Re: Popup windows: problems and some solutions

Chris Josephes (cpj1@winternet.com)
Fri, 22 Mar 1996 21:03:30 -0600 (CST)


Date: Fri, 22 Mar 1996 21:03:30 -0600 (CST)
From: Chris Josephes <cpj1@winternet.com>
To: Heikki Vesalainen <wes@clinet.fi>
Cc: The W3 HTML group <www-html@w3.org>
Subject: Re: Popup windows: problems and some solutions
In-Reply-To: <3152C6B6.5674@clinet.fi>
Message-Id: <Pine.SUN.3.91.960322204928.24924A-100000@subzero.winternet.com>

On Fri, 22 Mar 1996, Heikki Vesalainen wrote:

> Chris Josephes wrote:
> 
> > Your post to the previous digest clearly mentioned TARGET="_popup"
> 
> There has been a lot of talk after my first proposal on this subject.
> And in the problems & solutions text that you replyed to I clearly said, 
> that the targetting way will bring up problems with old browsers.
> 
> > But the point of the A element is to serve as an anchor.  I doubt that
> > the NAME attribute will go away due to it's amount of usage, but <A 
> > NAME> dosn't necessarily mean that the element will be a Footnote.  It
> > could just be a target to another point in the document.
> 
> Well it should go.(and be replaced by ID)

No arguement there, but a LOT of web pages do use NAME because ID isn't 
implemented .

> 
> > <FN> clearly identifies the selected text as a footnote.  Thus, the
> > browser, serach robots, etc, etc, clearly understand the purpose of
> > the enclosed text.
> 
> I think the browsers should know it's a popup window before they start 
> loading it.  Actually they must know, because otherwise they will first 
> clear the current page, then (as they receive the FN tag) they will have 
> to redraw the page they just cleared and display the popup note into it.
> 
> If we declare it a popup window (and meaby it's dimensions) before the 
> data is even received, the browsers won't have this problem.

Currently, a lot of browsers would have this problem, because none of 
them implement footnotes.  Also, how could a browser really guarauntee 
that the next file it is loading IS a footnote, or just a badly formatted 
HTML file?

>  
> > Huh?  You're saying <A NAME> dosn't require a closing </A>?  That 
> > dosn't sound right.
> 
> I'm saying that officially the </A> is required, but think from the 
> programmes point of view.  What would it do with it?  What practise would 
> it have? NONE! In the <A HREF> tag they nead the </A> to determine, where 
> the end of the sensitive area is, but with <A NAME> the case is 
> different.
> 

With your implementation I am going to guess that you are proposing sticking 
each individual footnote in a seperate file, which IMHO is wasteful.  Why 
not use the FN tag, and stick all the footnotes at the end of the file, 
that way it'd be a lot easier to edit them.  

> > Tables mess up older browsers.  I'm not saying we should always jump on
> 
> Tables should not mess up old browsers (the data will just be displayed 
> in the order it is in the HTML file).  If it does, then the old browsers 
> are not good.  The use of FN will mess up even the good old browsers.

By "old browsers", I mean browsers that can't handle table tags.
Try displaying a page with a table in Mosaic 1.0, or Netscape 1.0, 
especially a table with cells expanded down several rows.  I can 
guarauntee that it will not look good at all.

Yes, FN would mess up an old browser, but so would your new usage of the 
A/NAME tag with the name attribute because it is going by the old behavior.  
Like I said before, it's a hell of a lot easier to create a new tag than 
it is to redefine an old one.

> 
> > If you're really concerned about implementing footnotes right now, you
> 
> I'm consearned about there being too many tags for nearly same purposes.

I can understand the arguement that HTML has a few really bad tags that 
do tend to be abused, but there is such a thing as being TOO conservative 
on this issue.  It's not like the FN tag was just implemented by some 
cheezy browser company, it was included in the HTML 3.0 proposal.

> 
> -- 
> Heikki Vesalainen
> http://www.clinet.fi/~wes/
> wes@clinet.fi
> 

----------------------- Christopher P. Josephes ----------------------------
Email |  mailto:cpj1@winternet.com
Web   |  http://www.winternet.com/~cpj1/