Free software and fiduciary duty

Indeed, it's a summary judgement. However, in terms of the peril created for software authors that hardly matters from a practical perspective:

The cost of taking this to trial will be many millions of dollars.

If the defendants run out of funding along the way they will lose. Assuming they don't and they win, they still lose: In the UK the loser pays motion by motion, but usually only 70% (and even that diminished by earlier arguments that you weren't successful with). That might sound good until you realize that if you're on the hook for 10 million in trial costs and only recover 70%-- that would still leave most ordinary individual defendants completely bankrupt. All on account of contributing to some free software without remuneration.

We're extraordinarily fortunate to have sponsorship from the Bitcoin community for the time being-- though it can't be guaranteed to continue because there is no bound on the upper cost to the trial or the number of additional cases he can file (and, in fact, he's just filed a second lawsuit). This isn't merely a hypothetical, for example a journalist/podcaster that Wright sued was forced to abandon his defense mid-trial as a result of his sponsorship running out when he hit a million GBP in legal costs. But our good fortune this is a bit of historical accident, not a guarantee... and someone else in a similar situation might not be so fortunate.

Because of this, as an individual the outcome of going to trial is easily lose/lose: you're bankrupt regardless of if you win or lose. The plaintiff in this case was well away of this, bragging on social media long before filing the case that his goal was to ruin the defendants financially and destroy their families and that they would either way by the costs of the case. Because of this *it's* critical that free software authors have the greatest chance possible of being able to discharge spurious claims on a summary basis: after there isn't income from the work that would allow for absorbing legal costs as a "cost of doing business". If you can't discharge spurious legal attacks on a summary basis, -- then participating in the first place is a bad idea.

Going back to the article,

> The appeals-court ruling, though, makes no mention of warranties.

It's an error to read the appeals decision in isolation however, as it's effectively a diff of the lower court decision, where the subject is addressed https://www.bailii.org/ew/cases/EWHC/Ch/2022/667.html

It didn't arise in the appeal because that point wasn't being appealed. The original decision was made on the basis that this would just be a contrived application of fiduciary duties in the abstract. Unsurprising, it's not like you can allege that thieves stole a billion dollars from you and now the federal reserve has a duty to void the stolen bills by serial number and print you new bills. :) Positives duties are quite rare in the law, even a police officer holding a rope that could save your life as you fell off a cliff is not obligated to hand it to you.

> In other words, it is not the software license that might possibly create an obligation here.

Indeed, it doesn't create it but I would argue that it failed to disclaim it effectively enough.

Since the plaintiff claims to have used the system by way of the MIT (expat) licensed software written by the defendants (and the owner of the plaintiff actually redistributes that software) and the software formed the only basis of connection between the parties, I think there clearly was an opportunity to have more forcefully protected from this outcome: The trial court wrote, "It is true that the disclaimer is in broad terms, but it is not clear to me that it would reasonably be understood to mean that controllers of the BTC Network assume no responsibility for any aspect of its operation." but of course that *is* the intent, because who would publish their work for free to the world if it meant taking on potentially millions of dollars in liability for other peoples errors? We certainly wouldn't have.

So I agree with Corbet on the court's fundamental misunderstanding-- but that "misunderstanding" was very likely at this juncture by virtue of this being a summary determination: the court was constrained to find the facts in the plaintiff's favor. It wouldn't have been impossible for them to be wise enough to the free software to see through his control allegation, but in this case they weren't. But even in the (untrue) hypothetical world where the imagined control actually did exist, we would have expected the license we offered (and the plaintiff accepted) to form a serious shield against these claims.

Now, if that could be clarified while also staying within the norms of open source software or if doing so would be of general benefit (rather than just helping with this particularly weird case) I'm less sure of.