United States v. Eric Robert Johnson

United States Court of Appeals
For the First Circuit
No. 24-1467
UNITED STATES,
Appellee,
v.
ERIC ROBERT JOHNSON,
Defendant, Appellant.
APPEAL FROM THE UNITED STATES DISTRICT COURT
FOR THE DISTRICT OF MASSACHUSETTS
[Hon. Denise J. Casper, U.S. District Judge]
Before
Gelpí, Thompson, and Montecalvo,
Circuit Judges.
Christine DeMaso, Assistant Federal Public Defender, for
appellant.
Randall E. Kromm, Assistant United States Attorney, with whom
Leah B. Foley, United States Attorney, was on brief, for appellee.
July 7, 2026

-- 1 of 29 --

- 2 -
THOMPSON, Circuit Judge.
Overview
In the current era of technology, there are myriad ways
for people to share content and information with each other. One
such way is to participate in peer-to-peer, file-sharing networks,
like LimeWire, BitTorrent, or, pertinent to this case, Freenet.
These publicly-available, internet-based platforms are a
conglomerate of files that users can download from, upload to, and
share with each other. Freenet, in particular, advertises as a
network that preserves its users' anonymity, thereby providing
them a more privatized means to share files. And, as a result,
some choose to abuse the platform by circulating illicit content.
Well, in the present case, appellant Eric Robert
Johnson's Freenet activity led to his conviction in the District
of Massachusetts for possessing child pornography. But since the
pretrial phase, Johnson has maintained that the government's bust
violated the Fourth Amendment, and he's sought to suppress all
evidence derived from the bust. Unable to sway the district court
in his favor, however, Johnson decided to plead guilty and rehash
the suppression issue before us.
So, we lay out everything in detail below, including our
reasons for why the district court is affirmed.

-- 2 of 29 --

- 3 -
I. Background
We pull our facts from the district court's decision and
from the evidentiary hearing, "presenting them in the light most
compatible with [the district court's] ruling." United States v.
McGregor, 650 F.3d 813, 816 (1st Cir. 2011) (citation modified).
a. Freenet
Before getting into what happened in Johnson's case, we
need to lay some groundwork about Freenet and its mechanics to
understand today's issues. Freenet is an internet-based,
peer-to-peer, file-sharing network that focuses on preserving its
users' anonymity. The Freenet software is publicly-available and
free, and thus it can be downloaded and installed by anyone.
Freenet's file sharing operates through a network of linked
computers (or "nodes") that break down data into fragments (or
"blocks"), encrypts those fragments, and randomly distributes them
among its users' nodes for storage.1 This breakdown process occurs
when a user uploads a file onto Freenet. After Freenet distributes
the data blocks to a user's nodes for storage, it then generates
an index (or "manifest") that lists all the blocks of the file and
1 Freenet does not store complete files in one place, and,
once broken down into blocks, users cannot decrypt the individual
blocks to view the content stored on their devices. At most, the
users are able to see there is an encrypted block in their data
cache.

-- 3 of 29 --

- 4 -
a "manifest key"2 that can be used to locate the blocks so it can
"reconstruct" the file for download. To reconstruct and download
files on Freenet, users share manifest keys through various means,
such as email, text, or posting them on online forums.
How does the reconstruction process work? To download
a file, a user inputs a manifest key into Freenet and doing so
signals the user's node to initiate the hard work. The node first
sends out requests to the user's peers -- neighboring nodes -- for
the data blocks.3 As our usage of "requests" (plural) suggests,
the Freenet software doesn't contact one sole peer to retrieve the
data blocks. The requests are divvied up (in roughly equal parts)
and sent out to several of the user's peers. If the
receiving-peer's node does not have the respective data block, it
will "relay" the request to a group of its peers to help it gather
its portion of the requested data blocks (again, with the work
divided evenly). Each time a request is sent out is called a
"hop," and Freenet will cap a request at no more than 18 hops. We
say "no more than" because Freenet attempts to conceal the original
requestor's identity by randomizing the maximum number of hops for
each download, however, that number never exceeds 18.
2 Each manifest key is made up of a series of letters, numbers,
and special characters.
3 Data blocks are identified by their "hash value[s]" which
is, in essence, a unique digital signature for each block.

-- 4 of 29 --

- 5 -
Freenet is operable in either "Darknet" or "Opennet"
mode, based on each user's choice. In the more private and
anonymized Darknet mode, a user's computer will connect only to
peers selected by that user. But in the less-protected Opennet
mode, a user's computer will connect to unknown peers (strangers)
on the whole of Freenet's worldwide network (where Freenet is
legal, of course). Freenet does not conceal a user's IP address
when their node connects to a neighboring node, thus strangers can
view an Opennet user's IP address whenever they connect.
Upon the download and installation of Freenet, users are
warned explicitly before choosing an operation mode that, in low
security Opennet mode, "an attacker with moderate resources may be
able to trace [the user's] activity on Freenet back to [the user]."4
It further warns (on another screen) that in low security mode,
"[i]t may be quite easy for others to discover [the user's]
identity." Freenet also reminds users throughout the file
downloading process that they are operating in low security mode
4 One more caveat is worth noting. Opennet mode can operate
in either "low security" or "normal security" mode. By choosing
low security mode, the user agrees that they "do not care about
monitoring and want [Freenet's] maximum performance." By choosing
normal security mode, the user agrees that they "live in a
relatively free country, but [they] would like to make it more
difficult for others to monitor [their] communications." Also,
when selecting normal mode, Freenet warns that it "will be
reasonably careful to protect [the user's] anonymity, at some
performance cost. [The user] should add friends running Freenet
and upgrade to [Darknet mode] when [they] are able."

-- 5 of 29 --

- 6 -
and to "[b]e careful" when opening a newly-downloaded file because
"it might cause your web browser or other software to breach your
anonymity[] [or] give away your IP address."
b. Freenet Roundup
Freenet's anonymity features make it alluring to those
who want to share and retrieve illicit materials, such as child
sexual abuse material ("CSAM"), online. To combat this problem,
Dr. Brian Levine -- a professor at the University of
Massachusetts -- created "Freenet Roundup," a law enforcement tool
that can identify and trace CSAM file-sharing on Freenet. Freenet
Roundup is a modification of Freenet that operates in Opennet mode
and is available only to law enforcement officers. When using
Freenet Roundup, an officer's node operates as would a normal peer
in the Freenet network. Thus, the officer's node receives and
relays strangers' requests for data blocks like a typical Freenet
user. The difference with Freenet Roundup is that it logs requests
it receives and filters them through a two-step process.
At the first step, the law enforcement node looks at the
"Hops to Live" counter for the request -- which shows the remaining
number of hops out of the possible 18-hop maximum -- to determine
whether it has 16 or more hops remaining.5 If at least 16 hops do
remain, the request proceeds to step two of filtering. At step
5 Requests with less than 16 hops are not of interest because
they are highly unlikely to have come from an original requestor.

-- 6 of 29 --

- 7 -
two, the flagged requests are sent to a server managed by
Pennsylvania's Internet Crimes Against Children Task Force
("ICAC") to determine whether the requests were made for a known
CSAM file. The ICAC has a library of data blocks identified as
CSAM and compares their hash values with the blocks in the flagged
requests. If there's no match, there's no further processing of
that request. If there is a match, though, the request information
is entered into a spreadsheet that employs a formula to determine
whether its sender was the original requestor or merely a relayer
of another user's request.6 The request information includes the
timestamp (date and time) of when it was made, the number of
remaining hops, the number of peers in the node's network, and the
node's IP address.7 The information is then sent to the appropriate
jurisdiction for further investigation based on the geolocation
for the respective IP address.
c. The Arrest
Eric Robert Johnson was one of the Freenet users caught
in the web of Freenet Roundup. An investigation ensued after
Federal Bureau of Investigations ("FBI") Special Agent Bryce
6 Dr. Levine stated in his testimony that his report revealed
the false positive rate for identifying these matches as requestors
or relayers has been no more than 2.3 percent.
7 The node's number of peers, IP address, and the request's
remaining number of hops are all transmitted when using the Freenet
software in the normal course.

-- 7 of 29 --

- 8 -
Montoya was tipped off that Freenet Roundup identified a user (with
an IP address that was later determined to be Johnson's) sent out
three different requests for CSAM data blocks between May 29 and
June 8, 2021.8 In February of 2022, Agent Montoya successfully
applied for a search warrant for Johnson's residence in Billerica,
Massachusetts, aiming to find evidence of his possession and
receipt of child pornography.9 FBI agents executed the search,
and it yielded positive results. They found two laptops (at least
one of which that had the Freenet software downloaded), a tablet,
and multiple external hard drives (at least one of which that had
contained CSAM).
So, naturally (as they say), he caught a case, and we
now move on to recall the juridical play-by-play.
II. Procedural History
Johnson was charged with one count of possession of child
pornography in violation of 18 U.S.C. § 2252(a)(5)(B), (b)(2).
During the pretrial phase, Johnson moved the court to suppress all
8 To trace the IP address back to its owner, law enforcement
first determined that Verizon serviced the IP address. Thereafter,
the FBI subpoenaed Verizon for the subscriber information, which
was registered under Johnson's name and Billerica address.
9 In the affidavit, Agent Montoya explained the operations of
both Freenet and Freenet Roundup, and, from the evidence the FBI
had gathered and reviewed, that he believed a user with an IP
address registered to Johnson had sent out three different requests
for CSAM files. Agent Montoya also described the content of the
sought-after files.

-- 8 of 29 --

- 9 -
the evidence derived from law enforcement's use of Freenet Roundup.
He kicked off his motion by arguing that the FBI's use of its tool
to monitor Freenet and track his activity constituted a Fourth
Amendment search and that he had an expectation of privacy in the
requests he sent out on the platform.
To strengthen his challenge, Johnson said the
government's conduct was in contradiction to Kyllo v. United
States, 533 U.S. 27 (2001), being that it was only able to obtain
the at-issue evidence by using "sophisticated technology that is
not in general public use," which made the purported search
"presumptively unreasonable without a warrant." He also compared
law enforcement's warrantless use of Freenet Roundup with its
warrantless procurement of cell phone site information which the
Supreme Court deemed unconstitutional in Carpenter v. United
States, 585 U.S. 296 (2018).
In harmony with the Fourth Amendment framework for
answering these types of questions (fret not, we'll break this
down in more detail below), the remainder of Johnson's suppression
arguments can be placed in two categories: subjective and objective
expectations of privacy. We'll start with the former, the
subjective variant.
For this prong of his argument, Johnson explained that
a Freenet user relies on the software to provide anonymity, and
the limited data revealed through transmissions are a "dint of its

-- 9 of 29 --

- 10 -
operation," id. at 315, that allows Freenet to perform its
functions. A single node, Johnson said, cannot determine "the
overall structure of the network and identify individual users."
Thus, a user's subjective expectation of privacy in their Freenet
transmissions is violated when the government deploys its Freenet
Roundup tool to track activity "on the network, assemble the data,
and analyze it using special law enforcement-only software."
As to his claim of having an objective expectation of
privacy, Johnson focused on the process law enforcement undertook
to distinguish requestor nodes from relayer nodes (explained
above). He argued the "requirements and resources necessary to
de-anonymize Freenet users" make it "out-of-reach to all but state
actors." He supported his claim with a forensic report prepared
by Michael Miglianti,10 who described the "significant research and
validation resources" the government used to create Freenet
Roundup and explained how such techniques "are vastly different
than the resources associated with a typical Freenet user."11
10 Miglianti is an expert witness with "years of experience
in the digital forensics and cyber-security fields" who Johnson
called upon to testify at trial.
11 Miglianti stated (in his report) that the Freenet "software
is freely available," "easily downloaded from the website," and
"requires minimal user time and effort." He contrasted this
near-burdenless process with Freenet Roundup for which
"significant resources have been expended" to develop.

-- 10 of 29 --

- 11 -
In its opposition to the suppression motion, the
government argued Johnson failed to establish a subjective and
objective expectation of privacy in his Freenet transmissions. It
challenged Johnson's contention that his "mere use of Freenet, as
a file sharing platform specifically aimed at anonymizing its
users, alone[,] manifests a reasonable expectation of privacy[.]"
The government said Johnson lacked a reasonable expectation of
privacy when he "voluntarily shared requests for child pornography
files with an unknown collection of strangers on Freenet," homing
in on the rule that a person does not have a reasonable expectation
of privacy in information voluntarily disclosed to third parties.
And it further emphasized that Johnson chose to use Freenet in
Opennet mode, which warned him that his IP address "would be shared
with many virtual strangers."
Additionally, the government pointed to a breadth of
cases that held a person does "not have an expectation of privacy
with their peer-to-peer activity on the Internet." And it
ultimately sought to align Johnson's case with United States v.
Pobre, No. 8:19-CR-348-PX, 2022 WL 1136891 (D. Md. Apr. 15, 2022),
which determined the government's use of Freenet Roundup didn't
violate the Fourth Amendment because a Freenet user operating in
Opennet mode lacks an expectation of privacy in their activity on
the software.

-- 11 of 29 --

- 12 -
All things considered, the district court ultimately
agreed with the government and denied Johnson's motion to suppress,
reasoning that he lacked a reasonable expectation of privacy in
his Freenet activity. Despite Johnson's choice "to participate in
a peer-to-peer network designed to allow censorship-resistant
communication and with measures to protect anonymity," the court
said, Freenet "warned any user that [their] identification could
be discovered and that the choice of Opennet will allow connections
with strangers (which could or could not be law enforcement)."
The court distinguished "the wholesale cataloguing and disclosure
of cell site location information in Carpenter" from Freenet
Roundup, "which essentially logs the block requests that are likely
to contain suspected CSAM." Though law enforcement employed a
modified version of Freenet, as the court explained, it operated
"in a forum where Johnson already ha[d] no reasonable expectation
of privacy." Accordingly, Freenet Roundup allowed agents to "act[]
like an undercover officer crashing a public meeting." Pobre,
2022 WL 1136891, at *6. Lastly, the court acknowledged that it
was persuaded by "cases rejecting similar challenges to
traditional peer-to-peer networks[,]" and it found Pobre to be on
point with Johnson's case.
Following the district court's order and some additional
trial preparation, Johnson opted for a conditional guilty plea
that preserved his right to appeal the court's suppression ruling.

-- 12 of 29 --

- 13 -
Thereafter, the court sentenced him to 120 months' imprisonment
and five years' supervised release. He timely appealed, and here
we are.
III. Standard of Review
We embark on a bifurcated review for appeals of a
district court's denial of a suppression motion. See United States
v. Tiru-Plaza, 766 F.3d 111, 114 (1st Cir. 2014). Factual and
credibility determinations are assessed for clear error, and for
such determinations, "we grant significant deference to the
district court, overturning its finding only if, after a full
review of the record, we possess 'a definite and firm conviction'
that a mistake was made." Id. at 114-15 (quoting McGregor, 650
F.3d at 820). The district court's legal conclusions, however,
garner de novo review. United States v. Camacho, 661 F.3d 718,
724 (1st Cir. 2011). The "ultimate decision to grant or deny [a]
motion to suppress" qualifies as a legal conclusion subject to de
novo review. United States v. Centeno-González, 989 F.3d 36, 44
(1st Cir. 2021) (citation modified).
IV. Discussion
Johnson's claims of error closely track his arguments
below. But before diving in, we'll tease out the relevant legal
rules of the Fourth Amendment.

-- 13 of 29 --

- 14 -
a. The Fourth Amendment's Shield
Our exposé begins with the bedrock principle that the
Fourth Amendment protects "[t]he right of the people to be secure
in their persons, houses, papers, and effects, against
unreasonable searches and seizures." U.S. Const. amend. IV. At
bottom, this Amendment is intended "to safeguard the privacy and
security of individuals against arbitrary invasions by
governmental officials." Carpenter, 585 U.S. at 303 (quoting
Camara v. Mun. Ct. of City & Cnty. of S.F., 387 U.S. 523, 528
(1967)). A government actor searching one's home is "presumptively
unreasonable" in the absence of a warrant supported by probable
cause. See United States v. Karo, 468 U.S. 705, 714-15 (1984).
A Fourth Amendment search occurs when the government
violates a person's reasonable expectation of privacy. Kyllo, 533
U.S. at 33. The Amendment's protection therefore hinges on whether
the person trying to invoke it can demonstrate they had a
reasonable expectation of privacy in the area searched or item
seized by a government actor. See Carpenter, 585 U.S. at 304. As
we alluded to above, the "reasonable-expectation-of-privacy"
inquiry, commonly referred to as the Katz inquiry, see Katz v.
United States, 389 U.S. 347 (1967), has both a subjective and
objective component. Harper v. Werfel, 118 F.4th 100, 107 (1st
Cir. 2024) (quoting Smith v. Maryland, 442 U.S. 735, 740 (1979)
(citation modified)). A subjective expectation of privacy exists

-- 14 of 29 --

- 15 -
where a person, "by [their] conduct," demonstrates that "[they]
seek[] to preserve something as private." Smith, 442 U.S. at 740.
The objective counterpart of the test is satisfied if the person's
"subjective expectation of privacy is one that society is prepared
to recognize as reasonable." Id. (citation modified). If they
fail to meet either component of the Katz inquiry, the
warrantless-search challenge becomes futile, and our review will
come to a halt.12 See Centeno-González, 989 F.3d at 47 ("There can
only be a Fourth Amendment violation where the complainant had an
expectation of privacy in the item that was searched.").
At this point, the reader might be wondering how our
standard of review impacts the Katz inquiry. Wonder not, though,
for we shall explain. The existence of a person's subjective
expectation of privacy "is a question of fact subject to review
for clear error." United States v. Guzman, 149 F.4th 1132, 1140-41
(10th Cir. 2025) (quoting United States v. Wells, 739 F.3d 511,
522 (10th Cir. 2014)). The question of a person's objective
expectation of privacy lies in the realm of law and is therefore
12 The question of whether a defendant can successfully
demonstrate a reasonable expectation of privacy in the area
searched or item seized has been framed as a question of standing.
See United States v. Rheault, 561 F.3d 55, 58 n.8 (1st Cir. 2009).
Put plainly, a defendant who fails to sufficiently prove their
expectation of privacy will lack "standing to claim that an illegal
search or seizure occurred." United States v. Vilches-Navarrete,
523 F.3d 1, 13 (1st Cir. 2008) (quoting United States v. Mancini,
8 F.3d 104, 107 (1st Cir. 1993)).

-- 15 of 29 --

- 16 -
subject to de novo review. See id.; United States v. Rheault, 561
F.3d 55, 59 n.9 (1st Cir. 2009). Because this threshold "standing"
question is make-or-break for Johnson's appeal, we'll start (and,
as you'll see below, end) here. Let's get into it.
b. Johnson's Reasonable Expectation of Privacy
in His Freenet Activity
Johnson puts forth a number of arguments to justify his
claim of subjective privacy in his Freenet activity. But the
government's response doesn't necessarily oppose Johnson on this
point; it states tersely that the question is, "at best,
debatable." The government urges us (throughout its brief) not to
dwell on the subjective piece but to instead focus on whether
Johnson has proven his claim of objective privacy. Because this
is the most expeditious route to resolving Johnson's Fourth
Amendment claims, we oblige and assume without deciding that
Johnson has satisfied the subjective inquiry of the Katz test. So
the pièce de résistance of our discussion will be determining
whether Johnson adequately justifies his claim of objective
privacy in his Freenet activity.
Of first import is this: It is well established that
"[w]hat a person knowingly exposes to the public, even in his own
home or office, is not a subject of Fourth Amendment protection."
United States v. Morel, 922 F.3d 1, 10 (1st Cir. 2019) (quoting
Katz, 389 U.S. at 351). But what a person "seeks to preserve as

-- 16 of 29 --

- 17 -
private, even in an area accessible to the public, may be
constitutionally protected." Id. To make our determination, we
look at factors such as "ownership, possession and/or control;
historical use of the property searched or the thing seized;
ability to regulate access; the totality of the surrounding
circumstances; . . . and the objective reasonableness of such an
expectancy under the facts of a given case." Id. (quoting United
States v. Stokes, 829 F.3d 47, 53 (1st Cir. 2016)).
To reiterate, Freenet is a peer-to-peer, file-sharing
network that is available at no cost to anyone, anywhere (where
legally permitted) with internet access. This court and all of
our sister circuits that have considered this question hold that
a defendant lacks a reasonable expectation of privacy in content
uploaded to publicly-available, peer-to-peer networks. See Morel,
922 F.3d at 10; United States v. Shipton, 5 F.4th 933, 936 (8th
Cir. 2021) (stating that "a defendant has no objectively reasonable
expectation of privacy in files [they] share[] over a peer-to-peer
network, including those shared anonymously with law enforcement
officers"); United States v. Ewing, 140 F.4th 1339, 1348 (11th
Cir. 2025); United States v. Weast, 811 F.3d 743, 748 (5th Cir.
2016); United States v. Conner, 521 F. App'x 493, 497 (6th Cir.
2013); United States v. Ganoe, 538 F.3d 1117, 1127 (9th Cir. 2008);
United States v. Perrine, 518 F.3d 1196, 1205 (10th Cir. 2008).

-- 17 of 29 --

- 18 -
But Johnson sees things differently. Despite Freenet
being a publicly-accessible, file-sharing network, still, he feels
his claim of an objective expectation of privacy passes muster.
His position goes like this: Because the world is becoming ever so
digital, his "online movements," much like his physical movements,
are likely to reveal his "familial, political, professional,
religious, and sexual associations." Carpenter, 585 U.S. at 311.
Therefore, he asserts, the Fourth Amendment should prohibit law
enforcement from warrantlessly surveilling his online activity.
Unfortunately for Johnson, we must stop him in his tracks. With
this argument, he is misconstruing the scope of the government's
conduct at issue. And as he himself acknowledges, when performing
the Katz inquiry, we must pinpoint "the nature of the state
activity [being] challenged." Smith, 442 U.S. at 741.
The government contends it did not surveil Johnson's
online activity, and, at most, Freenet Roundup captures a user's
activity only at the moment the requestor's solicitation is
transmitted to the government's node. We agree. The government
didn't use Freenet Roundup to surveil the entirety of Johnson's
online movements -- in fact, it didn't necessarily surveil any of
Johnson's online activity. Through Freenet Roundup, the
government functioned like an ordinary Opennet-Freenet user -- by
using its node to receive others' voluntary requests and retrieve

-- 18 of 29 --

- 19 -
data blocks -- and by dint of luck (or unluck), some ended up being
Johnson's.
True, Freenet Roundup did deviate from the ordinary
Freenet software in two material ways. Upon receiving a user's
file request, it used the Hops to Live counter to determine whether
at least 16 hops out of the 18-Hop-maximum remained and, for those
that did, it logged the information transmitted along with the
request.13 The government took additional steps outside the
Freenet-verse to ascertain whether the request was for potentially
illicit content -- and those steps were: (1) comparing the data
blocks' hash values with those in the ICAC's CSAM database,
(2) applying a formula to determine whether the request was from
the originator, and (3) figuring out the subscriber information
for the IP address -- but even taking a conglomerate view of these
steps (as Johnson urges us to do thematically throughout his
briefs), the government's conduct is far short of surveilling
Johnson's online movements. The post-Freenet steps taken by the
government are based solely on the timestamped information that
Johnson transmitted voluntarily when he initiated his requests,
and nothing indicates that the government went beyond that
13 While testifying at the suppression hearing, Dr. Levine
agreed that "if [one] were to break up what the law enforcement
node does into two components, one being operation and one being
logging, the logging is what makes the law enforcement node
unique[,] but the operation is the same as any other user."

-- 19 of 29 --

- 20 -
information to access any other local files on his private devices.
See Ewing, 140 F.4th at 1347. The scope of the government's
challenged conduct is therefore limited to its participation on
Freenet as an Opennet user (through its Roundup variant) and its
use of the information that accompanied Johnson's voluntary
requests.
Adamant in his position that his case is on point with
Carpenter, Johnson persists. He says (in essence) that much like
a cell phone, "almost all of daily life involves the internet" so
much so that there's no easy way to avoid using it. And because
the internet is made of "a global network of linked computing
devices," there's no real means to completely privatize one's use
of it (again, his view). Johnson supports his claims by citing
portions of Dr. Levine's hearing testimony wherein he explained
the difficulty of achieving anonymity on the internet. And where
he stated, "[b]y virtue of using the internet, if you expect a
reply, which you always do on the internet [i.e., as Johnson
expected a reply to his CSAM solicitations], just like a letter in
the post office, you have to include your IP address in order to
get a response." Given the reality and necessity of operating in
such a realm, Johnson contends he should maintain a reasonable
expectation of privacy in his internet activity because the
mechanics of its operation make it impossible for him to avoid
some type of identifiable disclosure.

-- 20 of 29 --

- 21 -
We beg to differ for this reason. Johnson argues too
broadly and fails to direct his challenge towards his voluntary
Freenet-specific conduct, as opposed to what he might do or might
have done elsewhere on the internet. Indeed, to argue his claim,
the specific question he must address (as our precedent instructs)
is whether Freenet is so integrated with members of society's, or
even its users', everyday lives that it could provide the
government with an intimate window into their private details.
See, e.g., Harper, 118 F.4th at 109 (assessing whether Coinbase
digital currency exchanges are an indispensable part of daily
life).
In Carpenter, the Supreme Court explained that a person
maintains "a reasonable expectation of privacy in the whole of
their physical movements," which prohibits the government from
warrantlessly collecting one's cell phone location records to
discover their whereabouts. 585 U.S. at 310-11. In making its
narrow decision, the Court reasoned that a cell phone is "almost
a 'feature of human anatomy,'" id. at 311 (quoting Riley v.
California, 573 U.S. 373, 385 (2014)), that "logs a cell-site
record by dint of its operation," id. at 315 (emphasis added).
And "[a]part from disconnecting the phone from the network," said
the Court, "there is no way to avoid leaving behind a trail of
location data." Id. But, in our view, cell phones and their

-- 21 of 29 --

- 22 -
stockpile of location data are patently different from the
technology involved in this case, and here's why.
Unlike cell phones, Freenet is not so intertwined with
its users' lives that it could reveal the "whole of their physical
movements." Id. at 310; see also Pobre, 2022 WL 1136891, at *5
(stating that Freenet "is not a ubiquitous part of everyday life
as is a cell phone"). Far from it. Instead, it is a rarefied,
file-sharing software with limited information on its users'
personal lives (insofar as a user chooses to upload personal
details onto the platform). Freenet does not "follow[] its [users]
beyond public thoroughfares and into private residences, doctor's
offices, political headquarters, and other potentially revealing
locales," Carpenter, 585 U.S. at 311; cf. id.
("[I]ndividuals . . . compulsively carry cell phones with them all
the time."), allowing it to "amass a trail of location data,"
Pobre, 2022 WL 1136891, at *5 (citation modified). Additionally,
Freenet Roundup, as described, is unable to "achieve[] near perfect
surveillance" of a person's physical movements, thereby
categorically excluding it from the "species of technology" that
demands more rigid constitutional scrutiny as did the cell phone
technology involved in Carpenter. 585 U.S. at 318. The law
enforcement tool here simply takes note of bits of information
attached to a user's request -- which is shared and visible to
users on the normal version of Freenet -- and additional

-- 22 of 29 --

- 23 -
investigatory steps (explained above) are required to unveil the
location of the user's node when the request was sent out. This
conduct is in stark contrast to warrantlessly collecting "a
detailed chronicle of a person's physical presence compiled every
day, every moment, over several years" through a person's cell
phone location records. See id. at 315.
Resisting that logic, Johnson argues that we should not
deem his expectation of privacy unreasonable solely because
researchers have created software and come up with a mathematical
formula that allows the government to bypass Freenet's anonymity
and privacy protections. He directs us (once more) to Dr. Levine's
suppression hearing testimony, wherein he stated that his interest
in developing Freenet Roundup was "because law
enforcement . . . found it difficult to conduct investigations" on
that platform. In other words, says Johnson, routine law
enforcement tools would not have allowed the disclosure of his
identity. Continuing, Johnson says in light of Freenet's
assurances that it was better at protecting anonymity than similar
products, notwithstanding Freenet's cautionary warnings about the
risks of being identified when using the platform, it was
reasonable for him to believe that no ordinary Freenet user would
be capable of or even interested in "developing a mathematical
formula, modifying Freenet, and collecting the data necessary" to
reveal the contents of his requested files or of the requestor's

-- 23 of 29 --

- 24 -
origin. And, as Johnson urges, the Fourth Amendment must evolve
to account for these ever-emerging advancements in technology.
Put differently, his claim is that Freenet Roundup amounts to law
enforcement's use of advanced technology to gather information
akin to that which would normally require physically invading a
constitutionally protected area.
Color us unpersuaded. Freenet Roundup does not amount
to exclusive, enhanced technology that provides the government
with a backdoor into a person's private life. Cf. Kyllo, 533 U.S.
at 40 (prohibiting the government from using technology "that is
not in general public use" to learn the private details of a
person's home). Freenet Roundup works only to "facilitate[] [the
government's] membership into Freenet and record[] the file
requests that the [government node] receives." Pobre, 2022 WL
1136891, at *6. Aside from logging and filtering information that
accompanies each request voluntarily sent to other Freenet users,
as we've discussed, Freenet Roundup otherwise operates like a
normal user on the platform. These actions therefore could not
fairly be viewed as conduct involving technology that "is not in
general public use." Kyllo, 533 U.S. at 34; see Ewing, 140 F.4th
at 1349 (stating law enforcement used "Torrential Downpour" to
"merely scan[] publicly available information . . . not shrink the
realm of guaranteed privacy by exposing information that was not
already broadcast to the public" (citation modified)).

-- 24 of 29 --

- 25 -
Johnson's mention of the government's purportedly
sophisticated hash value cataloguing and application of its
mathematical formula does not change our minds. Dr. Levine
testified that he used "standard techniques" and that he "didn't
invent some crazy math to" design the formula. And as described,
it uses only the information transmitted as part of the user's
request -- such as the number of immediate peers a Freenet user
has and the remaining number of Hops for each request it logs -- to
calculate the likelihood that the request came from the originator.
Nothing on the record suggests to us that this conduct is so
advanced and out of reach, especially when Freenet itself warned
that users with "moderate resources" could ascertain similar
information. Further, once Johnson voluntarily sent out his
request, he could neither reasonably expect the information to
remain private nor control how strangers would use the information,
which applies equally when the government lawfully obtains
potentially incriminating information. Compare Boroian v.
Mueller, 616 F.3d 60, 67 (1st Cir. 2010) (noting that courts have
held "the government's matching of a lawfully obtained
identification record against other records in its lawful
possession does not infringe on an individual's legitimate
expectation of privacy"), with Carpenter, 585 U.S. at 314 (stating
that, under the third-party doctrine, "an individual has a reduced
expectation of privacy in information knowingly shared with

-- 25 of 29 --

- 26 -
another"). Thus, the government's use of information Johnson
voluntarily disseminated on Freenet does not alter our conclusion.
See United States v. Jacobsen, 466 U.S. 109, 117 (1984) ("Once
frustration of the original expectation of privacy occurs, the
Fourth Amendment does not prohibit governmental use of the
now-nonprivate information.").
The government posits that by using Freenet Roundup to
receive requests Johnson voluntarily sent to strangers on Freenet,
its actions are more akin to that of an undercover officer invited
to participate in an illicit transaction. See Lewis v. United
States, 385 U.S. 206, 210-11 (1966). But Johnson pushes back on
such a comparison and says it is inapt. He suggests that, unlike
being chatty with an undercover, he "relied on the
privacy-protection" afforded by Freenet, "not on his confidence in
other Freenet users." Johnson argues that he didn't announce to
anyone on Freenet (or otherwise) that he sought to download CSAM,
and, normally, the users receiving his requests have no idea of
the contents of any particular block or file. Thus, society should
see his expectation of privacy as reasonable.
We remain unconvinced by Johnson's argument and instead,
our view tracks that of the government's. Similar to inviting an
"undercover agent to [their] home for the specific purpose of
executing a felonious sale of narcotics," id. at 210, Johnson
voluntarily connected his computer to strangers on Freenet to

-- 26 of 29 --

- 27 -
retrieve files. Even when warned throughout his time on the
platform, Johnson never opted to toggle on Freenet's
more-privatized Darknet mode.14 And the government never targeted
Johnson specifically, nor did it ever solicit his Freenet requests.
What law enforcement did do was go on Freenet as an Opennet-user
(something Johnson does not argue was improper governmental
conduct in and of itself), and that is what caused it to receive
peers' requests, including the illicit ones Johnson initiated.
See Pobre, 2022 WL 1136891, at *6 ("From this, the [c]ourt cannot
conclude that the technology did anything more than receive that
which Pobre and others willingly shared in Freenet."). Such
conduct does not violate the Fourth Amendment.
Johnson won't budge. He says Freenet's warnings don't
defeat his privacy claims because they signal only that a user's
identity could be discovered, not that they could be made out as
the original requestor of a file or that the contents of a
requested file could be revealed. He's wrong. Freenet warned
Johnson -- like all users operating in Opennet mode -- that his IP
address was vulnerable to identification, that "an attacker with
moderate resources may be able to trace [his] activity on Freenet
back to [him,]" and that "[i]t may be quite easy for others to
14 To be clear, we are not saying that a Freenet user operating
in Darknet mode would have a reasonable expectation of privacy in
their activity on the platform. We leave that question for another
day.

-- 27 of 29 --

- 28 -
discover [his] identity." He was therefore adequately warned that
he was opening his Freenet transmissions to the world, including
the world of the government. See Ganoe, 538 F.3d at 1127. And
despite any purported ambiguity in the warnings, Johnson was at
least on notice that his activity could be traced back to him.
What's more, by continuing to use Freenet in Opennet mode after
the warnings were displayed, Johnson failed to take any
"affirmative steps to protect the [files]" by switching to Darknet
mode. Morel, 922 F.3d at 10. He presents to us no case, and we
are aware of none, that suggests society would deem his expectation
of privacy reasonable because he believed only some aspects of his
identity were discoverable.
Finally, Johnson asserts that Freenet is materially
different from other peer-to-peer networks, and therefore, any
cases involving those other platforms are inapplicable. As opposed
to traditional networks in which users open their computers and
files to the public, he contends that Freenet's users can only
access encrypted data blocks that are unidentifiable without a
manifest key. Johnson says these files are not made public in a
way that would call into question a Freenet user's expectation of
privacy.
Again, we disagree. Despite his many claims that he
used Freenet to anonymize his file-sharing activity to the furthest
extent possible, (to repeat) Johnson voluntarily operated on

-- 28 of 29 --

- 29 -
Freenet in low security Opennet mode when transmitting his requests
after being explicitly warned of that mode's shortcomings. Because
of his personal choice to connect with total strangers on Freenet,
we see no meaningful difference between Freenet and other
peer-to-peer, file-sharing networks. Once he chose to connect
openly in the Freenet-verse, "his expectation of privacy ha[d]
already been frustrated." Ewing, 140 F.4th at 1348 (citation
modified). And "[o]nce frustration of the original expectation of
privacy occurs, the Fourth Amendment does not prohibit
governmental use of the now nonprivate information." Id.
(alteration in original) (quoting Jacobsen, 466 U.S. at 117).
Against all of this, Johnson has failed to prove a
reasonable expectation of privacy in his Freenet activity, and we
needn't inquire any further. See United States v. Mayendía-Blanco,
905 F.3d 26, 37 (1st Cir. 2018) (stating that "[t]he simplest way
to decide a case is often the best" (alteration in original)
(quoting Stor/Gard, Inc. v. Strathmore Ins., 717 F.3d 242, 248
(1st Cir. 2013))).
V. Conclusion
Johnson's challenge to the district court's denial of
his suppression motion is unsuccessful. The matter is thus
affirmed.

-- 29 of 29 --