I think what you might want to look into, is LWP::Simple, using the getstore() method, which would look something like this, using your example:

   use strict;
   use LWP::Simple qw/getstore/;

   my $site      = getstore('http://www.google.com',
                            'test.html');
   # Or you could store the content in a scalar, with: 
   # my $content = get('http://www.google.com');
[download]

But you also mentioned you wanted images, so you'll want to look into using HTML::SimpleLinkExtor as well, which is used thusly:

   use strict;
   use HTML::SimpleLinkExtor;

   my $extract = HTML::SimpleLinkExtor->new();
   $extract->parse_file("test.html");
   # or you could use $extract->parse($content), as shown
   # above, using the raw HTML
   my @images = $extract->img;

   # feed each element of @images back to a getstore() for
   # each image found
[download]

If you want to maintain "browsability" of the site when you're done fetching it locally, you'll want to look into rewriting the links found in the document. I would suggest using URI to get the absolute url of the links, fetch those, and save them locally to disk. If you want to maintain the same filenames as referenced in the main HTML page you fetched, you might want to either regex off the filename portion itself, from the link you get, or use File::Basename to chop it off programatically. Here's a regex that can help split the file from the URL you get with URI's new_abs() method:

    $file =~ s/.*[\/\\](.*)/$1/;
[download]

Why would you want to use these modules instead of OLE? Because they're portable, and if your code happens to move from Windows to Linux or BSD or some other non-OLE-supporting variant, it will still continue to work, without you having to rewrite it.

Good luck in your quest.