Well, yes, actually one of my modules follows this naming style. Win32::Daemon::Simple provides a simplified (yet at the same time extended) interface to Win32::Daemon. The thing is that HTML::?Filter doesn't use HTML::TagFilter. Then there's CGI::Simple that provides part of the CGI.pm functionality and thus the code of the module is simpler ... this is not the case either. HTML::?Filter is totally independent to HTML::TagFilter. I did not know about HTML::TagFilter (if it even was already on CPAN which I doubt) when writing HTML::?Filter.

CGI::Lite is a more lightweight version of much of what CGI does. Likewise, OLE::Storage_Lite is a lighter alternative to OLE::Storage.

Note that it is "Lite", not "Simple". Simple indeed indicates a form of inheritance, such as HTML::TokeParser::Simple.

There's not a lot to document so I think the synopsis will suffice:

use HTML::JFilter;

my $filter = new HTML::JFilter <<'*END*'
b i code pre br
a: href name
font: color size style
*END*

$filteredHTML = $filter->doSTRING($enteredHTML);
$filter->doFILE($inputfile, $outputfile);
[download]

I don't like the interface either actually. Maybe the methods should be Filter() and FilterFile(). Or maybe I could provide a functional interface instead or in adition:

my $filter = HTML::JFilter::MakeFilter ($allowed_tags);
$filtered = $filter->( $html);
[download]

Yep, I vote for "Filter()" and "FilterFile()" instead... For the module, what do you think of HTML::Sieve as a name ?