back to URL.  Then you could set this to "always show URL", and *still*
> > see titles, by moving the cursor to each link.  The advantage of this is
> > that no extra screen space would be needed; the layout of the L-page
> > wouldn't be affected at all.
> 
> But I really want to see the ADDRESS (regardless of whether I've been
> there or not already); that's why I CHOSE to go to the L-page.
> 
> The BENEFIT of the L-page, to me at least (even if to no one else!),
> is that I can lean back in my chair and gaze at the screen and just
> LOOK AROUND without pushing ANY keys (except for going fwd or back
> a screen-page's worth of those addrs).
> 
> Hell, for sitting-back-and-gazing purposes, it would be EVEN BETTER
> to have BOTH the addresses AND THE TITLES TOO.

What you really want is more internal URLs, not more toggles.

  about:http://foo             -- the "=" page
  titles:http://foo            -- the "L" page
  links:http://foo             -- the "L" page using URLs instead of titles
             (and might as well show titles in the status line, when known)
  links-and-titles:http://foo  -- the "L" page using *both* URLs and titles

-- plus a way to bind to a key the notion "modify the CURRENT URL in the
following manner", so that today's commands could be written as:

  "L": "go to 'titles:CURRENT_URL'"
  "=": "go to 'about:CURRENT_URL'"

(and some additional magic that knows to *undo* rather than *repeat*
these things, from inside such a page).

All of the "L" page variants could include "*" for "you've been there".

> Re "without using extra screen space"; why is THAT so important?

Because, in large part, that was the original *purpose* of the "L" page:
to abstract a page down to only its links, to present as many as
possible of them in a short format.

> I say again, if the purpose of the L-page is for
> sitting backing and just gazing around that page,
> ignoring the status line (since you are not cursoring
> anywhere; the only movement being screen-page fwd 
> and back) -- then the more info there to gaze at,
> the better.  Seems to me.

Does it make more sense to you, now that you know that part of the
"info" you're asking for is only known after each page has been visited?

The code has the same philosophy you state: the more info to gaze at,
the better.  Only its idea of "more" is "more links on the screen", not
"more useless junk about each of those links".

>Bela<
reply via email to
[Prev in Thread] Current Thread [Next in Thread]