They didn't lose protection. They simply failed to establish such protection at the summary judgment stage.
In light of the summary judgment record, we conclude
that there are genuine issues of material fact as to whether
the moderators are LiveJournal’s agents. The factual dispute
is evident when we apply common law agency principles to the
evidentiary record.
DMCA safe harbor is an affirmative defense. An affirmative defense means that the defendant bares the responsibility to prove the defense. That can be very hard to do at the summary judgment stage when there are legitimate factual issues in dispute, as there are here.
The central issue here is that one of the moderators was an employee of the defendant. Notably, the moderators prescreen and curate material for publication. Furthermore, there was conflicting testimony about the role of the employee-moderator and his relationship with the other moderators, such as the extent to which he directs and guides them.
The only way the judge managed to discard the case at the summary judgment stage was by holding that the common law of agency didn't apply, not even with regard to the employee-moderator alone. (Which is to say, by holding that the factual dispute about the employee-moderator's knowledge, role, and relationship was irrelevant.) I haven't read his opinion so I have no idea how he arrived at that holding, but on its face it seems unsupportable.
> They simply failed to establish such protection at the summary judgment stage.
Of course. They lost the ability to use the standard affirmative "safe harbor" defense that many internet services currently rely on. At least one copyright lawyer believes[1] this may have significant implications for other places that moderate user content. Reddit and YouTube were specifically mentioned as possibly affected due to their own appointment of moderators.
Obviously, the details are decided in the retrial. However, the previously mentioned lawyer[1] suggests this puts LiveJournal in a poor position that encourages them to settled out of court, leaving the practical utility of safe harbor uncertain when appointed moderators are involved.
> DMCA safe harbor is an affirmative defense.
Yes, that was a key part of opinion.
> an employee of the defendant
Agent of the defendant, which includes non-employee unpaid positions. Regardless, I agree that the status of the moderators is indeed the central issue.
> on its face it seems unsupportable.
I agree, and so did the 9th circuit appellate court. That district covers YouTube and a lot of other internet businesses, which ads a lot more uncertainty for any of those businesses that moderate content and might be seen as having an agency relationship with their moderators.
> They lost the ability to use the standard affirmative "safe harbor" defense that many internet services currently rely on.
No, they retain the ability to use it. It's just that, at summary judgement (which views all disputed factual questions in the light least favorable to the party seeking summary judgement) their success in using that defense has not been established.
Which means that the elements of the defense will need to be established at trial, which is the norm for an affirmative defense.
> Obviously, the details are decided in the retrial.
Trial, not retrial. There was no first trial; the district courts (now reversed) summary judgement decision occurred before trial.
> That district covers YouTube and a lot of other internet businesses, which ads a lot more uncertainty for any of those businesses that moderate content and might be seen as having an agency relationship with their moderators.
Lots of sites with user-submitted content that have moderators engage in reactive moderation; LiveJournal, with ONTD, had moderator pre-clearance of content for posting. Even if moderators are agents, if they aren't approving content to be posted, their existence doesn't raise the issue that the content may be posted at the direction of the site owner (via an agent) rather than the outside user. It might raise issues of red-flag knowledge, but without blanket pre-clearance of the type LiveJournal practice, even if the content itself has somethibg like watermarks that would make a strong case for any viewer having red-flag knowledge, and the site has moderators that are unmistakably agents, it'll be much harder to make the case that the site had red-flag knowledge.
LiveJournal really seems to be in different position here than many sites with moderators.
The central issue here is that one of the moderators was an employee of the defendant. Notably, the moderators prescreen and curate material for publication. Furthermore, there was conflicting testimony about the role of the employee-moderator and his relationship with the other moderators, such as the extent to which he directs and guides them.
The only way the judge managed to discard the case at the summary judgment stage was by holding that the common law of agency didn't apply, not even with regard to the employee-moderator alone. (Which is to say, by holding that the factual dispute about the employee-moderator's knowledge, role, and relationship was irrelevant.) I haven't read his opinion so I have no idea how he arrived at that holding, but on its face it seems unsupportable.