Re: Why is there a necessity to submit your sitemap?

> Perhaps ... they don't want other search engines to profit from their
> work? :-)

Wouldn't that be a bit evil :-), as it's our work too, indeed we are
saving
the GoogleBot work by doing this. Anyway, the other search engine
would need to put in _a_lot_ of work to get a system which does every-
thing Google's does already.

> Also consider everything else in the Google Sitemaps control panel --
> you would still need to submit your sitemaps manually in their system
> to get any sort of similar power.

Yes, but that is not the feature I'm after.

> IMHO it doesn't make sense - at the moment

Of course, Google's sitemap feature has only been out a few months,
though, in the long-term, such a feature could benift everyone -- the
users especially.

> (but is certainly a great idea!).

Thanks :-)

> For an optimal coverage of most modern search engines, add a
> RSS-Feed for your site (list all your pages) and link that on your
> web-pages.

Nice idea, though what I'm interested in is telling the search engine
which pages have been updated, rather than just where those pages
are located. We already know that GoogleBot maintains some sort of
log of when it visited a site last, so if it loaded 'sitemap.xml' it
would
immediately be able to see that I modified or created some file nested
deep in the subdirectories and cache that file. Currently, however,
the bot is forced to blindly scrabble about my site in the hope of
finding new/updated pages; needless to say, this is most inefficient,
wasting both Google's and my bandwidth as well as running the risk
of not finding the updates after all that work.

To reiterate, I can see Google Sitemaps solving the problem of this
waste, but the method of submission of the sitemap url seems, at
least to me, wrong -- and at best inefficient -- though I could give
Google the benift of the doubt and assume that this is only because
Sitemaps is still in the beta phase and this issue has not yet been
considered. (Hopefully things will go faster than with groups, which
seems to be in eternal beta :-p .)

--
C
2005-12-03