frame.pl and robots

Uli Wortmann latex2html
Tue, 16 Mar 1999 12:28:11 +0100


>>>>> "Ross" == Ross MOORE <ross@mpce.mq.edu.au> writes:
 
    Ross> Currently there is a <META....> with an attribute value of
    Ross> 'norobots' Should this be removed ?  ... or retained in some
    Ross> files but no others ?

    Ross> Is it <META .. > tags that robots search, or the <TITLE..>
    Ross> or other things as well ?

Hi Ross,

haven't you got my mail last week? Whether the Meta-tag is considered
by bots I don't know. Most sources claim that only few bots evaluate
it at all. The problem with bots is caused by the fact that currently
all links from a framed node point to a file (file_mn.html) which
contains only the frameset, but no content. AFAIK, bots will skip the
\frame section of a file if it contains a frameset. As a result, the
bot won't follow any of the links of the entry file.

I worked around that problem by specifying NO_FRAMES, which resulted
however in all files of the frameset having the full textinformation
of the node in their \noframe section. I presume, that having all the
links (internal and external) of file.html (index.html), or better
having the full text-content of file.html in the \noframe section of
file_mn.html should be sufficient, since links from file_mn.html to
file-node1.html point to file-node1_mn.html. Thus the whole chain of a
document can be searched. In case of frame-only docs, on should
possibly put only the links into the \noframe part of file_mn.html.

The only special case to be considered is the file index.html which is
most likely produced from file_tf.html instead file_mn.html.


	What do you think?

		Uli

-- 
	Uli Wortmann
	Dept. of Geology       Fax (Switzerland) (1) 632  1080
	ETH-Zuerich            Fon                        3694
        Visit the SPOC-team at http://www.spoc.ethz.ch