TL;DR: check out Don't Moderate Me and sign up at dontmoderate.me.
People often consider social media, news sites, and online forums as public places for free exchange of ideas, but they are not. Behind every web site is a party, be it a megacorp or an individual, who exercises complete control over which content is served to users. You won't notice this if you avoid controversial topics, but ask anyone who has engaged in online potitical speech, or posted a critical product review to a shopping site, only to have their words redacted without explanation.
Content moderation is a force of good and evil. Moderators increase the quality of a discussion by removing ad-hominem attacks and spam, but they also silence uncomfortable or minority-held opinions which deserve fair consideration. Of course the owner of a site can remove content that they don't wish to serve - it is their site, after all! Still, there are troubling implications for the major social news and media sites where people engage in public (perhaps 'democratic') discussion. As we invest more social capital in these privately controlled venues, we become more vulnerable to editorial control that may not align with the collective interests of users.
One implication is that content moderation is often silent. A contributor's words may be removed completely, edited to fit someone else's agenda, or placed in a different context than the author intended, and he or she may never realize it. This is rare in some online communities but very common in others! Moderation is even less visible to subsequent visitors who have no idea what changes may have occurred before they arrive.
If a web site is publicly visible, then we have the capability to notice when content is changed or removed, so why don't we notice?
Unfortunately for readers, there are no universal/reliable "highlight changes" or "show previous version" buttons to see a page as it was served yesterday or a month ago. Enormous problems of scale thwart existing efforts toward this, in short because their goal is to collect and store a chronological history of all web pages on the internet ("over 472 billion" as of this writing).
The problem is tractable, however, when considering the needs of content authors, or anyone concerned about specific content on specific pages. For these targeted situations, a service could watch content on your behalf and notify you when it changes. Don't Moderate Me is that service.
Basically, you sign up and create some monitors. A monitor says "watch this text on this URL". Create one whenever you're concerned that something on a web page will be modified or removed. Every day, Don't Moderate Me loads the page and looks for the text that you specify. An audit trail is kept along with screenshots, and if the specified text ever changes or disappears, we notify you with an email. Using Don't Moderate Me, you can rest easy with the knowledge that if your content is moderated in the future, you'll be informed and equipped with evidence. Of course, what you do with that information is up to you.
Other services work by analyzing images of a page and detecting visual changes that exceed some threshold, or by detecting any changes on a page at all. DMM saves images too, but makes decisions based on text. Other services will generate false alarms when the style and presentation of a site changes, or when other contributors post new content after yours, but Don't Moderate Me tries not to. (DMM isn't intended to replace things like thread subscriptions and RSS feeds.)
One more caveat: Don't Moderate Me is a proof-of-concept and important problems remain unsolved. Certain pagination schemes are likely to cause false alarms, and if someone is determined to trick the service, they can hide your content from human readers while serving it in a way that a computer can still read (like white text on a white background). Still, I encourage you to try Don't Moderate Me, see if you can break it, and provide feedback.
Check out Don't Moderate Me and sign up at dontmoderate.me.