Topic
Network-resource Evidence
Within the Topic facet, Network-resource Evidence topic intelligence connects articles that share a specific subject, signal focus, or monitoring theme. The page gives readers a richer path through related reporting, source evidence, market actors, and infrastructure implications, with enough context to understand why the topic matters across company movements, governance decisions, regional exposure, and operational risk. Readers can compare recurring signals, affected organisations, public evidence, market context, service continuity, procurement, competition, compliance, and strategic planning questions behind the subject instead of stopping at a thin list of matching articles. It explains what the topic covers, which infrastructure actors or policies are involved, what evidence supports the coverage, and why the subject may matter for operators, customers, investors, and policy readers.
History
ROHC Began Without a Return Channel: RFC 3095
Before ROHC could trust an acknowledgment, it had to work where no acknowledgment could come back. RFC 3095 therefore made every compression context begin in unidirectional mode, then let the decompressor—not the compressor's optimism alone—request the move to feedback-driven…
IETF
The Peak Had a Timestamp. The Fifteen Minutes Still Had No Sequence
A new BMP statistics format can expose the peak that disappeared between two identical snapshots. It still cannot say whether the route surge lasted seconds, dominated the window or arrived before the collapse.
Europe and Middle East Cloud Services
The Routing-Policy Swap Behind AS210328: What the August 2026 Registry Edit Changed — and What It Did Not
The Routing-Policy Swap Behind AS210328: What the August 2026 Registry Edit Changed — and What It Did Not intelligence summary explains the development, the public evidence available to readers, the organisations involved, the regional context, market exposure, and the…
Story
The contractual tail: how African LIRs stayed RIPE NCC members while AFRINIC served them
The handover of Africa's internet number registry is usually told as a single act of recognition. The dated instruments tell a more layered story: for over ten months after AFRINIC began operating, African members were legally RIPE NCC members under the RIPE NCC Standard Service…
CASE FILE
The Draft Had Eight Numbers. The Registry Still Had None
On 1 October 2026, an OpenPGP implementer could place two official-looking documents side by side and receive two different answers. Revision 05 of a working-group draft called eight composite algorithms 37 through 44. The live IANA registry still called the entire range from 37…
History
Foo Entered the RFCs Before the RFCs Explained It: RFC 3092
In December 1971, RFC 269 told a user to rename `FOO` as `BAR` and assumed the example needed no cultural translation. Three decades later, RFC 3092 counted roughly seven per cent of the RFC Series carrying the same private vocabulary. The explanation arrived after the convention…
IETF
The Site Reported Zero. Its Routes Could Still Remain in BGP
A revised edge-metadata draft turns one zero-value update into a site-wide forwarding decision. The control is compact; proving which routes, routers and live flows actually obeyed it is not.
History
The Firewall Allowed HTTP. It Had Not Approved the Packet Inside: RFC 3093
RFC 3093 proposed putting a complete IP datagram inside an HTTP message, passing it through the ordinary web opening, and reinserting it behind the firewall. Published on 1 April 2001, the comic design exposed a serious limit: a policy decision about an outer envelope says…
IETF
The PSP Key Arrived. No Protected Packet Had Crossed the NIC
The IKEv2 exchange can finish with two authenticated peers, an agreed PSP proposal, traffic selectors, SPIs and a wrapped receive key travelling in each direction. The control plane has done real work. Yet neither peer has proved that its sender installed the returned key, that a…
IETF
The Allowlist Could Reach Zero Errors. First the Configuration Had to Be Complete
A new IntraSAV draft replaces route inference with explicit source-prefix authorization. Its strongest promise begins where the operational proof is still missing: the completeness of the configuration installed on each interface.
History
The Same Zone Was Secure Here and Unsecured Elsewhere: RFC 3090
In 2001, DNSSEC could not answer “is this zone secure?” without first asking who was resolving it, which trust anchor that resolver held, and which path it followed. RFC 3090 named that awkward reality. A signed zone could belong to a local island of security and still look…
History
The Root Returned a Referral, Not the Directory: RFC 3088
In a distributed directory, the first server may know where the next one lives without knowing the answer to the user's search. RFC 3088 made that distinction concrete: OpenLDAP's experimental root service read DNS service records and handed clients LDAP URLs, leaving the…
CASE FILE
Tideo Administration: The Registry Record That Refused to Move
Tideo Administration: The Registry Record That Refused to Move intelligence summary explains the development, the public evidence available to readers, the organisations involved, the regional context, market exposure, and the infrastructure consequences that may follow. The CASE…
History
The URI Chose the Voicemail Menu. It Did Not Authenticate the Caller: RFC 3087
In 2001, SIP could carry a call to voicemail without carrying every trustworthy clue that an old telephone system used to choose the first greeting or menu. RFC 3087 turned the request target itself into a service instruction. That solved a routing problem, but it deliberately…
IETF
The Debug Header Named Every Step. It Could Not Prove Any of Them Happened
A new DKIM2 draft gives interoperability testers a common forensic trail inside the message. Its precision is useful precisely because its authority is deliberately zero.
Europe and Middle East Cloud Services
NovaCloud-Hosting after FFM2: the registry footprint that refuses to move
Six weeks after the operator declared its 19-day Frankfurt outage resolved, a dated re-read of the registry and routing record shows a footprint that has neither corrected nor collapsed — it has simply stayed contradictory.
IETF
The RADIUS Server Parsed the Wi-Fi Metric. It Still Did Not Know What Happened
An Access-Request can arrive with a perfectly valid Connect-Info string: a transmit rate, an RSSI and enough syntax for a policy engine to read both. That is useful progress. It is not yet evidence that the radio measured the right frames, that another vendor would report the…
History
The Policy Server Pushed the Rules. A Reconnect Had to Reconcile What the Device Kept: RFC 3084
RFC 3084 tried to make policy provisioning a disciplined conversation between a server and a network device. Its most revealing work began after delivery: a transaction needed a receipt, a failure needed rollback, and a broken connection left two machines that could remember…
IETF
The Refresh Ended. The Collector Still Needed Proof It Had the Whole RIB
A new BMP proposal can repair one stale routing view without tearing down every monitored session. Its ending marker closes a replay; it does not independently certify that nothing was lost inside it.
History
A Revision Number Could Order a NewsItem. It Could Not Fetch One: RFC 3085
NewsML could let one logical news item live at several addresses, so RFC 3085 gave it one URN and a per-item revision number. The design was precise about which edition came later and what kind of update it represented; it left resolution and validation to a provider that might…