[Date Prev][Date Next][Thread Prev][Thread Next][Author Index][Date Index][Thread Index]
Re: timid lurker seeks help
- To: xanadu@xxxxxxxxxxxxx
- Subject: Re: timid lurker seeks help
- From: ____Textpert Alert____ <ianf@xxxxxxxxx>
- Date: Sun, 17 Dec 1995 00:21:14 +0100
- Reply-to: xanadu@xxxxxxxxxxxxx
Tue, 21 Nov writes Andrew Pam:
> Hyper-G [.. allows ..] users to transclude entire documents,
> though it doesn't currently support fine-grained transclusion
> of document fragments. This will require client support for
> inline text in the same way as the existing <IMG> tag. I am
> still mystified by the omission of this useful concept from
> the HTML standard, and hope that some sort of <TXT SRC=x>
> tag is implemented soon.
Elementary, my dear Pam... ;-)) I'm not mystified at all.
Roughly speaking, allowing translusion of text fragments
within _present_ HTML framework opens up a whole can of, er,
parser worms. For starters it'd mean equipping the clients
with much additional intelligence to distinguish and keep
track of the nested document states, and to suppress and/or
enhance cascades of tags nested within higher-level tags.
I don't think present WWWLibs on which majority of clients
are based are up to job and, anyway, transclusion is too
complex and definitive a concept not to have been designed
in right from the start. Besides, all that would require
more, much bigger memory partitions for the already RAM-
thirsty clients, bigger by a magnitude.
For a glimpse of parsing problems associated with trans-
clusion compare the rendering of any half-complex URL and
that of same page when fed to Zippy filter. For instance:
http://www.suck.som/
http://www.ua.com/cgi-bin/sfx/http://www.suck.com/
__Ian