I type this:

<tr valign="baseline">
    <th align="left">
        <label id="editionLabel" for="editionDropList" runat="server">#Edition:</label>
    </th>
    <td>xx</td>
</tr>

Then I switch to the Design view to drop a DropDownList control between the two xs. The idea is that it insert the pseudo-XML describing the control in to the pseudo-HTML of the page. What it actually doesis as follows:

<tr vAlign=baseline>
    <th align=left><label id=editionLabel 
        for=editionDropList 
    runat="server">#Edition:</LABEL></TH>
    <td><asp:dropdownlist id=editionDropList runat="server">
                <asp:ListItem Value="test">#Test</asp:ListItem>
                <asp:ListItem Value="live" Selected="True">#Live</asp:ListItem>
                    </asp:dropdownlist></TD></TR>

My tidy layout is arbitrarily scrambled. End tags are needlessly sucked on to the end of line and then random whitespace added to the starts of lines. Some attribute names are sTrangely cApitalized fOr nO rEason; some tag names are RANDOMLY CAPITALIZED which means the page is no longer well-formed XML, let alone valid XHTML (as if!). Quotation marks have been stripped from attribute values! Why?!

Even if the whole page has to be parsed and written out again as part of the process of inserting their control's code in to my text, would it not have been good form to make the mechanically generated code better-formatted rather than in every way worse?