Summary

  • The strongest physical evidence for Carolina Wireless Broadband is not a marketing map. It is a City of Laurinburg elevated-tank maintenance notice that identified the company's equipment on the Caledonia Road tank and the US 401 South/Purcell Road tank, then described how maintenance could force removal, de-energisation, or interruption.
  • The public internet evidence is weaker than the physical site evidence. ARIN still records AS30501, 23.157.16.0/24, and 2602:fd56::/36 for the company, but RIPEstat showed no announced prefixes for AS30501 at the 13 July 2026 check and placed the IPv4 block's last observed origin visibility in February 2026.
  • The network should therefore be assessed as a recent, real fixed-wireless access system with unverified current reachability, unproven backup power, unproven backhaul diversity, and a customer-dependency story that depends on local repair labour as much as radio coverage.

The network becomes legible at the water tank

Small fixed-wireless networks often begin with an unglamorous piece of civic infrastructure. In Laurinburg, North Carolina, Carolina Wireless Broadband becomes legible at two elevated water tanks. The City of Laurinburg's 6 August 2025 notice about maintenance on its elevated storage tanks said company equipment was present on the Caledonia Road tank and on the US 401 South tank at Purcell Road. The second tank also hosted three cellular carriers. The city separated the occupants in an important way: cellular carriers had contracts and paid rent, while Carolina Wireless Broadband had access without rent as a public service.

That notice is the anchor for this article because it describes an operating dependency rather than an advertised service area. A fixed-wireless operator can publish a coverage claim, but a municipal maintenance notice names the support structure, the owner, the maintenance conflict, and the consequence. Laurinburg was preparing to clean and repaint several tanks. To protect workers from unsafe radio-frequency energy and to protect nearby property, communications equipment could need to be removed or de-energised.

At Caledonia Road, the city said Carolina Wireless Broadband had been told to remove equipment and had reportedly made other arrangements to continue service. At US 401 South/Purcell Road, it said the city and painting contractor had been working with the company to minimise interruptions, while warning that interruptions might be required if equipment could not be relocated.

That is a resilience test in plain public language. It does not require a hurricane, a backhoe cut, or a carrier outage. The host structure itself needs work. A system that depends on that structure must either move radios, shift customers, route traffic through another site, or accept a service break. The notice does not say how many customers were affected, whether the Caledonia alternative arrangement succeeded, whether the Purcell Road work later caused interruptions, or whether the equipment returned to the tanks. Those limits matter.

But the notice still proves that Carolina Wireless Broadband was not just a name in a provider directory. It had physical equipment on municipal steel in 2025, and that equipment was important enough for the city to address customer continuity in public.

The geography is compact. Laurinburg is a small city in Scotland County, close to the South Carolina border and west of the Lumber River basin. A water tank gives a wireless operator height, line of sight, and a ready-made elevated platform without building a dedicated tower. From such a site, radios can reach roofs, small businesses, outlying homes, or relay points where trees, terrain, and distance permit. But a tank also belongs first to the water system. Its painting cycle, structural condition, site-access rules, electrical service, safety requirements, and municipal policy can interrupt a communications network that merely rides on it.

Carolina Wireless Broadband's access layer therefore sits at the intersection of broadband service and public works.

That intersection is useful to customers only if the rest of the chain holds. A customer premises radio must see the tank. The tank radio must be powered. The site switch, if present, must stay up. The backhaul from the tank must reach an aggregation point. The upstream provider, transport circuit, or autonomous-system path must carry traffic to the wider internet. A local support person must be able to reach a failed site, change a radio, realign a dish, replace a power supply, or talk a customer through a premises fault. The water tank is the visible piece, not the whole system.

Public-service access is not the same as control

The city notice creates a clear ownership boundary. Laurinburg owns the elevated tanks and controls the maintenance project. Carolina Wireless Broadband controls, or at least maintains, its own communications equipment. That division is not unusual. Many wireless internet providers use water towers, grain elevators, commercial rooftops, churches, schools, and other tall structures as host sites. The arrangement can be locally sensible: a small operator avoids the cost of a new tower, residents gain a service option, and the host earns rent or advances a public-service goal.

The Laurinburg arrangement was especially revealing because the city said Carolina Wireless Broadband received tank access free of charge as a public service. That can be read two ways. It is a sign that the city considered the service useful enough to support. It is also a reminder that the operator's right to occupy scarce elevated space may depend on municipal tolerance, maintenance windows, and public priorities that are not identical to a commercial tower lease. The notice did not publish contract terms, access windows, notice periods, outage commitments, liability allocation, site-power responsibilities, or restoration priority.

Without those details, free tank access cannot be treated as durable operational control.

The two confirmed host sites also share a common owner. Two water tanks can improve radio coverage and reduce the effect of a single antenna failure, but they are not automatically independent infrastructure. If both are city-owned, both can be affected by the same municipal maintenance programme, the same contractor schedule, the same safety rules, and potentially the same local public-works constraints. The August 2025 project covered three tanks, including both tanks that the notice linked to Carolina Wireless Broadband equipment. The sequence itself made the shared dependency visible.

For a small fixed-wireless operator, this boundary can be more important than formal network diagrams. A carrier hotel or large data centre is designed to host communications equipment and often publishes power, access, and interconnection conditions. A water tank is designed to store water. Communications tenants are secondary. When painters need to shroud a tank, when RF exposure must be controlled, when a municipal crew needs safe access, the broadband operator's equipment becomes something to move, shut down, or work around.

This does not make the arrangement fragile by definition. It can be resilient if the operator has alternate sites, spare radios, prebuilt mounts, customer-sector overlap, a portable mast, a temporary microwave hop, and a tested plan to move traffic. The Laurinburg notice hints at such planning for Caledonia Road by saying other arrangements had reportedly been made. But a reported arrangement is not the same as a documented failover result.

The public record does not identify the alternate site, the customers shifted, the capacity available there, the duration of service continuity, or whether the arrangement was still in place after the work. A resilience claim should wait for those details.

From a customer's roof to the wider internet

The customer path in a fixed-wireless system is both simple and unforgiving. A customer has an outdoor or window-facing radio, an indoor router, and power at the premises. The radio points toward a base station or relay. The base station aggregates many customers and sends traffic toward a backhaul circuit. At each hop, the network can fail for a different reason. A tree grows into the path. Rain fade or interference cuts modulation. A power supply overheats inside a weatherproof box. A switch loses power. A backhaul circuit goes dark. An upstream route withdraws. A field technician cannot reach a site because roads are blocked after a storm.

The public evidence for Carolina Wireless Broadband supports only parts of that chain. Commercial provider directories identify the company as fixed wireless and place it in Laurinburg and nearby Gibson. InternetServices.com listed Carolina Wireless Broadband in Laurinburg and Gibson at up to 40 Mbps down and 5 Mbps up. InMyArea described the service type as fixed wireless. BroadbandNow carried a similar maximum-speed signal. Those entries support the technology class and local market. They do not prove current orderability, customer count, measured speed, or current service after the 2025 tank work and the 2026 routing changes.

The base-station side is more concrete because Laurinburg named two host sites. The notice did not say how many radios were mounted, whether the antennas were sectors or point-to-point dishes, how high they sat on the tanks, what frequencies they used, whether the links were licensed or unlicensed, what backhaul method served each tank, or whether customers could be shifted between tanks without truck rolls.

The Federal Communications Commission's fixed-wireless supporting-data guidance is useful here because it shows the kinds of engineering facts needed to evaluate such a network: base-station location, height, carrier, link budget, and clutter assumptions. None of those Carolina Wireless Broadband details were found in public sources.

Backhaul is the largest missing piece. Laurinburg's Public Works Department says the city operates more than 70 miles of fibre and more than 72 miles of electric distribution, and that its fibre connects schools, Scotland Memorial Hospital, emergency medical services, the library, county administration, and other public bodies. That tells us useful local fibre exists. It does not tell us Carolina Wireless Broadband uses it. The company could have a city fibre handoff, a leased Ethernet circuit, a microwave relay, a connection from a cable or telephone provider, a private point-to-point link, or some combination. Treating the municipal fibre ring as the company's backhaul would erase an important ownership boundary without evidence.

At the wider internet edge, the record is stronger historically and weaker now. ARIN still records Carolina Wireless Broadband as the registrant for AS30501. ARIN also records 23.157.16.0/24 and 2602:fd56::/36 to the company. Those resources show that the company had the administrative capacity to operate with its own autonomous-system number and address space. RIPEstat routing history shows that 23.157.16.0/24 was visible with origin AS30501 from 2020 through February 2026. That is meaningful: the operator was not merely reselling an invisible upstream service during that period. It had a globally observed route.

The current picture is different. RIPEstat returned no announced prefixes for AS30501 at the 13 July 2026 check. Its routing-status data for 23.157.16.0/24 showed the prefix first seen with AS30501 in July 2020, last seen in February 2026, and visible to zero RIS IPv4 peers at the query time. The ASN-neighbours data returned no current neighbours. Hurricane Electric's BGP Toolkit had previously shown a single observed IPv4 neighbour, Charter Communications AS11426, and no current global visibility after February 2026. A RADB route object with Charter-maintained proxy metadata is an administrative signal, not a live route.

This creates the central tension. The physical access layer was publicly visible in 2025; the public route layer was no longer visible in mid-2026. That does not prove the company has no customers. A small provider can operate behind upstream-assigned addresses, renumber customers, use a different network, or serve only private local traffic that public BGP collectors cannot see. But it does mean that the original network-resource evidence no longer supports a present-day independent route claim. If a customer's packets still leave Laurinburg through Carolina Wireless Broadband, the public record does not show how.

Capacity evidence stops at advertised tiers and registered numbers

Capacity is where a small wireless network can look larger on paper than it is during a wet Saturday evening. The public sources give three capacity-like signals for Carolina Wireless Broadband. First, provider directories list an advertised retail tier around 40 Mbps down and 5 Mbps up. Second, ARIN records a /24 IPv4 allocation and a /36 IPv6 allocation. Third, routing history shows the IPv4 prefix was globally visible for years. None of these is installed, powered, lit, or customer-available capacity in the engineering sense.

The advertised 40/5 figure may have been a real product tier, but it is not a measured service level. It does not state how many customers shared a sector, what backhaul committed rate supported the site, what busy-hour speeds looked like, how the link performed through trees or rain, or whether the tier was still orderable in July 2026. It is also below the FCC's modern fixed broadband benchmark. The FCC's 2024 report raised the fixed benchmark from 25/3 Mbps to 100/20 Mbps, changing the policy context for older fixed-wireless tiers.

A 40/5 product can still be valuable where alternatives are poor, but it is no longer strong evidence of adequate broadband on its own.

The ARIN resources are also easy to overread. A /24 contains 256 IPv4 addresses, but address count is not subscriber capacity. A provider can place many customers behind carrier-grade NAT, reserve addresses for infrastructure, or use only a small share of the block. Conversely, a provider can serve customers through upstream-assigned addresses while its own prefix is withdrawn. The IPv6 /36 is generous on paper, but no current public IPv6 routing evidence was established. Administrative number resources show preparedness and autonomy; they do not show active lit service.

Installed radio capacity is absent from the public record. The tank notice says equipment was present, not how much. There is no sector count, antenna inventory, modulation plan, channel width, spectrum band, maximum simultaneous throughput, or backhaul rate. Powered capacity is also absent. No source identifies batteries, generator connections, transfer switches, fuel arrangements, or runtime at the Caledonia Road or US 401 South/Purcell Road sites. Customer-available capacity is therefore unknown.

A radio can be installed and still not provide useful capacity if the backhaul is small, the sector is congested, the power system is weak, or the customer premises links are obstructed.

Competition makes these distinctions sharper. The Lumber River Digital Access Plan, published in March 2026, listed current Scotland County providers and technologies but did not list Carolina Wireless Broadband. That omission is not proof of closure; small providers can be missed. It is still a negative signal because the plan did identify cable, DSL, fibre, and fixed-wireless alternatives in the region. FOCUS Broadband separately announced rural Scotland County fibre service near Wagram with speeds up to 5 Gbps and a USDA ReConnect-supported project across Robeson and Scotland counties. North Carolina's 2022 GREAT grant announcement named Spectrum Southeast for a Scotland County award with funded-project requirements of at least 100/20 Mbps and scalability to 100/100 Mbps.

The economic conclusion is not that Carolina Wireless Broadband must disappear. Small wireless providers can survive by serving awkward addresses, responding faster than large incumbents, pricing simply, or maintaining relationships in places where fibre deployment remains incomplete. But the capacity burden rises when subsidised fibre, cable upgrades, and mobile fixed wireless enter the same geography. A 40/5 directory listing may once have looked like useful broadband. In 2026, it looks like a capacity claim that needs current orderability, performance, and resilience proof.

Power is a local network input, not background scenery

Fixed wireless is often described through antennas and line of sight, but power is just as decisive. The base station on a water tank needs power for radios, switches, heaters or cooling if present, monitoring gear, and backhaul equipment. The customer needs power for the premises radio and router. Any upstream aggregation site needs power as well. A network can have clear spectrum and good routes while still failing because the utility feed drops and the backup system is too small or untested.

Laurinburg's public works page shows that the city is not a passive backdrop. It operates electric distribution, maintains fibre, and publishes utility contacts. The power-outage page gives separate business-hours and after-hours reporting channels. The city has also reported recognition for electric reliability. Those details are relevant because a tank-mounted wireless site depends on local power restoration and local crew capacity. They do not prove that Carolina Wireless Broadband's tank equipment has a dedicated meter, battery runtime, generator plug, priority restoration, or remote monitoring.

Scotland County's hazard context argues against assuming short outages only. County emergency-management material identifies hurricanes, tornadoes, winter storms, and floods as large-scale hazards. A county archive item about Hurricane Matthew described the hurricane and following power outage in October 2016, including spoiled-food disposal. The Pee Dee Lumber Regional Hazard Mitigation Plan documents exposure to hurricanes, floods, tornadoes, downed trees, and power-line damage. None of that is a Carolina Wireless Broadband outage record. It is the environment in which tank radios and customer CPE must operate.

The physical shape of fixed wireless makes severe weather especially awkward. A storm can remove power at the tower, at the customer's home, or at the upstream point. It can push trees into the line of sight. It can damage a roof mount. It can block roads for a repair truck. It can also create immediate demand for connectivity as residents check weather, contact family, process insurance claims, use telehealth, work remotely, or run small-business payments. The value of the network rises at the same time as its repair difficulty.

CISA's emergency-communications and resilient-power guidance is useful as an analytical standard. It stresses correctly sized primary and backup power, fuel access, testing, monitoring, and decisions between batteries and generators. Applied to Carolina Wireless Broadband, the public questions are straightforward. How long can each tank site run without utility power? Are batteries sized for radios and backhaul together, or only for access radios? Is there a generator connection? Who fuels it after a storm? Does the upstream handoff have its own backup power? Can the operator see site-voltage alarms remotely?

Are customer-premises radios likely to stay powered through consumer UPS units, or does the access network survive while homes go dark?

No public source answers those questions. That absence should not be filled with generic small-ISP optimism. Resilience is proven when the failure path is documented. For now, the power conclusion is limited: local municipal power and repair capacity exist, severe-weather exposure is real, and Carolina Wireless Broadband's site-level backup-power configuration is not public.

Maintenance is a better test than marketing

The tank-painting notice matters because it describes a controlled disruption. Many network failures are messy, but maintenance is planned. A planned tank project gives the operator time to prepare, notify customers, install temporary equipment, shift sectors, or arrange alternate backhaul. If a network cannot preserve service through planned host-site maintenance, its claims during an unplanned storm or power failure deserve skepticism.

At Caledonia Road, Laurinburg said Carolina Wireless Broadband had reportedly made other arrangements to continue service. That phrase is encouraging but incomplete. It could mean radios were moved to another structure. It could mean customers were temporarily pointed to a different tank. It could mean a portable mast was installed. It could mean only some customers were shifted. It could also mean the city had been told of a plan but had not measured the result. The source does not say.

At US 401 South/Purcell Road, the notice was more conditional. The city and contractor were working with the company to minimise interruptions, but interruptions might be required if the company could not relocate equipment. The presence of three cellular carriers on the same tank adds contrast. Cellular tenants often have standard tower processes, outage notifications, temporary cell-site equipment, and larger field organisations. Carolina Wireless Broadband may have had a much smaller toolkit. The notice does not say whether the operator had spare radios, duplicate mounts, or a second crew.

The maintenance project also exposes the customer-communication problem. A small fixed-wireless network can have close relationships with customers, but customers still need notice, expected downtime, support channels, and realistic restoration windows. The public record does not show a Carolina Wireless Broadband maintenance advisory, service-status page, outage post, or customer notice related to the tank work. The company domain, checked in July 2026, did not present a working public website under normal HTTPS validation and returned a Webador "Website not found" page when inspected with certificate verification bypassed.

That website observation is not proof of operational failure. It does make public customer communication harder to verify.

Maintenance is therefore the most concrete resilience lens. It shows that the access layer had a known dependency and that the host owner could require changes. It also shows that customer continuity was discussed before the disruption. What it does not show is whether continuity worked. A better record would include before-and-after route visibility, site photos, temporary-host details, customer notices, post-maintenance restoration, and speed or outage evidence. Without those, the maintenance episode remains a strong physical proof and an unresolved resilience test.

Upstream diversity has to be physical

The upstream route is the part of Carolina Wireless Broadband's system that became least visible by mid-2026. Public route collectors are not a complete view of every private circuit, but they are useful when an operator has its own ASN and address space. AS30501 and 23.157.16.0/24 were visible for years. Then the prefix disappeared from public observation in February 2026. At the July 2026 check, RIPEstat showed no announced prefixes for AS30501 and no current neighbours. Hurricane Electric's page in the source pass also showed no current global visibility.

Historical neighbour evidence pointed to a single observed upstream relationship with Charter Communications AS11426. A single observed upstream is not necessarily a single physical path. Charter could have delivered service over diverse plant. A proxy route object could survive while an operational arrangement changes. A small provider might also have backup connectivity that is not visible as a separate BGP neighbour. But the public evidence does not prove any of that.

The safe conclusion is that public routing evidence once showed an independent Carolina Wireless Broadband origin and now does not, and that no public source proves a diverse upstream route.

CISA's resilient local-access guidance warns that redundancy can be illusory when circuits share a physical link, conduit, pole line, bridge crossing, central office, or entrance. This is especially relevant in a small market. Two commercial services can converge before they leave town. A wireless backhaul and a fibre handoff can share the same powered aggregation site. Two tank radios can depend on one upstream router. A backup provider can still fail when a common power outage or cable cut hits the local handoff.

For Carolina Wireless Broadband, the proper upstream questions are practical. Where is the demarcation point from the tank network to upstream transport? Is there one upstream carrier or more than one? Are there physically separate paths from Caledonia Road and US 401 South/Purcell Road? Does one tank relay through the other? Does traffic enter a municipal facility, a Charter facility, a telephone-company facility, or an operator-owned point of presence? If AS30501 is no longer announced, are customers numbered from an upstream provider, from a different ASN, or not currently served?

Does the RADB metadata reflect a live arrangement or stale policy?

The public record does not answer those questions. That is not a minor omission. For a fixed-wireless operator, the access radio can be the most visible asset while upstream transport is the actual single point of failure. A customer sees an antenna on a tank and assumes the network is local. The packet may still depend on one fibre leaving town. Until a route map, transport invoice, carrier statement, or BGP record proves otherwise, upstream diversity remains unproven.

Repair labour is part of capacity

The Mara Voss question for infrastructure is not only "what is installed?" It is "what can be repaired when something fails?" Carolina Wireless Broadband's public sources are thin on labour. No current staffing list, contractor roster, support hours, spare-inventory statement, or field-response standard was found. Provider directories publish a phone number, but a phone number is not repair capacity. The City of Laurinburg notice says each company maintained its own equipment, which places the tank radios on Carolina Wireless Broadband's side of the responsibility line.

This matters because fixed wireless is labour-intensive in ways that are easy to miss. Customer radios drift, brackets loosen, trees grow, roofs leak, power injectors fail, and firmware has to be updated. Base-station equipment sits outside in heat, humidity, wind, and lightning exposure. When a tank site has to be cleared for painting, someone has to remove or relocate equipment without damaging the host or breaking customer links. After a storm, someone has to decide whether the problem is a dead customer router, a failed CPE power supply, an obstructed line of sight, a tank radio issue, a backhaul outage, or an upstream route problem.

The U.S. Government Accountability Office's telecommunications workforce report gives the national context: broadband deployment and maintenance need additional workers, and rural, low-density environments create labour and economic challenges. It also notes fixed-wireless limitations around line of sight, trees, and terrain. Those are not findings about Carolina Wireless Broadband's staff. They explain why a small operator's field capacity is not a soft business issue. It is network capacity.

Local labour can be an advantage. A small Laurinburg-based provider may know the roads, the customers, the tanks, and the troublesome tree lines better than a distant call centre. It may be able to send a familiar technician faster than a large incumbent dispatch queue. But local labour can also be a bottleneck if there are only a few people who can climb, align radios, configure routers, coordinate with the city, and speak to customers. During a widespread storm, the same worker cannot be at a tank, an upstream facility, and ten customer roofs at once.

The public record therefore supports a cautious statement: repair labour is central to Carolina Wireless Broadband's resilience, but its depth is unverified. The city has its own electric and fibre crews, yet the public-works page does not say those crews maintain Carolina Wireless Broadband equipment or prioritize its sites. City support for tank access is not the same as a city guarantee to restore broadband radios. A customer-reliability assessment would need named support hours, escalation channels, spare equipment, tower-access permissions, contractor availability, and post-incident repair records.

Customer dependency is local even when the network is small

The danger in a small-provider article is to treat a weak public footprint as evidence that little depends on the network. That is not safe. A provider can be small and still matter deeply to the customers it serves, especially if those customers sit outside the best cable or fibre footprint. The 2021 Scotland County commissioners' minutes recorded concern about homes and businesses lacking reliable high-speed internet and support for a fibre project proposed to reach 1,970 addresses.

The Census Bureau estimated that 87.3% of Scotland County households had a broadband subscription in 2020-2024, but subscription is not the same as resilient service, and the remaining gap can be concentrated in rural pockets.

For customers at the edge of coverage, a fixed-wireless provider may be the service that makes remote work, school assignments, telehealth, farm administration, card payments, cloud accounting, security cameras, or family communication possible. If a tank site goes dark, the effect is not abstract. A student loses a class connection. A shop loses payment processing. A home health call drops. A resident has fewer ways to follow weather alerts. A small business may fall back to a mobile hotspot, if coverage and data plans allow it. The dependency may be modest in scale but sharp in consequence.

This is why current-service verification matters. The public record shows 2025 physical equipment and historical public routing. It does not prove the size or present status of the customer base. A regional planning report omitted the company in 2026, while provider directories preserved listings. Those signals conflict. The fairest treatment is to avoid both extremes. Do not write the company out of existence because route collectors no longer see its prefix. Do not write it up as a fully active resilient ISP because a directory still lists a 40/5 tier. The customers, if still served, deserve an evidence standard that tests the full chain.

The customer dependency also has a public-policy side. Laurinburg's free tank access was described as a public service. If public infrastructure supports private broadband equipment, the public can reasonably ask what it receives in return: coverage for otherwise unserved homes, outage reporting, minimum service levels, emergency coordination, or at least transparent maintenance communication. The 2025 notice gave the city side of the maintenance issue. It did not publish a service obligation. That absence should be corrected in future public records if the arrangement continues.

The regional market is moving around the operator

Carolina Wireless Broadband's resilience cannot be judged only by its own equipment. The surrounding market changes what customers expect and what the operator must repair. Scotland County has incumbent cable and telephone providers, municipal fibre serving public institutions, subsidised broadband projects, and new fibre phases. The Scotland County Comprehensive Plan describes Spectrum, AT&T, and Windstream as primary providers and notes municipal fibre for public institutions. The Lumber River Digital Access Plan lists current provider technologies and highlights uneven coverage.

FOCUS Broadband's rural project and Spectrum's grant obligations point toward faster fixed networks in parts of the county.

For a small fixed-wireless provider, this can cut both ways. Subsidised fibre can reduce the number of customers who need a wireless alternative, lowering revenue available for spares, upgrades, and staff. It can also leave pockets where fibre economics still fail and where a nimble wireless operator remains useful. The key is not whether fixed wireless is good or bad. It is whether this specific fixed-wireless network has enough capacity, backhaul, power, and repair depth to remain credible against higher-speed alternatives.

The federal and state policy context also raises the evidence burden. North Carolina's BEAD challenge portal describes challenge reasons such as unavailable service, excessive latency, data caps, and incorrect technology. NTIA guidance on alternative broadband technologies discusses robust service and identifies interference and capacity risks for unlicensed fixed wireless, along with mitigations such as conservative link budgets and reserved capacity. Those documents do not judge Carolina Wireless Broadband. They explain why a coverage claim must be tested at the address and engineering level.

If Carolina Wireless Broadband is still serving customers, the operator's future may depend less on matching fibre headline speeds than on proving reliability in hard places. A 40/5 service may be insufficient for many households, but a well-maintained wireless link can still matter where the alternative is no service, a weak mobile signal, or a long wait for construction. The operator would need to show current orderability, realistic speeds, no hidden data caps that impair essential use, and a plan for host-site and upstream failure. Without that, it becomes a legacy directory entry rather than a visible infrastructure provider.

What would change the assessment

The evidence needed to improve the assessment is concrete. First, the host-site record should be updated. Are Carolina Wireless Broadband radios still on the Caledonia Road and US 401 South/Purcell Road tanks after the maintenance project? If not, where did they move? If yes, what is the access agreement, and what happens during the next inspection or repainting cycle? A current municipal agenda item, lease, permit, site photo, or operator notice would materially improve the physical record.

Second, the backhaul path should be identified. The public does not need proprietary diagrams to understand resilience, but it does need the basics: whether the tanks use fibre, microwave, leased Ethernet, or another transport method; where the upstream handoff occurs; and whether the two tank sites share the same aggregation point. If city fibre is involved, the city or operator should say so. If a commercial carrier provides transport, the operator should say whether there is a backup provider or physically separate route. CISA's guidance is clear that circuit diversity has to be physical, not just contractual.

Third, the routing status should be explained. If AS30501 and 23.157.16.0/24 are intentionally withdrawn because customers moved behind an upstream provider, that is an operating change, not necessarily a service failure. If the company stopped using its own route after February 2026, the replacement arrangement should be visible through an upstream statement, customer notice, or new routing record. If the route withdrawal reflects inactivity, the public should not rely on stale ASN evidence.

Fourth, capacity should be separated into installed, powered, lit, and customer-available categories. An operator can publish a simple table: active sites, access technology, maximum plan, typical speed range, backhaul capacity band, backup-power runtime, and whether service is currently orderable by address. It does not need to reveal sensitive customer data. It does need to prevent advertised maximums from standing in for usable capacity.

Fifth, repair and customer communications should be made visible. A status page, maintenance notices, storm updates, or even a stable working website would change the public evidence profile. The July 2026 web observation, where normal HTTPS validation failed and the endpoint returned a site-builder not-found page under bypassed verification, is a negative public-communication signal. It may be temporary. If it is temporary, the fix is straightforward: restore a working public site, publish contact and outage procedures, and keep them current.

Assessment

Carolina Wireless Broadband should be read neither as a proven resilient local ISP nor as a dead network. The better assessment is narrower and more useful. The company had a real local fixed-wireless physical presence in Laurinburg in 2025, confirmed by a municipal notice that named two elevated water-tank host sites and described the operational risk of tank maintenance. It had registered number resources and a historically visible public route. Those facts are enough to analyse the network's dependency chain.

The current public evidence does not carry the chain all the way through. The route evidence that once made AS30501 visible to the world had disappeared from RIPE RIS by the July 2026 check. The company website was not functioning as a normal public operator site during the same redo window. The regional digital access plan omitted the company from its current-provider inventory. No source proves tank-site backup power, route diversity, current equipment presence, current customer count, or current orderability. Those gaps keep the network evidence grade weak.

That weak grade is not a moral judgement. It is a resilience finding. A tank-mounted fixed-wireless system can be valuable precisely because it is local, but locality does not remove dependency. It concentrates dependency in host access, power, weather, backhaul, and repair labour. The city-owned tank gives height, but the city can repaint it. The ARIN record gives administrative identity, but BGP collectors can stop seeing the route. A provider directory gives a speed tier, but not a working sector in bad weather. A phone number gives contact, but not a field crew.

For customers, the test is simple. When the tank is under maintenance, when the power is out, when trees damage the line of sight, or when the upstream route disappears, does the connection still work, and who repairs it? The public record can answer the first half only in fragments. Carolina Wireless Broadband's confirmed assets begin at Laurinburg's elevated tanks. Its unproven resilience begins everywhere packets and repair crews must go after that.