ok I see what you are doing. I'd have used a regexp without split as many others are suggesting, e.g.:

perl -F"[:;]" -lane '
    print "<a href=mysite.net/$_ >$_</a>"
      for map { /^\s*(\S+)/; $_ = $1 } @F;
'
[download]

Because in the awk you only wanted $1 so don't need to run the regexp more than once per line (line after the -F implied split that is).