W3C home > Mailing lists > Public > www-html@w3.org > February 2001

Re: client side includes

From: Daniel Hiester <alatus@earthlink.net>
Date: Tue, 6 Feb 2001 12:00:41 -0800
Message-ID: <007e01c09077$7c330400$5927b3d1@sol>
To: "www-html" <www-html@w3.org>
snippet:
- - - - - - - - - - - - - - -
> just don't. I've said it before: Currently, in the big two, if there is a
> table (which there ALWAYS is), which sets the layout of the page, then the
> page will not be displayed until the final </table> is recieved. Now,
> while
[DJW:]
For this to be true, the browser has to be NS4, or the
author has to be ignorant.  The latter is normally the case.
With CSS compliant browsers, you can set a style on the table
then use COL elements to enable incremental rendering.  Column
widths are almost always known for the presentational
tables that are "always present".  You can also specify
widths and heights on images to allow correct size place
holders to be displayed.

Murray Wrote:
Most html pages are under 15k while images can easily be larger than that.
Also the HTML file must be partially received before the first image request
is even dipatched, let alone responed to.
- - - - - - - - - - - - - - -

I appologize for over generalizing. Most of the sites I frequent, alas, are
designed in just the way both of you condemn. The most blatent example that
I can think of is mp3.com, and even worse, is the entire gamespy network (
planet_insertvideogamehere_.com, whose html files are usually around the
area of 50-75k, not including any of the images they use). Some sites have
simply too much information that they have to structure (which is what html
is for, correct?). Unfortunately, I don't visit a huge variety of sites, but
from the sites I do visit, I've gotten into the mindset that virtually
everyone in the professional web design field codes their html in the
"uber-table" fashion, as though it were a de-facto standard.

To respond to murray: Now I understand by what you mean, although I don't
think it's fair to call an image a "client-side include" file. I'm thinking
of "csi files" as pertains to just a means of replacing the functionality of
ssi, for those who do not have ssi as an option. A site like cnn.com
certainly has some form of ssi available to them as an option. Chances are,
they do not use ssi on their javascript modules, and css files, so that
browsers can cache those, instead of having to download all of that text at
each page. However, someone building a site on a server without ssi could
not possibly have as much content to present (or information to structure)
as a site like cnn.com, and would logically not use as many different files
as something like cnn.com (or microsoft.com, or netscape.com, and all the
rest)

::maximum email length met::

Daniel
Received on Tuesday, 6 February 2001 14:54:53 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 27 March 2012 18:15:45 GMT