W3C validators and url encoding

W3C validators do not encode urls correctly in the links they make:

If I try to validate an url containing &, for example
by specifying:
  http://validator.w3.org/check?uri=http://somehost/?a=1%26b=1

I get the result correctly, but the links the page gives, are like:
  http://validator.w3.org/something?uri=http://somehost/?a=1&b=1
  - which gives b=1 to the validator instead of keeping it as a part of the url to be checked.

The validator services (all of them) should encode the urls correctly.

I hope I made my point clear...

-- 
Joel Yliluoma - http://iki.fi/bisqwit/
This message was sent with Winnie III