Re: HTML should not be a file format, but an output format

Stephanos Piperoglou (
Sun, 23 Mar 1997 21:53:22 +0200 (EET)

Date: Sun, 23 Mar 1997 21:53:22 +0200 (EET)
From: Stephanos Piperoglou <>
To: "nemo/Joel N. Weber II" <>
Subject: Re: HTML should not be a file format, but an output format
In-Reply-To: <>
Message-ID: <>

On Sat, 22 Mar 1997, nemo/Joel N. Weber II wrote:

>    Why not? The structure of a web site is arbitrarily divided up into 
>    multiple files. Each page and each picture must be in a separate file. 
>    Server-side image maps have to be in separate files. CGI scripts have to 
>    be in separate files. Is this the best way to edit a site?
> I think I now see your point.
> OTOH, I don't see a problem with spliting it across several files.  I
> usually split computer programs I write across several files; why shouldn't
> I split web pages similarily?

So do I, BUT

What if I want to create a large site, where every page has a certain header
and a certain footer, and some pages are dynamically created while others
are not, and some are mixed, and all fit into a tree-like structure.

I do NOT want to have that header and footer on every page. I do NOT want
the "next, prev, up, home, index, toc, glossary, help" etc. links static in
files so I have to update them every time I change a single document. I do
NOT want to resort to special mechanisms for my dynamically generated

A Web page, to me, is a complete HTML document that is served out by a Web
server. The way that it is created (running an external program, pasting
together header, section, footer and dynamically created navigation links
and so on) is irrelevant. But in 99% of the cases it is not practical to
have this HTML document in a single file on the server's disk. And in 99% of
the cases it isn't.

You see, most files on a web server that include server-side includes or
server-side script languages like PHP/FI or LiveScript are NOT valid HTML.
That confuses people. They do consider these standard HTML, they create
pages that don't use proper hyperlinks, because they're not educated in a
situation like this where hyperlinks have to be properly designed.

A CGI binary is not an HTML document. It *produces* an HTML document,
actually several HTML documents, depending on circumstances. As a matter of
fact, it doesn't, now that I think of it. It produces something that is a
combination of HTTP headers *and* an HTML document. So you see my point.
HTML is not a good storage format because most information on the Web is
stored in ways that HTML can not cater to.

Stephanos "Pippis" Piperoglou -
I've never finished anything I began, but this time I'm