Summary

  • AMS-IX says the Eastern Caribbean IX now reaches Saint Eustatius, Saba, Saint Martin, Saint Barths and Saint Kitts, combining EUTEL’s local presence, SSCS submarine fibre and the AMS-IX exchange platform.
  • Geographic reach is not a traffic measurement. The public evidence does not yet show post-expansion peer participation by island, local traffic share, latency change or customer outcomes.

On 7 October, AMS-IX announced that five Eastern Caribbean islands were now connected through the Eastern Caribbean Internet Exchange. The headline sounds like a completed regional network. The release describes something more specific: an interconnection platform whose physical reach crosses five named islands and whose value will increase if additional networks join.

Those are different claims. A cable can reach an island without every network on that island being present at an exchange. An exchange can list participants without proving which routes they announce, accept or use. And a shorter possible path does not, by itself, prove that a user’s traffic took it or that service improved.

The announcement identifies three partners with distinct roles. EUTEL provides local presence and infrastructure on Saint Eustatius. Saba Statia Cable System, or SSCS, supplies submarine-fibre connectivity linking the islands. AMS-IX provides the technology and expertise for the exchange platform, where networks and content providers can connect and exchange traffic. The five named endpoints are Saint Eustatius, Saba, Saint Martin, Saint Barths and Saint Kitts.

The division of work matters because it separates an exchange from the transport beneath it. An internet exchange point creates a place and a set of technical arrangements through which participating networks can peer. The fibre system carries bits between locations. Local access networks then connect households, public bodies and businesses to their providers. For one customer’s request to remain within the region, each layer needs a suitable route and the relevant networks must have chosen to exchange the traffic that way.

AMS-IX says the new connection can let networks exchange traffic more directly instead of sending it along longer international routes. The operator says this can reduce latency, improve performance and use international connectivity more efficiently. Those are plausible effects of local interconnection, but the release reports no before-and-after latency, traffic volumes, route shares, outage statistics or island-by-island measurements. It also does not name the large content provider it says is already connected. The release is evidence of the operator’s deployment and intent, not a field study of results.

A public PeeringDB entry offers a useful, if limited, reference point. Its AMS-IX Eastern Caribbean record lists the exchange in Sint Eustatius and shows seven networks and eight connections. The record says it was last updated on 3 April 2025, well before the October 2026 expansion. It therefore cannot establish the current post-expansion membership or route activity. It shows why a dated exchange directory must not be presented as a live traffic census.

Other evidence helps separate physical resilience from local routing. A Dutch government evaluation published in February 2026 describes the SSCS cable as linking Saba and Sint Eustatius with Sint Maarten, Saint-Barthélemy and St Kitts/Nevis. It says design features make the system substantially redundant, support purchases from different market parties and include a redundant route between Bonaire and the two northern islands toward European Netherlands. That is a finding about the cable design and route options.

It does not show how much traffic the new exchange keeps local, and it is not a guarantee that every access network or exchange service will remain available during a fault.

The same evaluation describes a public-interest case for SSCS: its connection is important to two small islands, while the surrounding telecom market is international and concentrated. It concludes that continued government participation remains justified and proportionate. It also identifies a different risk: governance and transparency needed strengthening. The government accepted the recommendation to establish a supervisory board, and SSCS later reported that its first board would begin work in September 2026. A cable can have redundant paths while its company still needs clear oversight and public reporting.

Physical resilience and institutional accountability are complementary controls, not substitutes.

The evaluation’s historical cost evidence also needs careful boundaries. EBA reported that IP-transit costs from Saba and Saint Eustatius to Miami fell from USD 450 per Mbit in 2012 by 99% since 2021. It linked the decline partly to lower transit costs, while also noting end-user subsidies, discounts, changing data allowances and prices. That record predates the new exchange. It does not isolate any effect from the October extension, and it cannot be used to promise further price cuts.

The next proof should be operational and dated. First, publish which networks have joined after the expansion and which island each serves. Second, document the route policy: which prefixes are announced, what is accepted through route servers or bilateral sessions, and under what conditions traffic can leave the region. Third, report aggregate local-exchange traffic and route availability over a stated period. Fourth, compare latency and packet loss from representative probes before and after, with endpoints, times, and measurement methods disclosed.

Separate cable outages, exchange incidents and local-access failures so a single “connected” label does not conceal where service depends on another layer.

The current evidence supports a real milestone: the Eastern Caribbean IX’s announced reach now spans five islands, using a named combination of local presence, submarine fibre and exchange technology. It supports the possibility of more direct regional interconnection. It does not yet establish universal island participation, a measured shift in traffic paths or a customer-level service gain. The exchange’s success will be visible not in the island count alone but in the networks that join, the routes they use and the outcomes operators can show.

Heng Lu’s Note 36 is an editorial lens for the evidence discipline here: describe how the system works and distinguish documented operation from advocacy. It is not evidence about the Eastern Caribbean IX or its performance.

Sources: AMS-IX’s 7 October announcement; the Dutch House of Representatives’ government letter on the SSCS evaluation and the EBA evaluation; the PeeringDB exchange record; Packet Clearing House’s PoP register; and SSCS’s company information.

Sources