Resolving without TLD

Barry Margolin barmar at
Tue Oct 23 14:18:02 UTC 2001

In article <9r32s4$dbg at>,
Eivind Olsen  <eivind.olsen at> wrote:
>--On 22. oktober 2001 21:24 -0700 "(adam brower)" <adambrower at> 
>> With all respect to this august body, I must demur.
>> When I entered "northpoint," let me emphasize again
>> that the actual *url* to which my request resolved was:
>>       http://northpoint/
>> exactly as typed, including the slash. No "expansion"
>> took place. At the time, I went to,
>> just to be sure I wasn't nuts. It was live
>> (with a message of regret.) I'm sure this was
>> some sort of DNS voodoo, and I'm sure there's a
>> witch doctor on this NG who knows the answer...
>Some web-browsers has built-in logic that does this - if you enter a 
>hostname that's not fully qualified, some web-browsers try to connect to 
>www.<name>.com, <name>.com and perhaps a few other variants as well.

But when they do that, they fill in the location bar with the actual URL
that they translated it to.  At least, Netscape and IE do that.  Notice
that it automatically added the "http://" prefix.

If the location field wasn't modified like that, it usually means that it
was "DNS voodoo", usually the automatic appending of a default suffix by
the resolver.  Since the resolver does this without telling the browser, it
has no way of knowing that the URL should be rewritten.

Barry Margolin, barmar at
Genuity, Woburn, MA
Please DON'T copy followups to me -- I'll assume it wasn't posted to the group.

More information about the bind-users mailing list