Edit: Mentally tracking and integrating policy reversions[0] sounds like a good time as well.
[0] https://github.com/SimpleFinance/policies/commit/b0a644035fb...
There is a great deal of definitive evidence that arbitration clauses undermine the purposes of class proceedings, those purposes being:
1. increased access to justice (i.e. one may be automatically included in a successful claim, as opposed to seeking prospective reward through a potentially lengthy and expensive adjudication);
2. lowering of overall cost of adjudication (i.e. many issues may be resolved at once, at lower expense to the plaintiffs, defendant, and the courts); and
3. behavioural correction (i.e. raise the likelihood of bad behaviour by corporations resulting in monetary consequence).
Contracting-away class procedural rights can undermine point #1 and #3. Most people do not litigate even when they have an excellent chance of success, and behavioural correction requires a certain percentage of individuals to bring and succeed in their cases. The percentage who tend to litigate, and whom can be confidentially settled with out of court, are empirically a fraction the aggregate exposure from a class proceeding.
Which is all to say, arbitration clauses are not a problem if you do not feel like corporations should have monetary consequence for bad behaviour. This is certainly good for the people who own the corporations.
There is a problem, whether perceived or real, of "lottery ticket litigation" where plaintiff class action lawyers receive a large payout on success. Certainly one wants to incentivize capable lawyers to take on hard and speculative projects by dangling good returns, but in the case of class actions I have always thought of it more as a public interest litigation. There should be an incentive to succeed, but properly employed class proceedings can have a substantial public benefit. Lottery litigation can attract the wrong lawyers, though in my experience there is a fairly wide spectrum. On the other hand, public service class proceedings have their own set of distinct issues.
No it doesn't. Both the extensive use of the class form and punitive damages (or close substitutes) are virtually unique to the United States. Ad hoc regulation by an entrepreneurial plaintiff's bar acting without meaningful client input is not the not the only effective means to regulate businesses. On the contrary, plenty of countries arguably do a better job without the use of them at all.
I also broadly wouldn't trust the US to be particularly good at government regulation until our elections are publicly funded. There's a lot of evidence that our legislation is distorted towards the interests of the moneyed. I don't think that's a recipe for energetic, effective business regulation.
> No it doesn't.
It is necessary for class proceedings to induce behavioural modification, which is what we are talking about here. There are other means to bring about behavioural correction, but that was beyond the scope of my comments.
> Both the extensive use of the class form and punitive damages (or close substitutes) are virtually unique to the United States.
At least a dozen states have class procedures, but you may be right on punitive damages. Canada is quite close, but punitive measures are handed out sparingly.
>> Ad hoc regulation by an entrepreneurial plaintiff's bar acting without meaningful client input is not the not the only effective means to regulate businesses.
I partially agree; class proceedings are a poor substitute for regulation. However, a few points. First, in tandem with victim compensation and reduced litigation costs, some argue the behavioural modification aspect from financial consequence can act as a check-and-balance that requires no arbitrary intervention by the government to correct and prevent wrongdoing.
Second, class action are procedural in nature and the substance of the case must be founded on existing laws – often common law torts, but sometimes regulatory or administrative. The cost of enforcement via a class proceeding is relieved from the public purse, and no additional rights are conferred on victims by way of a class proceeding – only the vehicle for exercising those rights is modified.
Third, regulators are often not as aligned with victims as plaintiff lawyers in class proceedings. It is not unheard of, and in some places quite common, for regulators to be hired by the very institutions they are charged with regulating.
Moreover, one can gauge the effectiveness of the class proceeding in behavioural modification by observing the broad resistance to it, namely the widespread introduction of arbitration clauses. Perhaps not the behaviour it was intended to modify, but that itself speaks to the point.
> On the contrary, plenty of countries arguably do a better job without the use of them at all.
I tend to agree, depending on the circumstances, but it is a hotly contested point.
The only contradictory 'evidence' is talk about why it's better; that is, companies claim it's an easier and cheaper process. They fail to mention that what they mean is, it's easier and cheaper to them.
As a consumer, I want the laws to be written by my elected politician, don't you? Sure, there's a lot of unfairness to it. But I can't imagine a selfish corporation, one that keeps claiming their only loyalty is the the share holder, will do much better.
Just one example. You know VWs recent emission scandal? In the USA those people signed not only an arbitration agreement, but one that disallows class actions. You won't hear the problems that arbitration is causing because those problems won't make it to a court of law to be a problem.
Honestly!
Slavery waits on the end of this route.
There could be a similar law preventing anyone from signing away rights we deem too important to jeopardize.
Diffs won't help - the aim from version 1.0 is to obfuscate.
Also, this principle already exists and has a decent track record. Which is why my lease is 3 pages long, simple, and relatively landmine-free (I had to initial key clauses) while my iTunes contract is 50 page long, impenetrable, and landmine rich. The principle just needs to be strengthened & applied in the tech world.
- states are on the side of corporations not the people
Power begets power. Good power structures redirect that concentration and redistribute power, wealth, and means of production. The question of how to go about doing this is older than civilisation, most recent example I've come up with is from a history of Caesar Augustus, addressing the optimates and populares.
Former are oligarches, latter proles.
Guess which side's plank was "strict property rights and no debt forgivenss", and which was "structured bankruptcies, land redistribution, and affordable grain"?
It's the same strict propertarian vs. living-wage argument we're having today.
I have to imagine that the companies would at least be paying more for lawyers to shuffle contracts around like that than to amend them slightly.
Lawyers call change sets "redlines" rather than "diffs". Their use in the profession long predates modern computing. Word has a serviceable built-in diff tool called "Compare Documents", and many law firms license superior document comparison software. Prose diff is hard, and specialized law tools far surpass diff, wdiff, GitHub prose diff, &c. in areas like move detection and treatment of punctuation.
Source code revision control systems are built for, and conducive to, modes of collaboration that only occasionally resemble legal practice. In the main, lawyers trade "patches" with commentary via e-mail, which is still the practice of some important open-source projects, like Git itself. (For a nice comparison of this workflow to "GitHub flow": http://zachholman.com/posts/git-commit-history/ . See also git-send-email and git-format-patch.) A developer might be tempted to squash commits to make a "pretty" patch. In an adversarial negotiation, hiding incremental revision history is essential for maintaining confidentiality and information advantage. If you're going to squash, anonymize, remove timestamps, &c. every time, Git is overhead and potentially dangerous. Git's architecture does not allow parallel, private histories for shared commits; you can't have your secrets and share them, too.
Fortunately, there are two common situations where Git makes sense for legal docs: standard forms and public terms. GitHub is in active use in both those areas.
A few pioneers, among them Jason Boehmig, whose Ironclad is a recent YC alum, and Casey Kuhlman, now of Eris Industries, have done work on "open" form contracts in plaintext markup tracked with Git. Jason lead the effort to make the Series Seed financing documents available on GitHub: https://github.com/seriesseed/equity I've followed in Jason's footsteps with an in-development community revision of Series Seed at https://github.com/seriesnext/seriesnext and an experimental company-to-company NDA at https://github.com/obviousnda/obviousnda (More announcements in this vein to come.) Casey's "Legal Markdown" and later writing for Eris were big inspiration for my current open-source work.
Public terms, like terms of use, fall somewhere in between working on a standard form for the common good and negotiating a contentious agreement. On the one hand, what's good for the service provider may come at user expense, as with arbitration clauses or limits on liability. On the other hand, courts require that users at least have notice and a way to review changes made, and transparency goes a long way to earning user trust and avoiding PR blow-up.
When the users on the other side are devs, Git makes a lot of sense. A number of developer-tools companies use version control to track their privacy policies, and sometimes other company policies. npm, Inc.'s policies, to give one example, are here: https://github.com/npm/policies Many of these use GitHub's "prose diff" support for Markdown.
If you're interested in terms of use, privacy policies, and the like, I've written about how my open-source work on "Common Form" (https://commonform.github.io), a schema and content-addressing system for modular legal documents, will apply to those terms: http://writing.kemitchell.com/2015/08/24/TOS-Already-Read.ht...
As a good example, the aforementioned Texas agreement seems to have a clause that states that the landlord can enter the property at any time without previous notice. Preventing the landlord from giving up that right is not a good thing.
Even if someone is willing and able to read a 44 page contract the first time, the next time version 1.0.1 of something comes out you are generally given a 44.5 page contract to re-agree to. "Somewhere" in there, they changed something; good luck figuring it out. So even if the version you read gave you permission to sue, the amendment that you clicked-through-in-agreement later may have given away that right (an amendment they probably made because somebody sued them).
We solved this type of thing years ago, the solution just hasn't made its way into legalese: use revision control. I wish that large documents were:
- Publicly revision-controlled with a system that has clients everyone can easily use (i.e. no arcane commands, and push-buttons for the most common things like "compare to last version").
- Public accountability for the entire revision history of every single line. If somebody wants to add [Horrible Clause A] two months after I first signed it, then I get to know exactly who made the change and if I want to I can take a look at other changes they've made to other documents.