I have a FreeBSD 5.4 system. There are several files that have been
created in MS Word and saved in UTF-8 format that I need to convert.
These files have the BOM attribute.

Here are a few links regarding UTF-8:

http://www.unicode.org/faq/utf_bom.html
http://www.cl.cam.ac.uk/~mgk25/unicode.html

I tried writing a simple Perl script to remove this attribute. This is the script.

#!/usr/local/bin/perl
@file=<>;
$file[0] =~ s/^\xEF\xBB\xBF//;
print(@file);

It appears to work, but it does not overwrite the file that it is
suppose to be fixing. I am at a lose as to what to do.

In addition, since this script may be included into a larger program,
it really should be able to use the 'use strict' attribute. If I use
it now, the program will fail.

I can supply examples of the files that need to be cleaned of the BOM attribute.

if you want to use strict, you must make the file variable local:

use strict;

my @file=<>;
$file[0] =~ s/^\xEF\xBB\xBF//;
print(@file);

The problem is not in the Perl language, but in the operating system.
Eigher UNIX or Windows do not allow to overwrite opened file. The
diamond operator opens a file (it calls open system call), the
operating system locks the file and it cannot be opened for writing.
No process is allowed to do that, even the process that opened it for
reading.

You need to write a script that will open the file in read-write mode
and "shift" the file content three bytes back. You have to use
open/read/close functions for that and a memory buffer (4096 bytes is
enough).

You can still use your solution, but you need to redirect the output
to a temporary file and then rename it back to original.

Hope this will help.