Real problems worth solving

In 2024, 80,162 qualified applicants were turned away from nursing programs across the US — not because they lacked qualifications, but because schools cannot hire enough faculty to teach them. The root cause is a salary inversion: nursing professors earn a median of $80,780 while bedside staff nurses earn $90,435+, and advanced practice nurses earn $129,000. Experienced nurses who move into teaching take pay cuts of up to $40,000. So what? Schools cannot expand enrollment. So what? The pipeline of new nurses is artificially capped at a time when demand is surging. So what? Hospitals cannot fill vacancies, patient care suffers, and the shortage compounds year over year. Why does this persist? Because nursing faculty salaries are set by university pay scales that have no mechanism to compete with clinical market rates, and there is no federal subsidy to close the gap. The result is a self-reinforcing cycle: too few faculty means too few graduates, which means too few future faculty.

The average US zookeeper earns $15.79 per hour ($32,843/year), barely above minimum wage in many states, despite typically holding a bachelor's degree in biology or zoology and performing physically demanding, hazardous work with dangerous animals. The average tenure across AZA facilities is just 2.5 years, indicating rapid turnover. This matters because animal care quality depends directly on keepers who know their individual animals' baseline behaviors, health patterns, and social dynamics — knowledge that takes years to build and is lost every time a keeper leaves. High turnover means animals are perpetually being handled by people who do not yet understand their needs, increasing stress, missed health indicators, and husbandry errors. The structural cause is that zoos exploit the 'passion premium': because people want to work with animals, employers pay below-market wages knowing there will always be a new graduate willing to accept poverty-level compensation. Zoos generate hundreds of millions in revenue collectively but allocate disproportionately little to frontline staff compensation, investing instead in capital exhibits designed to attract visitors.

Wild orcas travel up to 100 miles per day and dive to depths of 500-1,000 feet. Captive orca tanks are typically 100-140 feet long and 25-35 feet deep, meaning an orca would need to circle its tank approximately 1,900 times to cover the distance it would swim in a single wild day. Dorsal fin collapse — extremely rare in wild populations — occurs in 60-90% of captive male orcas, likely due to the combination of reduced swimming, gravitational effects from surface resting, and chronic dehydration from a fish-only diet. This matters because orcas are cognitively complex apex predators with culture, language, and multi-generational family bonds; confining them to a concrete pool is analogous to keeping a human in a bathroom for life. Captive orcas have a median lifespan roughly half that of wild populations. This persists because marine parks that hold orcas (primarily SeaWorld's remaining facilities) represent billions in sunk infrastructure costs, and while SeaWorld ended breeding in 2016, approximately 50 orcas remain in captivity globally with no viable release protocol, creating a decades-long ethical limbo.

At effectively every major zoo, psychotropic medications — SSRIs, benzodiazepines, antipsychotics — have been administered to animals exhibiting signs of psychological distress. Toledo Zoo dosed a gorilla with Prozac for anxiety, gave a tiger Valium for agitation, and put zebras and wildebeests on Haldol to 'relax' in new environments. Scarborough Sanctuary gave penguins antidepressants after storms. According to animal behaviorist Laurel Braitman, 'At every zoo where I spoke to someone, a psychopharmaceutical had been tried.' This matters because these drugs mask symptoms of a fundamentally inadequate environment rather than addressing root causes. There is virtually no peer-reviewed data on appropriate dosing, efficacy, or long-term side effects of psychotropic drugs in exotic species — veterinarians are extrapolating from human and domestic animal pharmacology. The animals are being medicated to make their distress invisible to visitors, not to improve their welfare. This persists because redesigning enclosures costs millions, while a daily pill costs pennies, and because there is no regulatory requirement to disclose psychotropic drug use to the public or to oversight bodies.

A 2009 assessment of more than 500 AZA Species Survival Plans (SSPs) concluded that few programs were estimated to be genetically and demographically sustainable over a 100-year timeframe. Population geneticists recommend an effective population of 250-500 individuals as the minimum to avoid steady genetic drift and inbreeding, but most zoo populations fall far below this threshold. This matters because SSPs are the primary justification zoos offer for keeping endangered species in captivity — the promise that they are maintaining a viable 'insurance population' against extinction. If these populations are not genetically sustainable, then the conservation argument collapses: zoos are maintaining shrinking gene pools that will eventually become too inbred to serve as reintroduction stock. The problem persists because zoo exhibit space is finite, inter-zoo animal transfers are logistically expensive and stressful, and some institutions resist SSP breeding recommendations that conflict with their own exhibit plans or visitor appeal. Additionally, there are now 295 SSPs competing for limited space, meaning each individual program gets fewer slots than it needs.

European zoos euthanize an estimated 3,000 to 5,000 healthy animals per year as 'surplus' — animals that are no longer needed for breeding programs and for whom no space exists at other facilities. These are not sick or suffering animals; they are healthy individuals killed because the zoo system overproduces animals it cannot house. The most visible case was Marius, a healthy 18-month-old giraffe publicly dissected at Copenhagen Zoo in 2014, but the practice is routine and largely hidden. This matters because zoos justify captive breeding as conservation, yet the system structurally produces animals it then kills, undermining both the ethical and conservation rationale. Some surplus animals are quietly transferred to unaccredited facilities, dealers, or even hunting ranches — Missouri's Dickerson Park Zoo transferred a giraffe, a greater kudu, and five red kangaroos to a dealer known to sell to hunting ranches. The problem persists because zoos need baby animals to drive visitor attendance and revenue, breeding programs produce more offspring than the captive population can absorb, and contraception is underused because it can complicate future breeding recommendations.

An estimated 5,000-7,000 tigers live in captivity in the United States — more than the approximately 4,500 remaining in the wild globally — yet only about 6% reside in AZA-accredited facilities. The remaining 94% live in private backyards, roadside attractions, and breeding operations where there is no standardized veterinary care, no genetic management, and no contribution to conservation. These tigers are often inbred, cross-bred between subspecies, and have zero conservation value for wild population recovery. The Big Cat Public Safety Act of 2022 banned new private ownership but grandfathered existing owners, meaning thousands of tigers remain in unregulated private hands with no sunset provision forcing transition to accredited sanctuaries. The people affected include neighbors with a 500-pound predator next door, emergency responders who face escaped big cats, and the animals themselves who receive amateur veterinary care. This persists because enforcement of the new federal law requires resources that USDA/APHIS does not have, and because state-level exotic animal laws remain a patchwork where some states still have no permit requirements at all.

In North American zoos, 67.4% of elephants (145 out of 215 sampled) had at least one foot abnormality, with nail pathologies in 92.4% of affected animals. Foot disease is considered the single most important health problem of captive elephants, and chronic, unresponsive foot conditions are a leading reason elephants are euthanized in captivity. In the wild, elephants walk 10-20 miles daily over varied natural terrain that naturally wears and conditions their foot pads. In zoos, they stand for hours on concrete, compacted soil, or other hard artificial surfaces that cause abnormal nail growth, abscesses, cracks, and infections. Arthritis was reported in 36% of surveyed zoo elephant populations. This matters because these are not cosmetic issues — they cause chronic pain that fundamentally degrades the animal's quality of life for years or decades, and ultimately shorten lifespan. The problem persists because retrofitting enclosures with appropriate substrates is expensive (often $1M+), and many facilities built their elephant exhibits decades ago with concrete infrastructure that cannot be easily replaced.

Eighty-five percent of elephants in North American zoos exhibit stereotypic behaviors — repetitive, functionless movements like head-bobbing, swaying, and pacing — that are widely recognized as indicators of chronic psychological distress. These behaviors consume 15.5% of daytime observations and 24.8% of nighttime observations, making stereotypy the second most common behavior after feeding. Asian elephants are 4 to 8 times more likely to exhibit stereotypies than African elephants. This matters because stereotypic behavior is not a quirk; it is the captive equivalent of a human institutionalized in solitary confinement developing compulsive rituals. It signals that the animal's fundamental behavioral needs — foraging across miles of terrain, complex social bonding, decision-making — are unmet. The structural cause is that even the best zoo enclosures are orders of magnitude smaller than wild elephant home ranges (which span 15-50 square miles), and zoo social groups are artificially composed of unrelated individuals who would never associate in the wild. Inter-zoo transfers, which disrupt social bonds, further increase stereotypy risk.

Between 2014 and 2018, USDA APHIS inspections of animal exhibitors fell from 9,489 to 8,354, while cited violations plummeted from 6,052 to just 1,716 — a 72% decline. By March 2025, Animal Care employed only 115 inspectors to cover thousands of licensed facilities. This means that violations are not decreasing because conditions improved; they are decreasing because there are fewer inspectors writing fewer citations. Animals in substandard facilities go uninspected for years, and when inspectors do visit, the reduced enforcement culture means marginal violations get overlooked. The people harmed are the animals themselves, who endure untreated injuries, inadequate space, and poor nutrition with no external accountability. This persists because APHIS staffing is subject to federal budget politics and hiring freezes, and because the agency's Risk-Based Inspection System was found in a 2025 audit to be non-functional — 95% of reviewed facilities were not inspected according to the system's own schedule.

Of the roughly 2,800 USDA-licensed animal exhibitors in the United States, fewer than 10% (about 238 facilities) hold AZA accreditation. The remaining 90%+ operate under bare-minimum USDA Animal Welfare Act standards, which set floor-level requirements for cage size, sanitation, and veterinary care but do not mandate enrichment programs, breeding management, or naturalistic enclosures. This matters because the animals in these unaccredited facilities — numbering in the hundreds of thousands — receive care governed only by regulations designed to prevent the worst abuse, not to promote genuine welfare. The structural reason this persists is that AZA accreditation is entirely voluntary, costs tens of thousands of dollars, and requires multi-year commitment to standards that many small operators cannot or will not meet. Meanwhile, USDA licensing is trivially easy to obtain and rarely revoked, creating a two-tier system where the vast majority of captive animals live under the lower tier with no public awareness of the distinction.

Every state prohibits employers from firing or penalizing employees for jury service, yet retaliation remains widespread because it is nearly impossible to prove. Employers rarely fire someone explicitly 'for jury duty' — instead, they reduce hours after the employee returns, reassign them to less desirable shifts, pass them over for promotions, or eliminate their position in a 'restructuring' that coincidentally follows their two-week absence. A restaurant server who returns from a 10-day trial to find their prime Friday and Saturday shifts reassigned to a coworker has effectively been punished, but proving the causal link requires an employment attorney most hourly workers cannot afford. The legal remedies that exist — compensatory damages, reinstatement, and in Texas up to 5 years of compensation — are only accessible through civil litigation that costs $10,000-$50,000 to pursue. For a worker earning $30,000/year who lost $2,000 in shifts, the math does not support filing suit. The practical result is that low-wage workers learn to avoid jury duty at all costs, further skewing jury pools toward people whose employers voluntarily provide paid leave. This persists because the laws protect against overt termination but not against the subtle forms of retaliation that actually occur, and because enforcement is complaint-driven with no proactive agency monitoring employer behavior.

Nearly 58% of judges surveyed by the National Center for State Courts reported noticing an increase in people ignoring jury summons entirely. Nationally, non-response and failure-to-appear rates rose from 14% in 2019 to 16% in 2022, with some urban courts exceeding 33%. In Harris County, TX, the rate hit 48%. Despite the fact that ignoring a jury summons is technically punishable by fines ($250-$1,500) or contempt of court, enforcement is virtually nonexistent. Courts lack the administrative resources to track down and prosecute tens of thousands of non-respondents, and judges are reluctant to jail people for avoiding a civic duty that pays $15-$50/day. The result is a self-reinforcing cycle: people learn through word of mouth that nothing happens if you ignore a summons, so more people ignore them, which shrinks the jury pool further, which increases the burden on the shrinking number of people who do comply. Those who comply are disproportionately older, whiter, wealthier, and more conservative — making jury composition increasingly unrepresentative. This persists because enforcement would require courts to spend more on tracking non-respondents than they save on juror fees, and because penalizing non-response with fines would disproportionately harm the low-income populations already least able to serve.

Jurors in complex financial fraud, patent, and antitrust cases are expected to evaluate expert testimony on derivatives pricing, DNA evidence, patent claims, or market concentration — yet they are strictly prohibited from conducting any independent research, including basic internet searches to understand terminology. A juror who does not understand what a 'credit default swap' is during a securities fraud trial cannot Google it, cannot ask Siri, and cannot consult any reference material outside the courtroom. They must rely entirely on whatever explanation the attorneys choose to provide, which is adversarial by design and optimized for persuasion rather than comprehension. The Jubilee Line corruption trial in the UK lasted 21 months and cost taxpayers over 60 million pounds before collapsing, in part because the jury could not follow the evidence. Juror confusion in complex cases does not just risk wrong verdicts — it creates pressure to decide based on the likeability of witnesses and attorneys rather than the merits of evidence. This persists because the prohibition on outside research is rooted in the adversarial system's assumption that all relevant information will be presented in court, and because allowing juror research would create appealable issues about what information influenced the verdict.

Courts send jury summons exclusively via USPS first-class mail to addresses drawn from voter registration rolls and DMV records that are often years out of date. The average American moves every 5-7 years, and younger Americans move far more frequently — yet there is no mechanism to verify addresses before mailing. In Harris County, Texas, the overall jury summons non-response rate reached 48% in 2022, up from 37% in 2019, and a significant portion of these non-responses are simply undelivered mail. When a summons goes to an old address, the intended juror never learns they were summoned. If the mail is not returned to sender (e.g., a new tenant discards it), the court may record the juror as a non-respondent and potentially issue a bench warrant or fine for failure to appear — punishing someone for mail they never received. Meanwhile, courts have no email notification, no SMS alerts, no digital verification system. The person has no way to know they were summoned and no way to prove they never received the summons. This persists because court administrative systems were built around physical mail infrastructure decades ago, modernization requires legislative authorization and IT budgets that court systems chronically lack, and there is no single federal standard for jury notification methods.

A systematic review published in the Journal of Traumatic Stress found that average traumatic stress symptoms tripled from baseline to follow-up among jurors, with 44% of participants meeting PTSD-type criteria just 7 days after exposure to graphic trial evidence. Jurors in murder and sexual assault cases are forced to view crime scene photographs, autopsy images, and recorded testimony depicting extreme violence — sometimes for weeks — with zero psychological preparation, screening for prior trauma history, or post-service mental health support. Unlike police officers, forensic examiners, and prosecutors who receive vicarious trauma training and have access to employee assistance programs, jurors are civilians with no institutional support system. They are dismissed after the verdict and sent home with no follow-up. Jurors with pre-existing trauma histories are particularly vulnerable: exposure to graphic evidence can trigger reminders of personal experiences and exacerbate prior symptoms. Courts offer no pre-screening for trauma vulnerability and no post-service counseling. This persists because the legal system treats jurors as fungible civic inputs rather than human beings absorbing traumatic material, and because providing mental health services would require courts to acknowledge that the adversarial process itself inflicts psychological harm on the civilians it conscripts.

Courts prohibit children in the courtroom and provide zero childcare facilities, yet most jurisdictions deny childcare burden as grounds for hardship exemption. A single parent or stay-at-home caregiver summoned for jury duty must arrange and pay for full-time childcare — typically $250-$350/week for one child — while receiving only $15-$50/day in juror fees. For a two-week trial with two children, out-of-pocket childcare costs can exceed $1,400 while juror compensation totals $150-$500. In California, caregivers can only claim hardship if they prove that no comparable care is economically feasible and it would create 'great financial hardship,' a standard that courts interpret narrowly. Many parents report having hardship requests denied and being told that childcare is not a valid excuse. The practical effect is that primary caregivers — disproportionately women — either serve at enormous personal cost, risk contempt charges for non-appearance (fines of $250-$1,500 and potential bench warrants), or are filtered out of jury pools entirely, reducing gender diversity on juries. This persists because courthouse infrastructure was designed around the assumption that jurors have a stay-at-home spouse, and retrofitting courthouses with childcare facilities would require capital investment that no court administrator has budget authority to approve.

In California, prosecutors used peremptory challenges to remove Black jurors in nearly 75% of cases studied and Latino jurors in 28% of cases, while striking white jurors in only 0.4% of cases. In North Carolina, prosecutors used 60% of their peremptory strikes against Black jurors who constituted only 32% of the jury pool. Despite the Supreme Court's 1986 Batson v. Kentucky ruling prohibiting race-based strikes, the decision is functionally unenforceable. Prosecutors need only provide a 'race-neutral' explanation for each strike — such as the juror's demeanor, neighborhood, or employment — and judges accept these explanations at face value. North Carolina's Supreme Court has never found a single Batson violation in three decades. The consequence is that defendants of color routinely face juries that do not reflect their community, undermining the constitutional guarantee of a fair trial by an impartial jury of one's peers. This persists because the Batson framework places the burden on the defense to prove discriminatory intent (which is nearly impossible when any facially neutral reason suffices), and because trial court judges are reluctant to accuse fellow officers of the court of racism.

Jury summons non-response rates show dramatic racial disparities: in 2024 data, 44% of Black summons recipients, 30% of Hispanic recipients, and 19% of Asian recipients never responded, compared to 12% of white recipients. This is not simply about civic disengagement — it reflects compounding structural barriers. Black and Hispanic Americans are more likely to work hourly jobs without paid jury leave, more likely to have moved recently (making summons undeliverable), more likely to distrust the court system due to historical and ongoing discrimination, and more likely to face financial hardship from unpaid service. The result is that jury pools systematically underrepresent the communities most affected by the criminal justice system. A Black defendant in a majority-Black city can face an almost entirely white jury because the summons-to-seated pipeline filters out Black jurors at every stage. This persists because courts rely on outdated source lists (voter rolls, DMV records) that undercount mobile and low-income populations, and because the $15-$50/day juror fee makes service economically impossible for those living paycheck to paycheck.

The Fair Labor Standards Act (FLSA) does not require employers to pay non-exempt employees for time spent responding to a jury summons or serving on a jury. Only 8 states and DC mandate employer-paid jury duty leave. In the remaining 42 states, hourly and part-time workers — the people least able to absorb income loss — receive zero pay from their employer during service. This creates a two-tier jury system: salaried professionals at large companies often get full pay during service as a corporate benefit, while hourly workers at small businesses, restaurants, and retail get nothing. The downstream effect is that low-income jurors aggressively seek hardship exemptions, skewing jury pools toward wealthier, older, and whiter demographics. This persists because jury duty leave is treated as a state issue, employers lobby against mandatory paid leave, and there is no federal political constituency pushing for change since the burden falls disproportionately on people with the least political power.

Federal court jurors are paid $50 per day of service, rising to $60/day only after 10 days of consecutive service. The median American worker earns roughly $250/day. For a three-week trial, a juror loses approximately $3,000-$4,200 in wages while receiving only $750-$900 in juror fees. This wage gap is not an inconvenience — it is a financial crisis for hourly workers, gig workers, and small business owners who have no paid leave. A single parent earning $18/hour who serves for 15 days loses $2,160 in wages and receives $770 in juror compensation, a net loss of $1,390 they cannot absorb. The reason this gap persists is that the federal juror fee of $50 was last meaningfully adjusted decades ago and Congress has no political incentive to raise it because jurors have no organized lobby, and the cost of raising fees across 94 federal districts would require appropriations nobody wants to fight for.

Approximately 20-30% of rural America has zero cell coverage -- complete dead zones where 911 calls cannot connect. The emerging T-Mobile/Starlink direct-to-cell service (announced 2023, beta 2024-2025) only supports text messaging and basic data, not voice calls. This means a hiker who breaks a leg on a trail, a farmer who rolls a tractor, or a domestic violence victim in a remote home still cannot call 911 through satellite-to-phone connectivity. Text-to-911 is available in some counties but requires the local PSAP (Public Safety Answering Point) to support it, and many rural PSAPs do not. The real pain: a woman on Highway 89A near Sedona spotted wildfire smoke and could not reach 911 due to a dead zone stretching miles -- she had to drive 10 minutes to the nearest fire station, during which time the fire grew into a 21,000-acre blaze. Direct-to-cell satellite is marketed as eliminating dead zones, but it eliminates dead zones for texting, not for the voice calls that save lives. This persists because satellite-to-phone bandwidth is extremely constrained (each satellite beam covers millions of square miles), making real-time voice encoding at acceptable latency technically infeasible with current direct-to-cell architectures.

Congress allocated $42.45 billion through the Broadband Equity, Access, and Deployment (BEAD) program in November 2021 to connect unserved and underserved communities. As of August 2025 -- nearly four years later -- zero dollars have been distributed for actual broadband deployment projects. States are still completing final program requirements, subgrantee selection processes, and NTIA approval cycles. The first BEAD-funded projects are not expected to break ground until early 2026, with major construction running 2026-2030. In June 2025, NTIA issued a 'Restructuring Policy Notice' that further delayed states by requiring compliance with modified requirements. The real pain: rural communities that were promised connectivity years ago remain unserved while bureaucratic approval processes grind on. A farmer in rural Iowa who was told in 2022 that fiber was coming through BEAD is now looking at 2028 or later before a cable reaches their property. Meanwhile, they are locked into Starlink at $120/month because no alternative exists. This persists because BEAD's implementation requires sequential federal-state approval stages (initial proposal, challenge process, final proposal, subgrantee selection, NTIA approval) with no mechanism to fast-track obvious cases.

Starlink's built-in dish heater ('snowmelt configuration') activates automatically in cold weather to keep the dish clear, drawing an additional 50-100W on top of the dish's normal 50-75W consumption -- roughly doubling power draw during winter. Despite this, the heater cannot keep up with heavy freezing rain, ice storms, or rapid snow accumulation. Ice builds up on the dish surface faster than it melts, causing signal degradation and complete outages lasting hours. The real pain: a rural family in northern Minnesota or upstate New York during a January ice storm loses internet at exactly the moment they most need it -- when roads are impassable, schools are closed, and they need connectivity for remote work, weather updates, and emergency communication. The power consumption spike also strains circuits in older rural homes with limited electrical capacity, and off-grid users may face total system shutdown as batteries drain. Manually clearing the dish means climbing onto an icy roof in dangerous conditions. This persists because the heater is a thermal solution constrained by the dish's surface area and power budget, and severe ice accretion simply exceeds what any flat-panel heater can handle without massive power draw.

The majority of rural homes that most need satellite internet are surrounded by trees -- the very geography that makes terrestrial broadband uneconomical also blocks satellite line-of-sight. Starlink requires a clear 100-degree cone of sky, and even a few trees within the field of view cause connection drops every 5-10 minutes lasting several seconds each. The Starlink app's obstruction checker shows red dots that correspond to real outages. The real pain: a homeowner in the Pacific Northwest or Appalachian region installs Starlink on their roof, but mature 80-foot Douglas firs or oaks surrounding their property cause 30-50 brief outages per day. Each drop is only 3-10 seconds, but that is enough to kill a Zoom call, drop a VoIP conversation, or cause a file upload to fail and restart. The only solutions are cutting down trees (often $1,000-3,000 per tree, potentially dozens needed, and may violate local ordinances), installing a 40-foot Rohn tower ($2,000-5,000 installed), or running 200+ feet of cable to place the dish in a clearing far from the house. This persists because satellite internet physically requires line-of-sight, and rural homes are disproportionately located in wooded areas.

Precision agriculture depends on uploading massive datasets -- drone survey footage (2-10 GB per flight), soil sensor telemetry streams, NDVI crop health imagery, and yield maps -- to cloud processing platforms. Starlink's upload speeds cap at 8-25 Mbps in practice, and an Ookla report found only 17.4% of U.S. Starlink users achieve broadband-level performance, largely due to insufficient upload capacity. The real pain: a farmer running a 2,000-acre operation with DJI drone surveys generating 5 GB of orthomosaic data per field pass has to wait 30-50 minutes per upload at 15 Mbps, and that is during off-peak hours. During evening congestion, uploads can stall entirely. Real-time sensor streams from soil moisture probes and weather stations become unreliable. The farmer paid $120/month plus $599 hardware expecting to modernize operations but cannot run the data pipeline that precision ag platforms require. This persists because satellite spectrum allocation inherently favors download bandwidth, and Starlink's architecture dedicates far more capacity to downstream (consumer-facing) traffic than upstream.

Starlink residential users in densely subscribed cells experience severe speed degradation during peak evening hours (5-11 PM), with download speeds dropping from 200+ Mbps off-peak to under 25 Mbps during congestion. Starlink's deprioritization system means that Standard plan users (the $120/month residential tier) are automatically throttled when cells are oversold. The real pain: a rural family that switched from a 10 Mbps DSL line to Starlink expecting 200 Mbps finds that during the hours they actually use the internet most -- evening homework, streaming, video calls with family -- speeds can crater below what their old DSL provided. They cannot predict when congestion will hit or how severe it will be, making it impossible to schedule important video calls or uploads during peak hours. The only fix is upgrading to Priority data at $250+/month, which is unaffordable for most rural households. This persists because Starlink has no per-cell subscriber caps and continues selling residential plans in already-congested cells, since turning away customers hurts revenue.

The FCC's Rural Digital Opportunity Fund awarded $9.2 billion to ISPs to build broadband in underserved areas. As of 2025, $3.3 billion of those awards have defaulted -- ISPs took the money (or won the auction) and never built the infrastructure. 1.9 million of the 5.2 million eligible locations are no longer scheduled to receive service. Three companies alone (Starry, Starlink, and LTD Broadband) account for $2.5 billion of the defaults. States like California (94% default), Massachusetts (89% default), and New Jersey (100% default) are catastrophically affected. The real pain: rural residents in these areas were told broadband was coming, some turned down other options or made housing decisions based on that promise, and now they face years more of no service while the FCC claws back funds and re-auctions. The first deployment deadline (40% of locations by December 31, 2024) has passed, revealing the full scale of non-compliance. This persists because RDOF's auction mechanism rewarded the lowest bidder regardless of feasibility, and the FCC lacked rigorous vetting of bidders' technical and financial capacity.

The FCC's National Broadband Map shows 19.6 million Americans lacking 100/20 Mbps broadband, but independent audits put the real number closer to 26 million -- a 33% undercount of 6.4 million people. This happens because ISPs self-report coverage and have every incentive to overstate it: claiming a census block is 'served' when only one address in that block can actually get service. The real pain: when BEAD and RDOF funding is allocated based on these maps, communities that are genuinely unserved get classified as 'served' and receive zero federal funding for broadband buildout. A ranch 2 miles outside a small town shows as 'covered' by a fixed-wireless ISP that technically serves one house on the town's edge, so no federal dollars flow there. The challenge process exists but pits individual residents (who must file FCC Form 477 challenges, provide speed test evidence, and document denied service requests) against ISP legal teams. This persists because the FCC relies on provider self-reporting with no independent verification, and ISPs face no penalty for overclaiming coverage.

Rural homesteaders and off-grid residents who are the most desperate for satellite internet face a cruel irony: Starlink's standard dish consumes 50-75W during active use (and up to 110-150W for the 12V mobile kit), translating to 1.5-3.6 kWh per day of continuous operation. For an off-grid solar setup, this means dedicating 400-600W of solar panels plus battery storage just for internet, on top of the household's existing power budget. In winter months with 3-4 hours of peak sunlight, the math gets worse. Add the dish's snow-melt heater and consumption spikes further. The real pain: a family running a 2kW off-grid solar system in rural Montana has to choose between running their refrigerator or keeping internet on during cloudy winter days. The Starlink Mini (25-40W) helps but has lower speeds and still costs $599 for hardware. This persists because Starlink's phased-array antenna requires significant power for beam-steering and signal processing, and no competitive pressure exists to optimize power draw since most customers are grid-connected.