A strange story about MSNBC host Joy Reid has been unfolding for a week. It began when a Twitter user with about 1,000 followers, @Jamie_Maz, dug up what appeared to be homophobic posts on Reid’s defunct blog, the Reid Report. They were similar in nature to posts that Reid apologized for as “insensitive” back in December, after @Jamie_Maz brought those to light.

The new round of posts contain a lot of cliche gay jokes about Charlie Crist and others, concerns that “adult gay men tend to be attracted to very young, post-pubescent types, bringing them ‘into the lifestyle,’” and commentary like “part of the intrinsic nature of ‘straightness’ is that the idea of homosexual sex is ... well ... gross ... even if you think that gay people are perfectly lovely individuals.”

The triumph of the gay-rights movement has been so complete and fast that it’s easy to forget that 10 years ago—in the same election that swept Barack Obama to the White House—California voters passed a state constitutional amendment banning same-sex marriage. Attitudes changed, the moral arc bent, and now, a lot fewer people disparage gay people like this than did in 2006. A liberal talk-show host would and should be embarrassed and ashamed by these posts popping up, but Reid apologized once, and could have done so again.

Instead, Reid released a statement to Mediaite saying that she’d been hacked and was not responsible for the posts. “In December I learned that an unknown, external party accessed and manipulated material from my now-defunct blog, The Reid Report, to include offensive and hateful references that are fabricated and run counter to my personal beliefs and ideology,” Reid said.

The posts had been dug up on the Internet Archive’s Wayback Machine, which maintains copies of many pages on the web. When Reid said she’d been hacked, many jumped to the conclusion that it was the Wayback Machine that had been hacked. On its blog, the Internet Archive said that Reid’s lawyers had contacted them about a possible hack, but that they had no indication that one had occurred.

“This past December, Reid’s lawyers contacted us, asking to have archives of the blog (blog.reidreport.com) taken down, stating that ‘fraudulent’ posts were ‘inserted into legitimate content’ in our archives of the blog,” they wrote. “Her attorneys stated that they didn’t know if the alleged insertion happened on the original site or with our archives (the point at which the manipulation is to have occurred, according to Reid, is still unclear to us).”

On review, the Internet Archive “found nothing to indicate tampering or hacking of the Wayback Machine versions.”

The Wayback Machine has been archiving posts for years and years, and in many instances, it re-crawls URLs. A blog post that went live in 2006 might have been indexed in 2007 and 2009 and 2011 and 2017. This is important because if Reid’s blog was hacked to insert new posts with old dates, the copies in the Internet Archive’s repository would have recent dates, even if they showed old time stamps on Reid’s site.

It’s not possible to view the Internet Archive’s public stash of Reid blog posts because Reid’s team recently inserted code into the site that prevents the Internet Archive from indexing it.

But the Internet Archive’s Brewster Kahle confirmed to me that the Wayback Machine crawled Reid’s site back in the 2000s—and that there was nothing suspicious about the way the posts appeared in the archives.

“We saw [the blog posts] in the ’00s soon after they were dated on the blog, and [they were] archived in normal course of operations,” Kahle told me.

Furthermore, the Old Dominion computer scientist Michael Nelson scoured the other public records of Reid’s site to see if he could find evidence for the screenshots that @Jamie_Maz posted. Even without precise URLs, which makes it much easier to look up the records, Nelson was able to identify five matches in the various archives from the screenshots.

There is a small caveat, which is that these other copies probably used the Wayback Machine’s copies as their original source, even if they have stored them separately.

“Nonetheless,” Nelson wrote, “with multiple copies geographically and administratively dispersed throughout the globe, an adversary would have had to hack multiple web archives and alter their contents (cf. lockss.org), or have hacked the original site (blog.reidreport.com) approximately 12 years ago for adulterated pages to have been hosted at all the different web archives.”

As he notes, it is possible that someone did this, but “extraordinarily unlikely.”

Among the implausible factors is that for Reid’s story to be correct, someone would have had to hack her blog back in the ’00s, but no one, including Reid herself, noticed the invalid posts.

Why is this all happening now, when Reid is famous and one of the most prominent women of color on news television?

It seems it really was due to the relentless digging of a single Twitter user, @Jamie_Maz. Reached via direct message, the user explained,“after her apology, I started looking through the posts again and found things that were problematic related to LGBTQ+ issues that were not related to Republican hypocrisy on gay marriage, which her apology cited as part of the explanation for what she wrote about Crist.”

Maz, who often tweets about archival finds and social justice, spent chunks of time in December and January reading through the posts, month by month, and screenshotting examples of what she found.

The Internet rots faster than anyone could have guessed. Almost every word written on the web gets flushed down the memory hole, sinking into obscurity; it’s basically impossible to find any reference to the original Reid Report just using Google.

But, as Joy Reid is finding out, all it takes is one person to chisel electronic words into a more permanent form.