The topic for debate is the US DOJ trying to seize anti-Trump protest site visitor logs, from the DisruptJ20 inaguration protest. See this Atlantic article.
Summary: The DisruptJ20 website was launched November 2016 to plan anti-Trump inauguration protests. On July 27, 2017, the US DOJ tried to seize "more than 1.3 million IP addresses, as well as the email addresses and photos of thousands of people" from the DreamHost web host's DisruptJ20 website. Dreamhost is resisting the warrant.
Warrant's legality? Constitutionality? Legitimacy in light of other concerns (like civil rights, digital privacy)? In the article, a digital privacy attorney states, "The Fourth Amendment was designed to prohibit fishing expeditions like this."
To reduce this to absurdity: if two million people click on a protest related website, and two hundred show up, and one commits a crime, then all two million names, email addresses, etc. are seizable by the big bad gummint??? This has the odor of an implicit we'll-know-where-you-live threat to discourage a/o rile J20 and any such future protestors, and to boot, wave the prospect of future legal harrassment of any web-hosts daring to sponsor pages protesting the administration's policies (if any are even remotely linked to a crime). This list would also be a marvelous list of persons to go around knocking on the doors of in the early- to mid stages of some greater authoritarian screw-turning. That Trump has authoritarian ambitions seems no great secret.
"A hearing in D.C. Superior Court is scheduled for Friday." (ie 8/18/17)
I think that’s a crucial question. The Atlantic article mentions, “In the physical world, the government can search only one apartment in an apartment building with a single warrant; it can’t search the entire apartment building.” And [IMO] this is what makes it so absurd, that, yes, they may get a warrant for a data dump for a website hosting illegal activity, but the DisruptJ20 website was hosting a legal activity (though some, later, committed illegal acts). There’s rather a big difference. This seems more like raiding the whole apartment building when only one apartment was the site of wrongdoing. DreamHost’s case, in my interpretation, seems to follow the “name the apartment, and we’ll give you the contents” lines, which seems a reasonable check on the governments reach here.
And here’s another article on the seizure from NPR, in which a Georgetown law professor notes that the Constitution was intended to get away from “general” warrants issued by the British, “which gave sweeping access to search any location with a single piece of paper.” He adds, “This smells like a general warrant. I think the framers would recognize a single request to get the reading habits of tens of thousands of people to essentially be the closest thing we have in modern times to a general warrant.”
The basic problem here is the lack of useful analogies between the physical world of searching a house and the search of a server’s electronic records.
A search warrant can authorize the search of a house for cocaine and guns. Agents can search the entire house: attic to basement, in between mattress and springs, in the kids’ rooms and the laundry baskets, to execute that warrant. But they don’t take anything except guns, cocaine, and perhaps any other object whose illegal nature is immediately apparent.
Now we discuss a search warrant for a server. The government contends that they have probable cause to believe that the server records contain evidence of crimes. The server owners object that the server also contains personal information for people with non-criminal, First Amendment protected exercise of rights. But how can the government search for the criminal records without also looking at the non-criminal records?
There are options. One is a firewall: one government team or a special master evaluates all the records, and passes to the prosecution team only the records showing criminal connection, and are forbidden from discussing other findings.
This has seldom been necessary, but it’s not unheard of. A similar solution can be used when the government needs to search a lawyer’s work product, as an example.
What crimes have been committed by the users other than viewing a legal website? For possibly reading or possibly even joining in a protest?
It shold be the other way around - they;ve arrested ‘john doe’ and they want to specifically find out if ‘john doe’ visited this site - those records are on his computer or they can narrow the search warrant to his specifics.
As it stands - this is a ‘tell me who my enemies are’ type of witch hunt.
Zeb and Yitzak conspire to rob a bank. Armed with probable cause to prove this agreement, the government procures a search warrant and seizes Zeb’s computer because they wish to find any evidence of communication with Yitzak, like an e-mail that says, “Hey Yitzak, just reminding you of our joint agreement to rob the bank tomorrow at noon. I have already taken several substantive steps toward the robbery, including buying the shotguns and drawing our escape route, which I have attached as an MS Paint .jpg.”
Can Zeb object because his e-mail is also full of other, noncriminal transactions? Can Zeb say, “How about I just give you all the mail I’ve exchanged with Yitzak?”
If the government’s search of Zeb’s mail also uncovers a note to Xanadu, thanking him for his help in composing the bank holdup note and promising him five percent of Zeb’s share of the robbery proceeds, must they ignore it?
Isn’t this analogy inapt, in that they are not targeting Zeb’s account in this case, but everyone who has used that mail service? That is, they are not saying, show me what content Zeb’s IP address viewed on your web page. They are instead saying, show me the information for everyone who looked at, uploaded, or downloaded content to your site.
Unless I have the facts wrong, I don’t think they claim that the site itself is somehow guilty of some crime, right?
I’m happy to be argued out of this position, though.
Czarcasm, the judge that issued the warrant presumably thought there is sufficient evidence of a crime, right? So, I say we should allow for the fact that some crime was committed by someone. I think the issue here is the over-broad warrant.
Yeah, I think the analogy would be more like Zeb and Yitzak had conspired to rob a bank using email through gmail.com, so the government subpoenas all of the email and login information that Gmail has, so they can sort through it to get Zeb and Yitzak’s conversations.
And to continue the analogy, Gmail has offered to give the government Zeb and Yitzak’s specific emails, but not their whole database.
They are saying that the section of the server devoted to hosting www.disruptJ20.org contains evidence related to violations of law. In other words, they’re not saying anyone is guilty; they’re saying a crime was committed and we have probable cause to believe we will find evidence of that crime at this site.
Would you be OK with a search warrant that asked Google to deliver all the content in Zeb’s and Yitzak’s Gmail accounts, and their Google drives, and any other storage they controlled or had access to? After all, what if the jpg drawing was simply stored on Google drive and then shared between them, but not mailed?
That’s not a relevant inquiry. The affidavit says that the government expects to find the fruits, evidence, or instrumentalities of the crime in the data stored by the user account that manages the website. The search warrant does not accuse any one of a crime. It says that the evidence of a crime will be probably be found on that server.
As I understand it, if the police search a home looking for a specific thing as listed on the warrant – let’s say Mr. Manafort’s business records – there are sharp limits on how any other items discovered – let’s say Trump’s pee-pee tapes – could be used.
So let’s say the authorities get all this data and are looking through it for evidence of those knuckleheads planning a riot. If they discover that Donald Trump posted his pee-pee tapes to the website, I would expect that the authorities would not be free to seize the tapes, because it doesn’t relate to the planning of the riot. Lawyer types, am I generally on the right track?
To me, that means that it seems reasonable that the government may not be wildly violating the Constitution with its search warrant. Especially if the basis of the warrant may have involved evidence like one of the arrested protesters saying, “Oh yeah, I was on this message board and I read about all these users talking about how hard they were going to riot, but I personally didn’t post in those threads, nor can I remember their user names, much less their real identities.” Seems to me like the police ought to be able to find out what was said in those online discussions, how many users posted to them, and have some means to identify the users.
That all being said, if there were ways to narrow down the scope of the warrant, I think that would be a good thing to do… but given the lack of details about the evidence being sought, I can’t conclusively say that the warrant is completely unreasonable as-is.
It appears from the search warrant itself that there is a two-step process, as mentioned above. Step one is that the company delivers all the data, lock stock and barrel. (See section I of Appendix B) Then any data delivered pursuant to section I that constitutes evidence regarding the rioting charge may be seized by the government. It doesn’t seem to say what happens to the data that is not seized, I would assume it is either returned or destroyed? Not sure.