4 ms·
I've noticed typically I agree with your views, but on this case we diverge. I think it's reasonable to assume that when the service cannot be accessed before
by reginaldjcooper 13y ago
I've noticed typically I agree with your views, but on this case we diverge.
I think it's reasonable to assume that when the service cannot be accessed before the accessor clicks "agree", and the accessor clicks "agree", that constitutes a valid contractual agreement. It's no more silly than mailing in a credit card application and being bound by those terms.
The fundamental attributes are present. There is a contract and you can agree to it or not. Your interaction with the company is viewing the terms the lawyers wrote and agreeing to them or leaving the website.
- levosmetalo 13y agoDo you also agree that a person infected with malware also accepted it just because he clicked "Ok" or "Yes" on some of the dialogs? There is no signature, there is no proof of identity of the signing party (copy of personal ID, witness, ...). Usually no confirmation that other side also accepted it. No copy of ToS received after "accepting". We all know that it could be that, for example, my 4yo child clicked on it when I went to the toilet after reading first few paragraphs of ToS, and now I'm somehow bound to it? Also, it happens very frequently that you "accept" some ToS, and after two years you find on some forum that Google, or PayPal just changed their ToS without sending you the updated copy for acceptance. Sorry, but too many things that are really different between ToS and a real acceptance of a Contract.
- reginaldjcooper 13y agoIf you agreed to a ToS by clicking yes, and the ToS allows them to install malware that's a failure on your part to read it. If your child clicks it, you probably have recourse in civil court because a 4-year-old is not able to execute contracts. Most ToS's include a clause that they can change it at any time; you might not be legally bound by the new one, check that with a lawyer. It seems like you are angered that you have to enter into a contract before using software, and you would prefer to just click "accept" and have that magically not be interpreted as you agreeing to the terms. "I'll use it but I don't accept the terms". I'd like that too, but I'm afraid that's not how it works.
- levosmetalo 13y agoNo, I just say that the fact that someone clicked accept button on some website is not a proof that I accepted anything. If things worked like that you would not be required to go to the bank to sign a contract and take a loan, one mouse click on bank website would be enough. I could as well write whatever ToS on my homepage with "accept" and "not accept" link with whatever terms, and then wait for Google bot to "accept" my terms. Do you think Google would be bound by their "acceptance" in the same way. It's not like I was using my digital signature or whatever, it's just that someone clicked somewhere. The burden of proof here is on the "vendor" or software provider. They need to prove that I accepted their terms, not the other way round. That's why many free services require credit card, copy of the ID and physical signature as a proof of identity. The fact that Google decided that this might be too expensive for them to enforce, and that is cheaper to just not do it is their choice.
- reginaldjcooper 13y agoWell I think the burden of proof is going to be on you if you are found willfully using the service afterwards. Otherwise you're probably right.
- mikeash 13y agoI'm not a lawyer, so take the following with much salt. Furthermore, I'm arguing more from an "ideal legal world" viewpoint rather than the current American legal environment. A contract must consist of: 1. A "meeting of the minds." This means both sides come to a common understanding of what's in the contract. Basically, it means there's an offer by one side and acceptance by the other without changes. 2. Intention to be legally bound. Basically, this means that the context of a contract matters. It seems to boil down to commercial agreements generally being enforceable while others (e.g. promising money to family) are not. 3. Consideration. This is some sort of exchange of value. In short, a contract cannot be one-way. This is why you sometimes see e.g. things being sold for $1 rather than being just plain given away for free. The way I see it, there are serious problems for click-through agreements for both 1 and 3. 2 is not a big deal, as despite the name, it doesn't appear to actually apply to the case where you click Agree without the intention to follow the contract. One problem with #1 is that essentially nobody ever reads these things. In general, "I didn't read the contract I signed" is not an excuse, of course. However, I think this does (or should) change when everybody doesn't read it, because it means that the other party knows, or should know with reasonable certainty, that you're unaware of the contents. Normally you presume that each party is aware of the contents of the contract, but you can't do that with click-through agreements since it's widely-known that they go unread. A contract which I don't read and which you are well aware I didn't read should not be enforceable. For high-value, important contracts like buying a house, a person familiar with the contract will go over each page with you and have you initial it to affirm that you read it, precisely to avoid this problem. The combination of widespread failure to read the contracts and a complete lack of effort to ensure that you, personally read the one in question means that, IMO, there is no meeting of the minds. Another problem with #1 is that there is either no opportunity to propose changes, or the other side never notices or reviews them. With a normal contract, I can cross out clauses, add or change wording, etc., at which point the other party can accept or reject my changes, or propose further changes. Where this really runs into trouble, I think, is when you edit the contract locally before clicking Agree. This is generally trivial using a DOM inspector for any click-through agreement found on the web. Imagine I take the ToS in question here and alter it to read, "In exchange for mikeash's use of the Google API, Google promises to pay mikeash the sum of ten million dollars." Then I click Agree. Should Google be bound by this agreement? I imagine you immediately and strongly say "No!" But why should I be bound by the original while they are not bound by the alteration? In a normal contractual setting, they have the opportunity to say yes or no to my proposed changes. They will, of course, say no. In this setting, they don't say no! They simply grant me access to their service after I "Agree" to the altered contract. Does this not imply acceptance on their part? If not, what's the difference between their situation and mine? Finally, #3 seems to completely destroy the concept of any click-through agreement that isn't part of some sort of payment process. Terms of service presented as part of a checkout process would seem to be fine in this respect, but when presented as part of a free web site, it doesn't seem to work. They give you access in exchange for... nothing. Similarly, click-through software licenses seem to fail here because they show up after the money has been exchanged. I pay for a copy of software which I obtain, and only later am I presented with the EULA. Since the money and product already changed hands, there's no further consideration, and so the EULA should not have force. That's my thinking. I welcome dissenting opinions.