This needs to be shared everywhere.

DISCLAIMER: This is by no means built to discourage people from voting. Please register to vote and make sure you participate in primaries and the mid-terms.

In addition to the equipment page from the other post, I also launched the source-focused election security page to CivicResultMaps:

https://www.civicresultmaps.org/security

The current topline is at least 227 reported threats between November 5 and November 9, 2024. There are 109 counties on the map, while 66 additional threats were reported only at the state level because no county was named.

That distinction is important. The 227 figure does not mean 227 confirmed bombs or 227 unique polling places. One message can name several facilities, and different sources count threats, polling locations, precincts, election offices, and tabulation sites differently. The page keeps those measurements separate instead of combining them into one misleading number.

You can switch between incident-source coverage and 2024 presidential winner or margin layers, filter by state and documented disruption, search by county, FIPS code, or source, and inspect the records behind individual counties. It also produces state and date summaries, shareable filtered reports, CSV data, Sources JSON, an SVG map, and a printable PDF.

The source trail shows what can and cannot be established. The FBI said many of the Election Day threats appeared to originate from Russian email domains and that none had been determined credible at the time of its statement. That statement did not provide a national count or county list, and an email domain is not proof of who actually sent a message. The mapped records rely on the Brennan Center's later public-source tracker, with reviewed official state and county detail added where available.

The election overlays are there to make geographic questions easier to examine. An overlap on a map is not proof of coordination or causation, and these records are not evidence of altered votes or an incorrect election result.

If anyone has additional official county records, after-action reports, court orders, or corrections, I would be interested in adding them to the source trail.

After the recent conversations here about voting systems, ports, modems, and "air-gapped" equipment, I wanted to share the equipment side of Civic Result Maps.

One problem I kept running into is that several different kinds of evidence get mixed together. A product sheet can show that a port exists, but not that it was active. A certification record can include an optional modem, but not show that a particular county bought or used it. A county record can name a system family without establishing the firmware or exact internal configuration of every unit.

I built this page to keep those distinctions visible and make the unknowns clear.

There are six source-linked equipment dossiers right now. You can search by state, see whether records name a specific product family or only a manufacturer, click through individual components, view official reference photos, and open an optional 3D schematic where you can isolate or hide parts.

Each dossier also includes software and version observations, configuration-change records, documented network paths and controls, power and UPS information, test findings, deployment records where available, and the archived sources behind the claims.

Where an exact product can be identified, the page also includes a narrowly scoped vulnerability and advisory review. It does not attach reports from a similar but different product just because the names happen to be close.

This is not a claim that any machine was connected, compromised, or changed votes. It is a way to see what the public record actually establishes, what it does not establish, and where better state or county records are still needed.

If you enjoy digging through primary sources, I would love for people here to stress-test it. If you find an overstatement, a missing caveat, or an official document that fills a gap, please send it my way.

Equipment explorer:

https://www.civicresultmaps.org/equipment

Source code:

https://github.com/Camreyn/civicresultmaps

So, why this book? Look, I get it. People do not like picking up books and typically "revealing" elements from books are reported in media anyways. This book is an exception, but I am not saying go purchase it, I can summarize some of what it has here (I did write this initially and then had ChatGPT format it/clean it up more since I **really** want to convey this topic properly/clearly):

1. A voting machine was compromised in under two minutes

The episode that frames the book occurred after Braun helped establish DEF CON’s Voting Village. A participant located in Europe reportedly compromised an American voting machine in less than two minutes. In subsequent Voting Villages, every machine made available to researchers was successfully compromised, sometimes within minutes.

The tests were not identical to a live election: researchers generally had access to the equipment and could study it freely. But Braun’s concern was that a nation-state could purchase the same retired equipment, examine it privately for months, develop an attack and deploy it only after extensive testing. The Voting Village machines had been purchased through government auctions and sites such as eBay.

2. Basic security precautions were sometimes simply absent

The technical findings associated with Braun’s work included equipment with:

* Externally accessible USB ports

* BIOS passwords that had never been configured

* Maintenance passwords stored in plain text

* Encryption keys stored beside the encrypted data

* Secure Boot disabled

* The ability to boot an attacker-controlled operating system from a USB drive

On one electronic poll book, researchers could boot Linux from an exposed USB port and then access, modify or delete voter-registration information stored on the device. The same tablet still contained ordinary consumer applications such as Netflix, Hulu and Amazon, showing that it had barely been hardened for election use.

3. Some vulnerabilities had remained unresolved for more than a decade

The 2019 Voting Village report said researchers repeatedly found vulnerabilities that had been documented in earlier California and Ohio security studies but were still present. Every device tested that year was certified for use in at least one American jurisdiction, and researchers found ways to compromise all of them.

This creates a particularly dangerous environment: election equipment may remain in service for many years, while attackers continually improve their tools and retain unlimited time to study older models.

4. An attacker does not necessarily need to change the vote count

Braun emphasizes that attacking voter registration, electronic poll books or election websites may be easier and nearly as destructive as altering tabulator totals.

An attacker could potentially:

* Remove or modify registration records

* Change a voter’s address, party or precinct assignment

* disable check-in devices

* Force large numbers of voters to use provisional ballots

* Produce hours-long lines at selected locations

* Crash election-reporting websites on election night

Even when every affected voter is eventually allowed to vote, delays and confusion can disproportionately discourage participation. Braun’s campaign experience had already shown him that ordinary equipment failures, insufficient machines and staffing shortages could produce serious disenfranchisement without any hacker being involved.

5. Paperless voting can make meaningful verification impossible

A central warning is that a computerized system should not be trusted merely because it produces a plausible total. With paperless direct-recording electronic machines, the ballot display, stored votes and reported totals may all be controlled by the same software. A compromised machine could therefore alter the record and the mechanism intended to verify that record.

The Voting Village report concluded that voter-marked paper ballots were necessary for meaningful post-election verification. It also stressed that simply having paper is insufficient: jurisdictions must conduct rigorous post-election audits comparing physical ballots against the electronic totals. Optical scanners themselves can be compromised, but preserved paper ballots make it possible to recover the correct outcome.

6. A mundane malfunction can become indistinguishable from an attack

One of Braun’s most unsettling observations is that election security is also about public confidence. An adversary may not need to alter a single ballot. It may be enough to cause a database outage, delay results or amplify an authentic equipment failure.

Braun explained the reinforcing cycle sort of like this:

Equipment failure or minor cyberattack
↓
Long lines, delayed results or missing records
↓
Rumors and accusations appear online
↓
Bots, trolls and political actors amplify them
↓
Voters lose confidence in the eventual result

Because many cyberattacks are difficult to attribute quickly, officials may be unable to demonstrate during a rapidly developing crisis whether a failure was accidental or malicious. Braun later described this interaction among fragile infrastructure, inexpensive cyberattacks and disinformation as a “perfect storm” capable of damaging an election even when the final count is correct.

7. The people defending elections often had fewer resources than the attackers

American elections are administered across thousands of state and local jurisdictions with widely varying budgets, staffing and technical expertise. Braun argues that election technology was frequently purchased cheaply and designed to be operated by volunteers or nontechnical employees. Local administrators might have only occasional vendor support, while a foreign intelligence service could employ teams of specialists continuously.

The institutional response was another concern. Braun encountered resistance from officials and vendors who feared that openly discussing vulnerabilities would reduce voter confidence. His position was essentially the opposite: concealing weaknesses prevents independent testing and leaves adversaries as the only people studying the systems seriously.

The important limitation

The book documents capability, exposure and inadequate safeguards. It does not establish that hackers secretly changed the reported winner of a particular American election. Vulnerability is not itself proof of exploitation.

Some of the book also reflects conditions between 2016 and 2019. Braun said in 2024 that election security had become substantially better than in 2016, particularly because of the increased use of paper ballots and greater recognition of the threat. He nevertheless continued to warn that internet-connected election infrastructure creates many potential entry points that did not exist in fully analog elections.

The deepest warning of the book is therefore not simply “machines can be hacked.” It is that an election system without independently auditable evidence may be unable to prove that the correct outcome survived a hack, malfunction or coordinated attempt to generate chaos.

A summary on the topics regarding internet connectivity in elections

The election system is an ecosystem, not one machine

A simplified version of the architecture Braun discusses looks like this:

Online voter-registration system
│
▼
Central voter database ─────► Electronic poll books
│                  Wi-Fi/cellular/Bluetooth
│
▼
Election-management computer
│
USB/memory cards
│
▼
Ballot scanners and voting devices
│
modem/removable media
│
▼
Unofficial-results system ───► Election-night website

Not every arrow exists in every jurisdiction, and the ballot-counting equipment may be physically isolated. Braun’s point is that compromising an earlier or adjacent component can still affect election operations farther down the chain.

An offline voting machine can receive data from an online computer

Before an election, an election-management system, or EMS, prepares ballot definitions, candidate lists, precinct information and configuration files. Those files may then be transferred to scanners or voting machines on memory cards or USB media.

NIST specifically warns that when an EMS is exposed to the internet, malware can infect it and then travel on the memory cards configured for individual voting machines and optical scanners. NIST therefore recommends mapping every wired and wireless connection and physically disconnecting election-management networks rather than relying only on a software setting.

This is why Braun called the voting-machine “air gap” “erroneously named” during his 2019 congressional testimony. He said an attack demonstrated by researchers could cross that supposed separation and take complete control of a machine. His broader point was not that every isolated machine could simply be contacted over the internet. It was that removable media, upstream computers and maintenance processes can create bridges into an otherwise disconnected device.

Conceptually:

Internet
│
▼
Compromised election-management computer
│
▼
Infected election files or removable media
│
▼
Nominally offline scanner

That type of path resembles the general lesson of Stuxnet: an isolated system can still be reached through its supply chain, maintenance process or removable media. Braun invoked Stuxnet when discussing why physical isolation alone should not be treated as an absolute guarantee.

Connectivity makes attacks scalable

Traditional election fraud generally requires people to act locally: interfering with a ballot box, manipulating records in one office or corrupting officials in one jurisdiction. Internet-connected infrastructure potentially allows one attacker to act remotely against many locations.

Braun therefore treats connectivity as a force multiplier:

Traditional attack:
One person → one location → limited number of ballots
Cyberattack:
One attacker → shared vulnerability → many systems or jurisdictions

He also extends this concern to the technology supply chain. Voting equipment of the same model can be deployed in numerous states, with common firmware, operating systems and election software. Braun testified that malicious firmware introduced upstream could theoretically compromise an entire class of machines, rather than requiring someone to tamper separately with every device. That was presented as a possible nation-state attack scenario, not evidence that such an attack had occurred in a U.S. election.

In a 2024 interview, Braun summarized this distinction by observing that elections conducted before the internet lacked today’s remotely accessible entry points. He acknowledged that modern elections had become substantially more secure since 2016, particularly through greater use of paper ballots, while arguing that computerization creates attack possibilities that fully analog systems did not have.

Electronic poll books may be more exposed than ballot scanners

Electronic poll books are the laptops or tablets used to determine whether a person is registered, locate the correct precinct and record that the person has checked in. They are often connected because different polling places need current registration and check-in information.

In written answers to Congress, Braun described research finding vulnerabilities through:

* Wi-Fi or other wireless networks

* Bluetooth

* Cellular connections

* Cloud-hosted backup or database infrastructure

He warned that a compromised polling-place connection could potentially provide a path toward a jurisdiction’s central voter database. Depending on the system, that database might contain names, addresses, birth dates, signatures, voting histories, driver’s-license information and partial or complete identification numbers.

But the operational danger may be more immediate than stealing information. An attacker could alter or disable records so that voters appear:

* unregistered;

* registered at the wrong address;

* assigned to the wrong precinct;

* already checked in; or

* ineligible to receive a normal ballot.

The likely result would not necessarily be invisible vote manipulation. It could be confusion, provisional ballots and multi-hour lines while poll workers attempt to reconstruct who is eligible. Braun emphasized that this alone could suppress participation and create the impression that the election was “rigged,” even when the ballot-counting equipment remained untouched.

Obviously, pirating is not allowed via reddit's ToS. So, if you do wish to seek out the book, I recommend you blacklist the site "Anna's Archive" site to prevent yourself from accidentally acquiring the book through illegitimate means.

This is a tool that pulls all of the state's election data together that processes them through known, peer reviewed, scientific methods of identifying symptoms of election tampering.

I wanted to share a project I’ve been working on: Civic Result Maps.

​

The goal is pretty simple: make election result data easier to inspect, compare, and trace back to sources.

​

Current features include:

​

State-by-state 2024 presidential result maps

​

County-level results where data is loaded

​

Source links for imported data

​

Readiness/status tracking so users can see what is loaded, partial, missing, or still being collected

​

Turnout data work in progress, including registered-voter denominator tracking

​

Public API endpoints for results, turnout, sources, coverage, and review rows

​

Historical baseline support where older data is available

​

Review/indicator views for spotting things that may deserve closer human inspection

​

Export options for CSVs

​

Import-run records so people can see what data was promoted and when

​

​

A big thing I’m trying to avoid is black-box claims. If the app shows a number, I want users to be able to trace where it came from. If a state is incomplete, I want that to be visible instead of hidden.

​

Wisconsin has been the main test case recently because I’ve been working through turnout and registered-voter denominator issues there. I’ve also started collecting 2024 EAC fallback turnout data for other states, with swing states prioritized.

​

The site is here:

​

https://civicresultmaps.org

​

There is also now a support link on the site for anyone who wants to help with server costs.

​

And if something is confusing or looks wrong in the UI, please reach out!

There was a new update in Election Truth Alliance, et al. v. Al Schmidt, et al., the federal case in the Western District of Pennsylvania involving alleged issues in the 2024 election.

​

The short version is that the case was mostly dismissed, but there is still a narrow path left for future-election claims.

​

The plaintiffs alleged that electronic voting machines and tabulation systems used in Allegheny, Cambria, and Erie Counties were improperly programmed or manipulated during the 2024 general election. Their argument was that the reported vote totals did not accurately reflect the votes that were actually cast.

​

The court’s opinion describes several categories of allegations. Those included mail-in ballot total changes between Election Day reporting and the final certified totals, an allegation that Cambria County ordered more ballots than the number of registered voters, Erie County mail-ballot delivery problems, and statistical analysis claims based on methods like the Klimek methodology and Dr. Walter Mebane’s eforensics model.

​

Judge Susan Paradise Baxter denied the plaintiffs’ request to file a Second Amended Complaint and granted the defendants’ motions to dismiss the First Amended Complaint.

​

That does not mean the court held a trial and decided every factual allegation was false. It also does not mean the plaintiffs proved election manipulation.

​

The case was dismissed at the pleading stage. That means the court was mainly looking at whether the plaintiffs had pleaded enough to bring the case in federal court.

​

The major issue was standing. In plain English, the court said the plaintiffs did not plead enough specific facts showing that each named plaintiff personally suffered a concrete injury, that the injury was traceable to the named defendants, and that the court could fix it.

​

The court also had a problem with parts of the theory relying too much on unnamed third-party vendors or software actors as the main link in the chain, instead of specific conduct by the defendants themselves.

​

What was dismissed permanently

​

The court dismissed the backward-looking claims about the 2024 election results with prejudice.

​

That includes requests asking the court to disturb the certified results, order a recount, order an audit, or make a judicial determination about whether the reported 2024 vote totals were accurate.

​

“With prejudice” means those claims cannot just be brought again in this federal case.

​

What can still possibly continue

​

The future-looking claims were dismissed without prejudice.

​

That means the plaintiffs were given a limited chance to try again, but only for prospective claims about future elections.

​

The court gave them 30 days to file an amended pleading if they want to continue on that path. To do that, they would need to be much more specific. They would have to explain which named plaintiff faces a future injury, what future election they intend to vote in, whether the same or comparable equipment will be used, why the alleged problem is likely to happen again, and how each named defendant’s conduct caused the future risk.

​

Plain language version

​

The court basically said this:

​

You have not shown enough specific, personal, legally traceable harm to keep this version of the case alive in federal court. You also cannot use this case to reopen the certified 2024 results. If you want to bring a future-election case, you need to plead it with much more detail and connect the alleged harm to specific defendants.

​

My read

​

This is a major loss for the part of the case aimed at the 2024 results. The federal court shut the door on using this case to force a recount, audit, or judicial review of the certified November 2024 totals.

​

But the ruling is not the same as the court saying the election data was investigated and proven clean. It is also not the same as the court saying the plaintiffs proved their claims.

​

It is a standing and pleading ruling. The 2024-result claims are effectively done in this federal case, while a much narrower future-election claim might still be possible if the plaintiffs can plead it correctly.

Been working on this for a while now, it's a lot of work and not fully complete yet, but largely almost everything immediately relevant for 2024 is present with sources.

I am working on 2012, 2016, and 2020 too.

Use the tabs at the top to navigate. I tried to have plain language explanations for everything, typically you just need to hover over elements to see explanations. Apologies that the mobile version isn't *super* friendly in that regard.

After listening to Ashley St Clair, I realized I remembered vaguely that a ton of the satellite were submitted for de-orbit post election but I wasn't sure about the relevancy.

So I made a tool that collects the data and maps it out in a timeline player.

I'm not all the way there yet in knowing if this has any use but I'm comfortable enough publishing it for others to look at because, who knows, maybe one of you can interpret this or see something I'm not.

Edit: The timeline tab shows a visual timeline of the satellites and the documented planned de-orbit.

After listening to Ashley St Clair, I realized I remembered vaguely that a ton of the satellite were submitted for de-orbit post election but I wasn't sure about the relevancy.

So I made a tool that collects the data and maps it out in a timeline player.

I'm not all the way there yet in knowing if this has any use but I'm comfortable enough publishing it for others to look at because, who knows, maybe one of you can interpret this or see something I'm not.

The last document processed was on May 22nd. The case was not dismissed.

Basically, the defendants had filed motions trying to get the original lawsuit thrown out. But then the plaintiffs filed an updated version of the lawsuit, called a first amended complaint.

Because of that, the judge said the old motions to dismiss do not really apply anymore, since they were aimed at the old version of the complaint.

So in plain English:

They tried to dismiss Version 1, but Version 2 is now the active lawsuit, so the court is not deciding the old dismissal motions.

This does not mean ETA won the case. It does not mean the judge ruled that the claims are proven.

But it does mean the lawsuit is still alive, and the fight now moves to the updated complaint. The defendants can still try to dismiss that version too, but the first attempt to dismiss the old version is off the table.

The lawsuit is still moving forward. This was a procedural step, not the final decision.

I wanted to share my experience trying to get a clear answer from Pennsylvania officials about whether the issues being observed in the 2024 election data would actually be detectable under Pennsylvania’s post-election audit process.

The main thing I am trying to understand is if outside groups are identifying possible targeted anomalies in presidential election data, what part of Pennsylvania’s 2024 audit process would actually confirm or disprove those concerns against the paper record?

I sent a detailed email to the Pennsylvania Department of State press address after calling the Governor’s Office and being directed there. In that email, I made clear that I was not asking Pennsylvania officials to accept Election Truth Alliance’s conclusions as proven. I was asking them to explain what the audit process does and does not verify.

The distinction I keep running into is this:

Every valid vote may have a voter-verifiable paper ballot record.

That does not mean every vote was manually counted or hand-audited after the election.

That distinction matters because Pennsylvania’s public audit materials describe a county-level 2% statistical recount, or 2,000 ballots, whichever is fewer, and a statewide risk-limiting audit. My understanding is that the 2024 statewide RLA selected the State Treasurer race, not the presidential race.

So the technical question I am trying to get answered is:

If the alleged issue is specific to the presidential race, and if the statewide RLA audited the State Treasurer race, then what statewide audit specifically checked the presidential contest against the paper record?

And if the county-level audit only checks a limited sample, would it necessarily detect a targeted issue that exists outside the sampled precincts, batches, or reporting units?

That is the concern.

When I called, one response I received seemed to treat the existence of voter-verifiable paper records as though it meant every vote was recounted in the audit. But based on Pennsylvania’s own audit descriptions, that does not appear to be correct. Every vote having a paper record is not the same as every paper record being manually reviewed in the audit.

That phone exchange was frustrating, but it is not the main point.

The main point is that I am trying to get an answer to whether the specific patterns being raised in the data could be checked in a meaningful way by Pennsylvania’s actual audit process.

They have not responded to the email yet. I sort of do not expect that they will.

I wanted to share my experience trying to contact Marc Elias’s firm about election-related data I believed was time-sensitive.

On March 24, I called the firm and explained that I wanted to share information/data with them. I was given an email address and told that after I sent it, they would send a confirmation email verifying that they received it.

I sent the information.

I did not receive a confirmation.

On March 25, I followed up to check whether they had received it or sent a confirmation. Nothing.

On April 5, I checked again. Still nothing.

Later, I called again. When I called from my normal number, it went straight to voicemail. I tried from another number, and that also went to voicemail. When I called from a different number, someone answered. They told me they had never received the emails. I verified the email address I had been using, and they confirmed it was the correct one. They then asked me to forward the email again, and this time they said they received it.

After that, I followed up again because I was told someone would contact me within 48 hours. I did not receive a response within that window. I sent another follow-up after seven days.

This then turned into roughly two months of trying to get any kind of update.

Eventually (May 21st), I called again and was told they had no record of anyone responding to my inquiry. They put me on hold to check with someone. The answer I received was a very general statement along the lines of: thank you for the information, and it may be used for future needs.

I told the person on the phone that this felt extremely general for something I had explained was time-sensitive and had spent two months trying to get an answer on. I also said it felt like a non-answer.

They later followed up and said the suggestion from their side was to look at the Center for Secure and Modern Elections / modernelections.org as an example of a group that does the kind of work I was trying to pursue.

I am not posting this to claim I know exactly what happened internally. Maybe it was just disorganization, maybe the emails got missed, maybe they receive too many inquiries, or maybe they were never going to seriously engage with it. I do not know.

But I do think the timeline is worth sharing because, from my perspective, I made repeated good-faith attempts to provide information, followed the process I was given, and still had to spend about two months trying to get even a basic response.

Make of that what you will.

I’m trying to better understand what normal civic options exist for Watertown residents when they believe a school board member should be removed or replaced.

From what I’ve found, Wisconsin does not appear to have a simple “public vote at a board meeting” process to remove a school board member. The main public tool seems to be a recall election, plus the ordinary election process when terms expire.

My understanding of the Wisconsin school board recall process is:

Confirm the board member is recall-eligible.

Wisconsin recall rules generally require the elected official to have served at least one year in their current term before a recall petition can be filed.

Contact the school district clerk.

The district clerk should be able to confirm whether a specific board member is currently eligible for recall, the official filing officer, the exact number of signatures required, and what forms/process must be used.

File/register before collecting signatures.

A recall effort has to be registered before petitions are circulated. You cannot just collect random signatures first and turn them in later.

State a reason related to the office.

For Wisconsin school district officials, the recall petition has to state a reason connected to the official responsibilities of the board member.

Collect enough valid signatures within the deadline.

Wisconsin law appears to require signatures equal to 25% of the votes cast for governor in the last gubernatorial election in the relevant district. The filing officer is supposed to certify the required number.

Submit the petitions for review.

Signatures can be challenged, so accuracy matters: names, addresses, dates, circulator info, and eligibility of signers all matter.

If the petition is sufficient, a recall election is triggered.

A recall does not automatically remove the board member. It makes them stand for election again before their term would normally end.

Separately, there may be a legal “removal for cause” route for misconduct, neglect of duty, malfeasance, etc., but that seems much more formal/legal and not the same as a public recall petition.

For people who know Wisconsin local government or school board law:

Am I understanding this correctly? Is there anything important I’m missing before someone even considers organizing around this?

I’m not asking for harassment or anything extreme. I’m trying to understand the lawful civic process: recall, elections, public comment, records requests, and what residents are actually allowed to do.