I know I'm a bit of a grumpy old man on legal topics, but I think an awful lot of HN readers view legal issues through the lens of a half-baked libertarian ideology [1] and raise their risk profile as a result. If a project does hit paydirt and see massive user growth in a short period, then statistically there are bound to be conflicts of interest between some of those users. Many such conflicts can and should be smoothed away by good community management, but a few will escalate into disputes whose outcome can have significant economic implications for the firm. OMO the legal system is not some sort of stick to beat people with or deliberate crafted to keep upstarts at a disadvantage, as in Marxist social theory, but the aggregated history of past disputes within a (relatively) simple procedural framework. The role I envision for a CLO is not to sit around all day worrying about lawsuits or drafting ever more aggressive contracts, but to maximize the economic benefit of a firm's customer relationships over the long term. So if you want to provide people with (for example) a new way to consume media, then part of your long-term strategy should include a vision of how you would like the copyright system to function and a plan for how to move the needle on the way it does function.
1. I say 'half-baked libertarian' because a lot of people resent government administration or even recognition of things like patents or other intellectual property while having very absolutist views on the transfer of other economic interests within markets, for which they would like ironclad contractual protection - even though contractual protections depend upon the enforcability of court judgments, which are in turn founded upon a social consensus for governance. Often when people complain about the unfairness of a legal ruling, it seems to me that they've simply failed to consider the existence of interests other than their own.
I also think you're fundamentally right about the way HN looks at legal risks. In particular, I think the average HN reader drastically discounts the cost of legal risks that apply only once a particular enterprise has traction. I think those are some of the most pernicious risks, because they can lead to people investing years of their lives in businesses for counterfeit upsides.
But having said that: do you really need on-staff legal to learn that a contract with no explicit acceptance and with unilateral change provisions is legally suspect? Over the last couple years, I've been trying to keep a little tally of all the redlines we've gotten on our contract reviews (all of our MSAs get legal review), and I feel like much smaller details have come closer to derailing projects than these two details.
The court takes this amendment power to its logical conclusion. If Zappos can change the terms at any time, then it can change the arbitration clause at any time. Thus, citing to a long list of cases, the court says that such unilateral power to change the arbitration clause makes the clause “illusory”–and thus unenforceable.
Oh so very useful to know...
