Behind the Scenes: Member Profiles Are for Members

There are parts of building SwingCircles that look exciting in screenshots, and there are parts that involve spending an afternoon firmly telling robots to keep their noses out.

Today was mostly the second kind.

One of the principles behind SwingCircles has always been straightforward: member profiles are for members. They should not become search-engine results, AI training material or convenient collections of personal information for automated crawlers.

So today we added several layers of protection around anything that displays a member.

What has changed?

Member directories, filtered member searches and individual profile pages are now explicitly marked so that search engines should not index, archive or generate snippets from them.

The same protection covers WordPress author archives and dedicated user-profile pages that might otherwise provide another route to somebody’s information.

Member-related pages have also been removed from the website’s XML sitemaps. In other words, SwingCircles is no longer helpfully handing search engines a list of places where members can be found.

And then there are the AI bots

Recognised AI crawlers are now blocked across the entire website.

That includes automated crawlers operated for AI training, AI-generated search results and similar large-scale content collection. They are told not to crawl the site and, when a recognised crawler attempts to access it anyway, the server refuses the request.

The only deliberate exception is OpenAI’s user-triggered ChatGPT visitor. That is not an automatic crawler: it is used when I specifically ask Archie—my increasingly indispensable AI co-builder—to inspect or test something on the site.

Archie may visit when invited. His extended robot family may not move in.

More than one “no entry” sign

This was not simply a matter of ticking a single privacy box.

The protection now works across several different layers:

• Member content remains behind the login requirement.
• Member pages carry search-engine noindex instructions.
• The server sends an additional privacy instruction in its response.
• Member listings and author pages are excluded from website sitemaps.
• Recognised AI crawlers are blocked in the site’s crawler rules.
• Automated requests from those crawlers are actively refused.

We tested the finished setup from outside the site. The member profile remained protected, the indexing instructions were present, the sitemap no longer advertised it, an AI training crawler was refused and an invited ChatGPT visit still worked.

Privacy is an ongoing job

No website owner should pretend that a line in a robots.txt file is an impenetrable force field. Respectable crawlers follow the rules; dishonest ones can disguise themselves.

That is why SwingCircles is using several overlapping controls instead of relying upon one polite request. These protections will continue to be reviewed as the site develops and as crawler technology changes.

It is not the most glamorous part of building a new community, but it is one of the most important.

Private. Verified. Chaos-Free. only means something if we do the unglamorous work behind it.