TheOptimizationIdiot - 12:17 am on May 2, 2013 (gmt 0)
If the response is 410, the HTML content is irellevant to bots.
Well, you would think it would be, but there have been occasions where Google has indexed 404 pages, including their content, so I try to "err or the side of caution" and make sure they "get what I mean" without any question when I remove a page.
Also, the content of a page, even 410 Gone, is not irrelevant to search engines. I don't remember where I read (or heard) it (could have been a video), but it was from an "official source" and was something about them evaluating error pages for user experience and a custom error page that provides relevant links is a "good sign" to Google at least.
Thinking those error pages, 404, 403, 410, etc. "just don't count" and a SE will not ever take into account what's displayed is a bad move, imo. My personal opinion is: Providing a good visitor experience with a custom error page and relevant links (if possible and "important" links if not) while making sure search engines "get the point" the page is not there for them and not intended to manipulate anything is a much better route.
And, thinking they "don't know" what's on an error page isn't a good idea, because to "not know" they have to use a HEAD request rather than a GET, but you won't see HEAD requests from gBot (or bBot) in your logs, because when they request a page (URL), they use GET and they GET the whole page (URL) just like a visitor would, so they know exactly what's on it, regardless of status code served in response to the request.
Lastly, if what you're saying is correct and I'm not, then the noindex,nofollow on the page does no harm at all, so why not "err on the side of caution" and include it?