Can you give a short example showing when a program would want to use these new methods? Thanks.

Sure. I am writing a static site generator which renders pages by combining a layout HTML file with HTML files containing the content of each page. The repetitive parts of the pages reside in the layout file and the unique content are in the content files. The content files start with a front-matter containing metadata about the page, such as its title and description.

For example if we have the content page

<!--{
"title": "The title of the page",
"description": "Page description"
}-->

<p>here is a paragraph</p>

<h2>And a subheading</h2>
<p>And another paragraph</p>

the program would first insert all elements after the front-matter into the layout. It would then insert a h1 and p based on the metadata:

// Insert the page title heading
pageTitle := &html.Node{
Type: html.ElementNode,
Data: "h1",
DataAtom: atom.H1,
}
pageTitle.AppendChild(&html.Node{
Type: html.TextNode,
Data: titleFieldInFrontMatter,
})
main.PrependChild(pageTitle)

// Insert the page description paragraph
pageDescription := &html.Node{
Type: html.ElementNode,
Data: "p",
DataAtom: atom.P,
}
pageDescription.AppendChild(&html.Node{
Type: html.TextNode,
Data: descriptionFieldInFrontMatter,
})
pageTitle.Parent.InsertAfter(pageDescription, pageTitle)

Sorry, this did not turn out short after all :D

/cc @nigeltao

This proposal has been added to the active column of the proposals project
and will now be reviewed at the weekly proposal review meetings.
— rsc for the proposal review group

The methods that are already there - AppendChild and InsertBefore - are a complete set in the sense that they provide all possible functionality. Do we really need to add a second complete set?

Also, if your program finds that it needs these frequently, it is easy to write them as local helper functions, either using AppendChild and InsertBefore or using direct manipulation of the Nodes.

Based on the discussion above, this proposal seems like a likely decline.
— rsc for the proposal review group

No change in consensus, so declined.
— rsc for the proposal review group