Maybe there is enough knowledge in this community to seriously attack the problem of reporting problems. Apparently the free software communities tend to use gnats as their bug coordination system. This is a simple email driven database that collects information from filled in forms that people email to a given address. But that's all it does. We have language processing systems and modules and applications all largely written in a very flexible and widely used language. Doesn't that make it easy to add several levels of debugging output that could be routed to circular buffers? Wouldn't it be pretty easy to collect the contents of such buffers and mail it off to the collection site with the bug report? Wouldn't it be easy to present that information to an expert as an HTML page with internal links to the various bits of information? In particular, wouldn't it be easy to present that information back to the possibly naive user in a way that offers lots of explanation for what has been happening just before the error was encountered? I think this applies to applications and would ease debugging for individual programmers and for teams and for whole user populations. Is that right? Are there problems that get in the way of widely deploying such an enhanced debugging system? Are there problems that forbid applying the same approach to library modules and versions of Mac-Perl itself? Are there already tools for testing how much of your Perl application has been exercised by the test cases you have made so far? Is there any advice offered for building in good testing facilities? Are there other, better ways to make problem solving easier for cooperating communities including newbie programmers doing things somewhat beyond their proven skills? Dick ***** Want to unsubscribe from this list? ***** Send mail with body "unsubscribe" to mac-perl-request@iis.ee.ethz.ch