The judgement begins with a disappearing network trail
Worldstar Network should not be dismissed as a pure name without operating history. Its official site at https://www.worldstar.net.tw/ presents a Taiwan hosting and infrastructure provider offering physical servers, server colocation, cloud servers, outsourced IT service, BGP network service, and a NoNAS cloud-disk product. The same site says the company was established in 2005 and served small and medium-sized businesses with colocation and dedicated server rental. PeeringDB still carries a Worldstar Network profile for AS131603 at https://www.peeringdb.com/net/10331, listing the network as a Cable/DSL/ISP, pointing to https://www.worldstar.net.tw, showing a Taiwan Internet Exchange presence, and naming interconnection facilities in Taipei and New Taipei City. Taiwan company-record mirrors also connect Worldstar Network Co., Ltd. to the Taiwan company number 27714634, a Taipei registration, NT$5 million in capital, and the English name Worldstar Network Co., Ltd.
That is enough to say there was a real company and a real technical footprint. The harder question is whether the current public footprint still supports a live infrastructure operator under the Worldstar name. Here the evidence turns sharply cautious. Hurricane Electric's BGP Toolkit page for AS131603 at https://bgp.he.net/AS131603 says the ASN has not been visible in the global routing table since September 21, 2022, and that some displayed information is from that time. The same page now labels AS131603 as Skycloud Computing co., Ltd., not Worldstar Network, and its APNIC whois block names SKYCLOUD-TW-AS with a last-modified date of April 15, 2024. APNIC RDAP at https://rdap.apnic.net/autnum/131603 also names SKYCLOUD-TW-AS, gives Skycloud Computing co., Ltd. as the description, and shows the same 2024 change date. IPinfo at https://ipinfo.io/AS131603 likewise displays AS131603 as Skycloud Computing co., Ltd., inactive, with zero current IPv4 addresses, zero current IPv6 addresses, no listed peers, and no listed upstreams.
The practical conclusion is not that Worldstar never mattered. It is that Worldstar's current value cannot be measured as if every old network and product claim still belongs to an active Worldstar-operated platform. The most defensible reading is historical continuity with unresolved present control. The company name, site, older license and company records, and stale PeeringDB network profile support a small Taiwan hosting and colocation business with technical substance. Current APNIC, HE, and IPinfo evidence shows the ASN and address trail that once made Worldstar look like a direct network operator has moved, changed names, or gone inactive in a way that weakens current operating proof. If a buyer, partner, or analyst is testing whether "Worldstar Network" is an active local network business rather than a broad name attached to sparse registry traces, the answer is mixed: the old operating surface is real; the current standalone network proof is thin.
That distinction matters economically. A local hosting provider can create value in three ways: by owning and operating its own network resources, by packaging leased data-center and carrier inputs into a reliable service, or by keeping a brand alive while the actual operating assets have moved elsewhere. Worldstar's public trail contains pieces of all three stories. The website speaks like a provider with its own service ladder and technical team. PeeringDB preserves the language of an autonomous network with exchange and facility presence. Current routing and registry sources point away from Worldstar and toward Skycloud. The economic lens should therefore focus less on the global-sounding name and more on the control surfaces still visible: company identity, working website, facilities, AS history, DNS, licensing, support channels, and whether customers can verify who actually delivers service today.
Identity: a registered company, multiple addresses, and a brand that outlived its clearest route
The company identity is more credible than the current network identity. Searchable Taiwan company-record mirrors identify Worldstar Network Co., Ltd. with Taiwan company number 27714634, a responsible person named Hsieh Tsu-Hui or Shieh, NT$5 million in capital, and a Taipei registration. Findcompany's public page at https://www.findcompany.com.tw/%E8%8F%AF%E4%B8%96%E9%81%94%E6%9C%89%E9%99%90%E5%85%AC%E5%8F%B8 lists Worldstar Network Co., Ltd., company number 27714634, status as registered, a current address at 2F, No. 1, Section 4, Xinyi Road, Da'an District, Taipei City, and a last change date of January 8, 2024. Taiwan business-information mirrors such as https://www.twbi.com.tw/company/284390 and https://twincn.com/item.aspx?no=27714634 show similar company identifiers and connect the English company name to Worldstar Network Co., Ltd. These are secondary mirrors, not the same as a live official registry extract, but the consistency across mirrors gives the identity more weight than a random hosting directory entry.
The official website gives a different operational address. Its footer at https://www.worldstar.net.tw/ lists a Taipei Shilin District address on Fulin Road, Lane 100, Alley 77, No. 9, 2F. Older public records and NCC lists point to a Datong District address on Yanping North Road. The PeeringDB organization page at https://www.peeringdb.com/org/14206 gives only "Taipei, Taiwan, 111" and a website override of http://www.worldstar.pro/. This address drift is not unusual for a small service company that has operated for many years, but it does matter. For hosting, the office address is not just stationery. It affects who signs contracts, who receives legal notices, where tax invoices point, who holds telecom permissions, and whether a customer can reconcile the brand with the entity behind support and billing.
There is also a name-language issue. The English identity is Worldstar Network Co., Ltd. The Chinese-language website uses the Worldstar brand in a Taiwan hosting context and describes the company as established in 2005. The PeeringDB network page names the organization simply as Worldstar Network. Current APNIC/RDAP, however, names AS131603 as Skycloud Computing co., Ltd. That means the brand, legal company, and network resource are not cleanly aligned in the public record as of 2026. In the best case, Worldstar may have transferred or ceased using a network resource while continuing as a hosting-service brand or legal company. In the weaker case, the website and older profiles may represent a legacy brand whose active infrastructure relevance has faded.
The important commercial question is not whether every name variation can be explained. It is whether a customer today can tell which entity holds the service contract and which operator controls the service. The answer is not visible enough from public sources. A customer buying a dedicated server, colocation slot, or managed service should ask for the legal counterparty, company number, service address, data-center location, network ASN, upstream providers, abuse contact, and SLA. If those documents point to Worldstar Network Co., Ltd., the brand has a current operating basis. If they point to Skycloud, Chief, Chunghwa, or another provider, then Worldstar's role may be sales, legacy brand, reseller, or historical wrapper rather than current network operator.
The website sells an old-fashioned but coherent hosting ladder
Worldstar's official site looks dated, but its product ladder is coherent. The homepage at https://www.worldstar.net.tw/ lists physical servers, BGP network service, server colocation, Worldstar NoNAS cloud disk, cloud servers, and IT outsourcing. It describes system planning, deployment, upgrades, maintenance, information security, backup and redundancy, and support-team service. It says a business can reduce cost and staffing load by using Worldstar's server-host management service. That is the classic pitch of a small managed infrastructure provider: the customer does not need to hire a full-time systems team if the provider can rent server capacity, monitor it, and maintain the operating system and network environment.
The physical-server pitch is aimed at customers that want more control than shared web hosting but do not want to buy and operate hardware themselves. The site says Worldstar can provide flexible server-rental plans according to hardware, bandwidth, operating-system, and management-interface needs. This points toward a dedicated-server or bare-metal rental model rather than a hyperscale cloud model. In that model, revenue depends on hardware utilization, monthly rental commitments, support efficiency, and the provider's ability to keep servers productive long enough to recover purchase cost. The official homepage also says Worldstar purchased Dell and Cisco professional server equipment and that architectures were planned by engineers. That claim, if current, matters because dedicated hosting margins are stronger when the provider owns hardware and controls configuration rather than simply passing through another host's inventory.
The colocation pitch is similarly direct. Worldstar says customers can place their own servers in professional facilities and select plans based on private-line and bandwidth needs. That is a credible Taiwan small-business product: a customer may own a server for a website, application, trading system, school system, internal business process, or storage service but need professional power, cooling, bandwidth, and remote access. In this model, Worldstar's margin depends on rack units, power allocation, bandwidth resale, cross-connects, remote hands, and support. If Worldstar controls rack space directly, it can package service and support. If it resells capacity inside Chief Telecom or Chunghwa facilities, its margin is thinner but still can be viable if it adds local support and billing simplicity.
The cloud-server pitch appears as a managed IaaS-style offering, not a modern developer-platform story. The homepage says the cloud servers use a high-availability architecture and provide instance types and operating-system choices. It does not publish a detailed price list, API, region map, backup architecture, object storage product, compliance documentation, or current capacity statement. That does not make the service false. It means the service is likely sales-led and relationship-led, or the public website is stale. Taiwan has many small providers that sell cloud servers through human quote flows rather than self-service dashboards. For enterprise trust, however, lack of transparent pricing and current documentation raises the burden on sales documents.
The IT outsourcing offer is also economically important. Worldstar says its technical team can plan mail servers, web servers, file servers, VPN, virtualization, and private clouds according to customer needs. This is where a small provider can escape commodity hosting economics. Hosting alone is often priced down by competitors. Managed service, migration, emergency repair, monitoring, and private infrastructure design can carry better margin because the buyer is paying for local competence and accountability. But this also requires visible proof of current staff, support hours, security controls, and customer references. The official site asserts a technical team with Cisco, Linux, storage, and virtualization capability, yet public staff evidence is sparse.
The NoNAS cloud disk product rounds out the picture. The homepage links to https://www.nonas.io and describes a cloud-storage service supporting large cloud space, downloads, cloud media, VPN, and related functions. Search traces around NoNAS are thin, and the current commercial strength of that product is not clear. Still, its inclusion shows Worldstar's service portfolio was broader than simple web hosting. It tried to move from server rental and colocation into customer-facing cloud storage and managed private-cloud style services. In a small Taiwan provider, that strategy makes sense: hosting infrastructure becomes more defensible when it is attached to sticky data, backup, and managed IT workflows.
The strongest historical proof is PeeringDB, and the weakness is its age
PeeringDB is the strongest public evidence that Worldstar once had a real network operator surface. The AS131603 page at https://www.peeringdb.com/net/10331 names Worldstar Network as the organization, gives the company website override as https://www.worldstar.net.tw, classifies the network type as Cable/DSL/ISP, and lists five IPv4 prefixes and zero IPv6 prefixes. It shows traffic levels of 100-1000 Mbps, a balanced traffic ratio, Asia-Pacific geographic scope, and an open peering policy with no ratio requirement and no contract requirement. It lists sales, NOC, and technical contacts with Taiwanese phone numbers and worldstar.pro email addresses. It also lists TWIX as the public peering exchange, with a 1G capacity and IPv4 address 210.62.255.31, and two interconnection facilities: Chief LY Building Taipei and CHT Taipei Banqiao IDC.
Those details are too specific to be meaningless. A brand-only reseller usually does not leave behind a populated PeeringDB network profile with exchange, facility, contact, and peering-policy details. The Chief LY Building and CHT Taipei Banqiao IDC names also match the kind of facilities a Taiwan hosting and colocation provider would use. Chief's English IDC page at https://en.chief.com.tw/idc/ describes Chief as a professional telecommunication service provider and IDC/Tele-Center provider. Chief's data-service page at https://en.chief.com.tw/data/ calls Chief IDC a leading carrier-neutral IDC ecosystem in Taiwan. The CHT Banqiao facility fits the Chunghwa Telecom side of Worldstar's own facility and carrier story. PeeringDB therefore gives Worldstar's historical service claims a real infrastructure anchor.
The problem is update age. PeeringDB shows the Worldstar network record last updated in 2022, public peering information last updated in 2021, facility information last updated in 2017, and contact information last updated in 2020. Those dates are not disqualifying by themselves because many small network operators leave PeeringDB unchanged for years. But when the current APNIC record and BGP mirrors now point away from Worldstar, stale PeeringDB cannot carry the whole current judgement. It tells us what the network said about itself before. It does not prove that AS131603 still operates as Worldstar Network in 2026.
The Hurricane Electric page is more damaging to a current operating claim. It says AS131603 has not been visible in the global routing table since September 21, 2022. It shows one historic peer, AS17408 AboveNet Communications Taiwan, and one originated IPv4 prefix, 43.251.57.0/24, described under Skycloud Computing co., Ltd. It also shows the current APNIC aut-num block as SKYCLOUD-TW-AS with a 2024 last-modified date. IPinfo's AS131603 page shows an inactive ASN, zero hosted IP addresses, zero current ranges, zero peers, zero upstreams, and a Skycloud name. APNIC RDAP confirms the Skycloud name and Skycloud address for the aut-num. That set of sources is stronger for current routing status than PeeringDB because it reflects live or recent registry and routing-state views.
The address blocks strengthen the same conclusion. APNIC RDAP for https://rdap.apnic.net/ip/43.251.56.0 names 43.251.56.0/22 as SKYCLOUD-NET, allocated portable, with a last-changed date of April 15, 2024. APNIC's mirrored TWNIC RDAP view for https://rdap.apnic.net/ip/103.42.144.0 names 103.42.144.0/22 as SKYCLOUD-NET and points to Skycloud Computing co., Ltd. Search traces from IP databases still associate those ranges historically with Worldstar Network, but the current registry label is Skycloud. That is exactly the kind of evidence that can confuse an analyst: old IP intelligence says Worldstar; current RDAP says Skycloud. The economically conservative reading is that the valuable current network resources should not be attributed to Worldstar unless Worldstar or Skycloud publishes a current relationship.
The website itself is live enough to matter but stale enough to discount
Worldstar's website creates the second major evidence tension. The page is accessible through the web and indexed with the title "Worldstar Network Co., Ltd.: virtual hosting, cloud hosting, colocation, dedicated host, physical host, Taiwan server, Taiwan server colocation" in Chinese-language search surfaces. The homepage content at https://www.worldstar.net.tw/ is concrete: services, technical claims, product categories, company background, address, and social links. EasyCounter's page at https://www.easycounter.com/report/worldstar.net.tw shows the site resolving to 202.133.230.223, running Apache and Joomla, using Net-Chinese nameservers, and using Google mail exchangers. Local DNS checks also show worldstar.net.tw and www.worldstar.net.tw resolving to 202.133.230.223, with Net-Chinese nameservers and Google MX records.
That proves the web presence is not just an expired parked domain. It has DNS, mail routing, and indexed content. But it also reveals a weak current signal. The site is a dated Joomla-style page and even displays template residue in the page chrome. It does not offer a modern customer portal, current public price sheets, recent news, service-status posts, security certifications, audited uptime data, route collector links, or a current ASN explanation. The official contact email is protected behind JavaScript in the page output, and attempts to fetch the site directly through a local command-line client timed out while the indexed web view remained accessible. A slow or awkward website does not mean service is inactive, but for an infrastructure provider it reduces public confidence.
The domain's own hosting is also not clean proof of Worldstar-owned infrastructure. The A record points to 202.133.230.223, not obviously to AS131603 in the current public sources reviewed here. The APNIC RDAP result for the wider 202.153.160.0/19 range used by one Worldstar-associated IPinfo example points to Unigate Telecom and Chief-related contacts, not to Worldstar as resource holder. The website may be hosted on a third-party or legacy provider. That is normal for a small company, but it means the website does not verify the company's own network control.
This matters because the official site makes strong infrastructure claims. It says Worldstar has 10 professional facilities across Taipei, Taoyuan, and Hong Kong; uses lines from Chunghwa Telecom and Asia Pacific Telecom; has BGP interconnections with carriers and services including Chief Telecom, Google, and TPIX; owns Dell and Cisco equipment; and provides 24-hour monitoring. Those are useful claims, but they need current verification. "Ten facilities" and "Hong Kong" are meaningful only if the company can name current locations, contracts, remote-hands arrangements, network routes, and service levels. If the claims describe an earlier phase of the business, they should be treated as historical marketing rather than present capacity.
The lack of transparent pricing is also a signal. Many local hosting providers publish at least entry prices for shared hosting, VPS, dedicated servers, cloud servers, or rack units. Worldstar's public site, as indexed, emphasizes service categories and technical capabilities rather than a current package table. That pushes the business toward quote-based infrastructure and support rather than high-volume self-service hosting. Quote-based sales can be healthy if the company serves bespoke SME and colocation customers. It is less convincing if the website is the only current customer-acquisition surface and cannot show recent product detail.
Revenue logic: local support and rented control, not a hyperscale cloud story
Worldstar's best possible business is a local-support and rented-control business. Taiwan SMEs, schools, software houses, e-commerce sellers, and small enterprises often need practical infrastructure outcomes: a server hosted in Taiwan, stable bandwidth, a person to call, firewall and backup help, migration from a broken host, private file storage, or an in-country machine for latency and data-control reasons. A company like Worldstar can sell those outcomes without being a large carrier. It needs enough facility access, enough server stock, enough carrier relationships, and enough technical staff to make the customer feel less exposed than if it bought unmanaged capacity alone.
The revenue ladder would start with rented physical servers and colocation. A physical server creates monthly recurring revenue from hardware, bandwidth, IP address allocation, management, backup, and support. Colocation creates monthly revenue from rack space, power, ports, cross-connects, remote hands, and bandwidth resale. Cloud server revenue adds virtualization and resource pooling, but only if the provider can keep utilization high and avoid excessive support cost. IT outsourcing creates project and retainer revenue. NoNAS-style storage could create sticky recurring revenue if customers store family, business, or media files there. This is a sensible small-provider portfolio.
The margin problem is that almost every input is expensive or borrowed. If Worldstar uses Chief Telecom or Chunghwa facilities, it pays for rack space, power, access, cross-connects, and network service. If it uses third-party carrier lines, it pays wholesale bandwidth and must manage contention. If it rents or resells server capacity, it gives up part of the margin to the underlying owner. If it owns Dell and Cisco hardware, it faces capital recovery, refresh cycles, spares, warranties, and technician time. If it supports Linux, storage, virtualization, mail, VPN, and private clouds, it needs skilled engineers. If it includes public IPv4 addresses, it faces scarcity and allocation discipline. The value proposition is not "cheap bandwidth because of a global name." It is "local operations can assemble these inputs into a reliable service."
That is why the AS131603 uncertainty matters so much. A provider with its own active ASN, address space, exchange port, and peering policy has more control over routing, reputation, and cost. It can tune inbound and outbound traffic, negotiate peering, publish abuse contacts, and provide a more convincing network map. A provider without an active ASN can still serve customers, but it is more dependent on upstream carriers and data-center partners. It can be a good managed host, but not as strong an independent network operator. Worldstar's historical PeeringDB page suggested direct network control. Current APNIC and HE evidence weakens that claim.
The absence of public pricing also affects revenue interpretation. If Worldstar still serves a small base of relationship customers, price opacity may be normal because each deployment is customized. If it is trying to win new hosting customers, price opacity is a disadvantage against providers that publish VPS and dedicated-server prices. Taiwan buyers can compare against local providers, global cloud providers, and cheap VPS brands quickly. A company with a dated site and no current pricing must rely on existing relationships, referrals, specialized Taiwan routing, or hands-on support to win.
The best reading is therefore a modest business model, not a scale platform. Worldstar's valuable asset is not the word "Worldstar" or the existence of an old ASN. It is any remaining relationship with Taiwan facilities, carriers, support staff, and SME customers. If those are active, the company can still be economically useful. If those are gone or have moved to Skycloud or another operator, the Worldstar name mainly carries historical residue.
Supplier dependency is the operating story
Worldstar's public service claims openly rely on suppliers. The homepage says it uses Chunghwa Telecom and Asia Pacific Telecom lines and interconnection with Chief Telecom, Google, and TPIX. PeeringDB places AS131603 at TWIX, Chief LY Building Taipei, and CHT Taipei Banqiao IDC. EasyCounter and DNS records show Net-Chinese as domain nameserver provider and Google as mail exchanger. Chief and Chunghwa appear repeatedly as infrastructure anchors. None of this is a problem by itself. Hosting providers are assembly businesses. The question is whether the assembler has enough control and accountability to justify customer trust.
Chief is a particularly important dependency because it is not just another supplier. Chief's public materials describe a large Taiwan carrier-neutral IDC ecosystem and a broad interconnection environment. Data Center Map's Chief Telecom profile at https://www.datacentermap.com/c/chief-telecom/ calls Chief one of Taiwan's most important carrier-neutral data-center and carrier-hotel operators, with major domestic and international carrier presence in its Neihu buildings. If Worldstar is or was present in Chief LY, Worldstar benefited from being inside a dense interconnection environment without needing to own a major data-center building. That is economically rational. It gives a small provider access to carriers, exchange fabric, remote hands, and customer trust signals.
Chunghwa Telecom is the other anchor. CHT Taipei Banqiao IDC in PeeringDB indicates a facility relationship or presence in a Chunghwa environment. Chunghwa's IDC pages at https://www.idc.hinet.net/chtidc_eng/About_History.html position Chunghwa as a major Taiwan data-center services provider. For a small host, Chunghwa reach can help with domestic coverage, customer comfort, and latency. It can also raise cost. A small provider reselling or renting inside major facilities must compete against those same major operators and their direct channels.
The Skycloud shift is the supplier-or-successor question that public sources do not answer. APNIC RDAP now names Skycloud for AS131603 and related address space. Search results describe Skycloud Computing as a Taiwan cloud, CDN acceleration, edge-computing, and DDoS mitigation provider; Acer eDC's page at https://www.aceredc.com/en/serve-2-en/serve05-en/more-security-integration-2/itw-taiwan-ddos-protection-solution/ says Acer eDC joined with Skycloud as a strategic partner for a Taiwan DDoS protection project. That does not prove a Worldstar-Skycloud corporate transaction. It does show that the current resource holder named in APNIC is an active-looking Taiwan network/cloud provider, while Worldstar's public network record is older. If Worldstar's customers are now actually served through Skycloud infrastructure, the distinction should be disclosed in contracts and technical diagrams.
Supplier dependency also shapes risk in abuse and reputation. Hosting providers deal with spam, scanning, copyright complaints, compromised servers, and customer misuse. PeeringDB's old Worldstar contacts include sales, NOC, and technical emails at worldstar.pro. Current APNIC abuse contacts for AS131603 point to Skycloud. If a complaint or security incident involves historical Worldstar IP space, a customer or peer needs to know which desk is authoritative. Unclear abuse authority is not a small administrative flaw; it affects deliverability, route reputation, response time, and customer trust.
Customer dependency: SMEs that value a reachable operator
Worldstar's official site tells us its intended customer base: small and medium-sized companies that need server colocation, physical server rental, system planning, deployment, upgrade, maintenance, information security, backup, redundancy, and support. That is a customer base with practical needs rather than platform-engineering sophistication. The buyer likely values Taiwan hosting geography, Chinese-language support, a local phone number, and a provider willing to plan mail servers, web servers, file servers, VPN, virtualization, and private cloud setups. The buyer may not want AWS account design, Terraform, Kubernetes, or multi-cloud architecture.
That customer dependency can be attractive. SMEs often stay with a provider for years if the provider solves migration pain and answers the phone. Colocation customers can be sticky because moving hardware between facilities is inconvenient. Dedicated server customers can be sticky if their applications are customized. Managed IT customers can be sticky because the provider knows their mail, VPN, storage, and backups. This is the local-network economics behind a small provider: revenue is not created by owning a global cloud. It is created by being close enough, trusted enough, and technically useful enough that the customer does not switch over a small monthly price difference.
The same customer base is also fragile. SMEs are price-sensitive and often underinvest in redundancy. They may run old software, create high support load, or blame the hosting provider for application problems. They may accept vague infrastructure claims during purchase and then demand accountability during outages. If Worldstar does not publish clear service levels, support hours, backup responsibilities, security boundaries, and data-center access rules, those relationships can become expensive. A small provider's margin can vanish when low-value customers consume engineer time.
There is little public customer chatter to validate satisfaction. Searches for Worldstar Network and its Chinese brand return official pages, company-record mirrors, directory snippets, technical traces, and unrelated WorldStar entertainment results. There are not many recent customer reviews, case studies, press releases, outage reports, public testimonials, or staff posts. That absence is not proof of failure. Many small Taiwanese B2B infrastructure providers operate quietly. But it does limit the confidence of an outside analyst. The lack of market noise is itself a signal: Worldstar's brand does not appear to be generating a broad current retail hosting community.
Weak chatter should be used carefully. A Web66 inquiry page at https://0228820922.web66.com.tw/ repeats Worldstar's company number, contact person, phone, business hours, and service pitch around Taiwan data-center lines from HiNet, Chief, Sparq, and Taiwan Fixed Network. SitePrice and similar-site pages repeat the old hosting and colocation description. These are useful as echoes of the service positioning, not as proof of customer volume. They show that Worldstar's market identity was understood as Taiwan hosting, IDC lines, and server service. They do not show current order flow.
The most important customer question is therefore simple: who still depends on Worldstar today? If the company can show active colocation customers, dedicated-server rentals, managed IT retainers, and current support tickets, the business may still be meaningful despite weak public routing. If it cannot, the public footprint is mostly a legacy trace.
Competition is stronger than the old website suggests
Worldstar operates, or historically operated, in a crowded Taiwan infrastructure market. At the facility and interconnection end, Chief Telecom and Chunghwa Telecom are much stronger public anchors. Chief's materials describe carrier-neutral IDC buildings, cloud exchange services, and a broad domestic and international carrier ecosystem. Chunghwa has a national telecom and IDC position. A BusinessWire summary of a Taiwan data-center colocation market report at https://www.businesswire.com/news/home/20241022330628/en/Taiwan-Data-Center-Colocation-Supply-Demand-Analysis-Report-2023-2029-Chunghwa-Telecom-and-Chief-Telecom-Stand-as-Prominent-Providers---ResearchAndMarkets.com says Chunghwa Telecom and Chief Telecom are prominent Taiwan colocation providers and together contribute a large share of market IT load capacity. Even if the exact market-share figure depends on the paid report, the strategic point is obvious: Taiwan colocation is not short of larger infrastructure suppliers.
At the cloud and security-services end, Skycloud, Acer eDC, Taiwan AI Cloud, global cloud providers, and regional VPS brands create more current-looking alternatives. Skycloud's public mentions emphasize CDN acceleration, edge computing, and DDoS mitigation. Acer eDC's partnership page frames Skycloud as a strategic partner in a Taiwan DDoS protection project. Taiwan AI Cloud at https://www.twcloud.ai/en_us/company/ positions itself around sovereign AI and high-performance cloud infrastructure. These competitors may not chase every SME colocation customer, but they raise buyer expectations for current documentation, security positioning, and active engineering presence.
At the low-end hosting and VPS layer, price competition is severe. Taiwan and Asia-Pacific buyers can choose local dedicated-server providers, international VPS platforms with Taiwan locations, and regional hosts selling cheap monthly plans. Yuan-Jhen's Taiwan colocation page at https://yuanjhen.com/chunghwa-telecom-hosting-service markets direct Chunghwa Telecom routing, Taiwan server colocation, smart hands, and Asia-Pacific latency. LightNode's Taiwan VPS page at https://go.lightnode.com/taiwan-vps advertises hourly billing, NVMe SSD, KVM virtualization, and a Taipei data center. These offers are easy to compare online. A provider with no current price list must win on relationship and specialized support, not on digital acquisition.
The global cloud layer adds another kind of pressure. AWS, Google Cloud, Microsoft Azure, Oracle, Alibaba Cloud, and others either operate in Taiwan's region, nearby Asia-Pacific regions, or connect through Taiwan carrier hotels and cloud exchanges. They do not replace every local dedicated server or colocation service. They do change procurement psychology. A serious enterprise workload now asks why it should use a small provider unless that provider offers lower operational friction, local support, in-country network advantages, special routing, private hardware, or practical managed service. "We have servers in Taiwan" is not enough.
Worldstar's competitive path, if active, is therefore narrow but plausible. It can win if it is a specialist for Taiwan SMEs that need hands-on infrastructure help, local colocation, private server management, and a known support person. It cannot win as a broad global cloud brand. It cannot rely on stale AS evidence. It needs current service proof: named facilities, current network map, clear legal counterparty, current pricing, support commitments, and customer references.
Regulatory and accountability evidence is helpful, but historical
Regulatory traces improve Worldstar's credibility but do not solve the current operating question. Taiwan's National Communications Commission published historical lists of second-class telecommunications operators. Searchable NCC PDF text from the 2016 list at https://api.ncc.gov.tw/chncc/app/data/doc?aplistdn=undefined&detailNo=12&id=1020&module=commonMessage1020&preview=undefined&serno=35162_2001_news&type=s includes Worldstar's Chinese legal name in the operator list under a license number. A 2019 NCC operator list at https://api.ncc.gov.tw/chncc/app/data/doc?aplistdn=undefined&detailNo=9&id=1019&module=commonMessage1019&preview=undefined&serno=41051_2002_news&type=s also appears in search text with the company name, Datong District address, and phone/fax details. Taiwan business-record mirrors list a second-class telecommunications business item among the company's registered business activities.
This is meaningful because telecom permission is not the same as a generic web-design company registration. It suggests Worldstar was recognized in a regulated communications-service category. It also fits the official service portfolio: bandwidth resale, internet access, colocation, and managed network service. In a historical analysis, that supports the conclusion that Worldstar was a real telecom-adjacent infrastructure provider.
But the regulatory evidence reviewed here is not current enough to prove today's service state. Taiwan's telecom framework changed over time, and older second-class telecom lists may not directly map to current registration obligations. The company-record mirrors show a January 2024 company change, but a company change does not prove telecom service activity. A current official registration extract, current telecom registration or filing, current service terms, and current data-center/customer contracts would materially change the confidence level.
Operational accountability is also unresolved. PeeringDB's old Worldstar profile publishes contacts, but APNIC's current AS131603 abuse and technical contacts point to Skycloud. The website's footer contact is obfuscated. DNS and web traces prove a domain, but not a support desk. For infrastructure customers, accountability means knowing who answers abuse, who handles outages, who owns the hardware, who pays the facility, who signs the SLA, who can access the rack, and who has authority over routing. Public sources do not answer those questions cleanly.
This is the core risk for Worldstar's public image. The company looks more legitimate than a shell because of registration, old telecom traces, official service pages, and PeeringDB detail. It also looks less convincing than an active infrastructure operator because current routing and contact authority are not aligned under Worldstar. The result is a mid-confidence identity and a low-confidence current network-control claim.
What facts would change the judgement
The first fact that would change the judgement is a current statement explaining AS131603. If Worldstar sold, transferred, leased, discontinued, or rebranded the ASN and associated address ranges, that should be stated. If Skycloud operates the network on Worldstar's behalf, that relationship should be documented. If Worldstar no longer runs an autonomous network but still provides managed hosting through supplier networks, that is also acceptable, but customers should know. A simple current network page naming the ASN used for customer service, facility locations, upstreams, peering points, and abuse contacts would remove much of the uncertainty.
The second fact is current facility evidence. PeeringDB names Chief LY Building Taipei and CHT Taipei Banqiao IDC, but the facility data is old. A current list of service locations, cabinet capacity, power options, cross-connect providers, remote-hands terms, and data-center operator relationships would show whether Worldstar still has real operating surface. A customer contract or sales document that names the facility and legal counterparty would be stronger than a public marketing phrase about "10 facilities."
The third fact is current customer and support proof. Case studies, public maintenance notices, service-status history, a knowledge base with recent articles, staff profiles, or customer references would show whether the business is alive beyond the website. Even modest evidence would help. A small provider does not need to look like a hyperscaler. It does need to look reachable and accountable.
The fourth fact is current pricing or quote structure. Worldstar does not need to publish every enterprise price, but current starter packages for server rental, colocation, cloud servers, and managed support would clarify the business model. Prices would reveal whether the company is competing on low-cost commodity hosting, bespoke managed service, or data-center access. Without prices, outside analysis has to infer economics from service categories.
The fifth fact is current regulatory and company status. A fresh official company extract, telecom registration, tax status, and current business address would reconcile the different addresses in public records. That would matter for buyers who need formal invoices, data-processing terms, or procurement approval. It would also help separate Worldstar Network Co., Ltd. from unrelated "Worldstar" logistics, entertainment, and shipping names that now crowd search results.
Public evidence to keep close
The most useful official anchor is https://www.worldstar.net.tw/. It shows the brand's self-description, service categories, 2005 founding claim, SME infrastructure positioning, facility and carrier claims, and footer contact/address information. It supports the conclusion that Worldstar has a real public service identity, but its dated presentation and lack of current pricing limit how much current operating strength can be inferred.
The most useful network-history anchor is https://www.peeringdb.com/net/10331. It shows AS131603 under Worldstar Network, website override to worldstar.net.tw, Cable/DSL/ISP network type, traffic level, open peering policy, TWIX 1G presence, and Chief/CHT facility listings. It supports a historical network-operator footprint. Its own update dates make it weak as a current proof source.
The most useful current-routing anchors are https://bgp.he.net/AS131603, https://rdap.apnic.net/autnum/131603, and https://ipinfo.io/AS131603. Together they show that AS131603 is not currently visible as Worldstar in the way PeeringDB suggests. HE says it has not been visible in the global routing table since September 21, 2022. APNIC RDAP names SKYCLOUD-TW-AS and Skycloud Computing co., Ltd. IPinfo shows Skycloud, inactive status, and zero current addresses, peers, and upstreams.
The useful address-space anchors are https://rdap.apnic.net/ip/43.251.56.0 and https://rdap.apnic.net/ip/103.42.144.0. These show Skycloud labels on address ranges that appear in historical Worldstar-related traces. They support the judgement that old IP-intelligence associations should not be treated as current Worldstar control without updated confirmation.
The useful company-identity anchors are https://www.findcompany.com.tw/%E8%8F%AF%E4%B8%96%E9%81%94%E6%9C%89%E9%99%90%E5%85%AC%E5%8F%B8, https://www.twbi.com.tw/company/284390, https://twincn.com/item.aspx?no=27714634, and the January 2024 company-change open-data result at https://gcis.nat.gov.tw/mainNew/opendata/11301/company_change.csv?v=11301. They support the registered-company identity, company number, capital, address history, and English name, while remaining secondary or open-data sources rather than a fresh customer-contract proof.
The useful market-context anchors are Chief's IDC and data pages at https://en.chief.com.tw/idc/ and https://en.chief.com.tw/data/, Data Center Map's Chief profile at https://www.datacentermap.com/c/chief-telecom/, the Taiwan colocation-market summary at https://www.businesswire.com/news/home/20241022330628/en/Taiwan-Data-Center-Colocation-Supply-Demand-Analysis-Report-2023-2029-Chunghwa-Telecom-and-Chief-Telecom-Stand-as-Prominent-Providers---ResearchAndMarkets.com, Yuan-Jhen's Taiwan colocation page at https://yuanjhen.com/chunghwa-telecom-hosting-service, and LightNode's Taiwan VPS page at https://go.lightnode.com/taiwan-vps. These sources explain why a small provider must justify itself against stronger facility operators, more current cloud/security providers, and transparent low-cost VPS alternatives.
Final judgement
Worldstar Network's public footprint supports a real historical Taiwan hosting and network-services operator, but it does not support a strong current claim to active independent network control. The official site, company records, NCC traces, and PeeringDB profile show that the name was attached to colocation, server rental, cloud server, BGP, and managed IT services. That is more than a generic broad name. Yet the current evidence that matters most for network operations has shifted: AS131603 is inactive in HE's routing view, APNIC now names Skycloud for the aut-num, IPinfo shows no current AS resources, and related address ranges are currently labeled Skycloud. The old Worldstar network profile remains useful history, not enough present proof.
The economic value, if Worldstar is still active, is likely local and relationship-based: Taiwan facility access, server hosting, outsourced IT, private infrastructure work, and support for SMEs that need a reachable operator. The economic weakness is the lack of public currentness: no clear active ASN, no current route map, no transparent pricing, sparse recent customer chatter, inconsistent addresses, and no visible explanation of the Skycloud transition. For a buyer, Worldstar may still be a viable service name if the current contract, facility, support desk, and technical diagrams check out. For an outside analyst, the public evidence is too thin to price it as a strong independent ISP. The burden of proof now sits with current operating documents, not with the legacy brand or PeeringDB page.