Hm, I downloaded your code and ran it; it ran, just didn't get the titles, as I think it found multiple title tags and wasn't sure what to do. I changed url to http://www.burvil.org, and it gets the title there OK. Plus, if you print the results of $mechanize, you'll see there's no title field.

As to your original question, as I wasn't able to recreate your problem, I can't help much. However, some changes below might help give you more verbose info, so you might know how to troubleshoot.

• Replacing the top of your code with the following will give you more info:

  !/usr/bin/perl -w
  use strict;
  use warnings;
  [download]

• To print what's in $mechanize, include the following:

  use Data::Dumper;
  [download]

  
  Then, after you've gotten the url, include the following:
  

  print Dumper($mechanize);
  [download]

When I did it for your original xml file, it didn't have a title field, so complained later that it was uninitialized (the latter you'd only get if you use strict;), but when I did it for http://www.burvil.org, it worked OK, and gave a title. That page is a normal HTML page, not xml.

It strikes me as odd that it seems to have been working for you before. There is apparently something I'm missing, but not sure what it is.... In any case, hopefully this will point you in the right direction.

Update: You might also try using XML::Parser to parse the xml itself. If you look at the Dumper output I mentioned earlier, it has the actual XML output; you can get your titles from that.