Francisco Tolmasky

However, a comic book has never accidentally deleted the reader’s entire comic book collection & fan fiction has never accidentally leaked credit card information. As much as it pains me to say it, much software mirrors more giving untrained legal advice than sharing a fun story.

Software is an incredibly powerful thing that is uniquely simultaneously tremendously accessible and tremendously impactful. A chair design can be dangerous, but it’s hard to post the plans and have a non-skilled carpenter build them — “trying“ the chair isn’t very accessible.

On the other hand, posting writings or art online is very accessible, but not likely to cause unintentional physical damage (avoiding spiraling into the argument of how “dangerous” an idea can be). Generally speaking, most literature online isn’t accidentally a security nightmare

The intentionality is somewhat important here too: any endeavor can potentially be deliberately dangerous, but software, unclear whether incidentally or essentially, seems to be particularly prone to unintentional catastrophe.

As such I think the software industry in general, and open source community specifically, has been wrestling with the fubdemantal problem of responsibility. What do we owe each other? Nothing? It can feel that way since the acts of creation feel as innocent as drawing a picture.

And open source seems to provide the perfect cover of plausible deniability. No money changes hands, every license has an “as is” clause. But I think there’s a more important question here than whether we should be *legally liable* for our software.

Because we seem to want to have our cake and eat it too. We want to fall back to the “no promises” position when it’s beneficial, but lean hard on the marketing benefits of open source to maximize usage and popularity. I feel these are somewhat in conflict.

It’s one thing to leave something sitting out with a clear warning, it’s another to pitch it heavily with fine print stating “caveat emptor”.