Found out how StackOverflow handles suggesting friendly urls... prob worth looking into as it handles titles with some international characters and other strange chars.

http://stackoverflow.com/questions/25259/how-do-you-include-a-webpage-title-as-part-of-a-webpage-url/25486#25486

http://meta.stackoverflow.com/questions/7435/non-us-ascii-characters-dropped-from-full-profile-url/7696#7696

/// <summary>
/// Remap International Chars To Ascii
/// http://meta.stackoverflow.com/questions/7435/non-us-ascii-characters-dropped-from-full-profile-url/7696#7696
/// </summary>
/// <param name="c"></param>
/// <returns></returns>
public static string RemapInternationalCharToAscii(char c)
{
  string s = c.ToString().ToLowerInvariant();
  if ("àåáâäãåą".Contains(s))
  {
    return "a";
  }
  else if ("èéêëę".Contains(s))
  {
    return "e";
  }
  else if ("ìíîïı".Contains(s))
  {
    return "i";
  }
  else if ("òóôõöøőð".Contains(s))
  {
    return "o";
  }
  else if ("ùúûüŭů".Contains(s))
  {
    return "u";
  }
  else if ("çćčĉ".Contains(s))
  {
    return "c";
  }
  else if ("żźž".Contains(s))
  {
    return "z";
  }
  else if ("śşšŝ".Contains(s))
  {
    return "s";
  }
  else if ("ñń".Contains(s))
  {
    return "n";
  }
  else if ("ýÿ".Contains(s))
  {
    return "y";
  }
  else if ("ğĝ".Contains(s))
  {
    return "g";
  }
  else if (c == 'ř')
  {
    return "r";
  }
  else if (c == 'ł')
  {
    return "l";
  }
  else if (c == 'đ')
  {
    return "d";
  }
  else if (c == 'ß')
  {
    return "ss";
  }
  else if (c == 'Þ')
  {
    return "th";
  }
  else if (c == 'ĥ')
  {
    return "h";
  }
  else if (c == 'ĵ')
  {
    return "j";
  }
  else
  {
    return "";
  }
}

/// <summary>
/// Produces optional, URL-friendly version of a title, "like-this-one".
/// hand-tuned for speed, reflects performance refactoring contributed by John Gietzen (user otac0n)
/// http://stackoverflow.com/questions/25259/how-do-you-include-a-webpage-title-as-part-of-a-webpage-url/25486#25486
/// </summary>
public static string URLFriendly(string title)
{
  if (title == null) return "";

  const int maxlen = 80;
  int len = title.Length;
  bool prevdash = false;
  var sb = new StringBuilder(len);
  char c;

  for (int i = 0; i < len; i++)
  {
    c = title[i];
    if ((c >= 'a' && c <= 'z') || (c >= '0' && c <= '9'))
    {
      sb.Append(c);
      prevdash = false;
    }
    else if (c >= 'A' && c <= 'Z')
    {
      // tricky way to convert to lowercase
      sb.Append((char)(c | 32));
      prevdash = false;
    }
    else if (c == ' ' || c == ',' || c == '.' || c == '/' || c == '\\' || c == '-' || c == '_' || c == '=')
    {
      if (!prevdash && sb.Length > 0)
      {
        sb.Append('-');
        prevdash = true;
      }
    }
    else if ((int)c >= 128)
    {
      int prevlen = sb.Length;
      sb.Append(RemapInternationalCharToAscii(c));
      if (prevlen != sb.Length) prevdash = false;
    }
    if (i == maxlen) break;
  }

  if (prevdash)
    return sb.ToString().Substring(0, sb.Length - 1);
  else
    return sb.ToString();
}

Thanks Joe,

I know my site has been affected by some of the weird chars that my users have been able to put un urls,

I like how Jeff Atwood's (and stackoverflow products) try to convert the chars, if there is no conversion it strips it out.

My users find the strangest chars in MS Word and throw them in titles...

:)

<add key="UseClosestAsciiCharsForUrls" value="true" /> 

I just tried this key with Chinese Characters and it seems not working well. 

I changed the page name several times with different Chinese characters and the url change between .aspx and -1.aspx alternatively.