Topic
Software Lifecycle and Lock-in
Within the Topic facet, Software Lifecycle and Lock-in 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.
IETF
The First Packet Left Before the Image Existed. It Still Did Not Prove Low Latency: RFC 9828
A sender can now begin carrying a JPEG 2000 image while that image is still being encoded. That is a useful reduction in one waiting stage. It is not a receipt for the time at which a receiver recovered, decoded and displayed a usable picture.
Creators
Ivan Pepelnjak and the discipline of testing network automation
Ivan Pepelnjak’s career runs from Slovenia’s early internet and exchange infrastructure to ipSpace.net and the open-source netlab project. His recurring argument is practical: automation becomes trustworthy only when intended state is modelled explicitly, vendor claims are…
IETF
The Merge Was Clean. The Baseline May Not Be: NETCONF Private Candidates
`draft-ietf-netconf-privcand-10` gives each client a private place to prepare configuration and a defined way to detect and resolve overlap before commit. The operational mistake would be to let isolation, an empty comparison or a successful merge inherit authority over…
History
One Router Drew the Path. No Router Needed the Whole Map: RFC 1992
Nimrod’s sharpest scaling decision was to abandon the comforting fiction of a single, complete map: knowledge could be partial, private and stale, provided that a chosen path did not depend on every router improvising from a different picture.
IETF
The Transform Was Agreed. The Replay Was Not Yet Rejected: RFC 9827
RFC 9827 gives IKEv2 a broader and more honest contract for packet sequence numbers. The selected Transform ID describes what should be true when an SA's packets enter the network; it does not certify sender coordination, receiver anti-replay policy, the packet stream that…
IETF
A Diagnostic Plan Is Not a Root Cause: Eight Receipts for Scheduled OAM
`draft-ietf-opsawg-scheduling-oam-tests-07` makes network diagnosis programmable as timed, ordered test sequences. The discipline begins when operators refuse to let a stored plan, a green status or a returned metric stand in for the execution, causality, authority and outcome…
IETF
The color matched. The proof did not: RFC 9832’s transport-class evidence chain
Three BGP fields can carry the same 32-bit value and still answer different questions. RFC 9832 supplies an ordered mechanism for classful transport; operators still need a chain of receipts proving which namespace, attribute, database, fallback and forwarding state actually…
History
The Grant Was Automatic. The Boundary Was Not: RFC 1988
RFC 1988 removed the need to ask Hewlett-Packard for the same narrow permission implementation by implementation. It did not remove the perimeter around that permission—or the switch that could end it.
History
The Packet That Could Not Be Decrypted: RFC 1969’s Cross-Packet Chain
A lost encrypted packet in RFC 1969 did more than remove its own contents. It also took away the input needed to decrypt the next packet—yet the ciphertext of that unreadable successor carried enough state to let the receiver continue after it. This peculiar two-packet shadow…
IETF
The watts fell. The proof did not: RFC 9845 and the green-network control loop
A power graph can move down while the network’s carbon cost, workload allocation and service risk remain unknown. RFC 9845 makes visibility useful; it does not let one meter speak for the whole system.
History
The Uncompressed Packet That Changed the Dictionary: RFC 1967
An LZS-DCP packet can arrive with its data plainly uncompressed and still alter the shared state needed to decode the next compressed packet. RFC 1967 made that choice explicit, split state into numbered histories, and carried recovery signals inside later data packets. The…
IETF
CMIS Revision 04 Adds Control Handover but Leaves the Last Write Unsettled
Imagine a remote optical controller starting a three-register tuning sequence. Before the final write lands, the host removes that CMIS page from the remote-write allowlist and restores its own configuration. Which system owns the half-finished state? Revision 04 of a new IETF…
IETF
Revision 04 Promises UUID Identity, but Its Energy Schema Still Points to a Local Name
A data model can be syntactically valid and still give readers two different answers to the same governance question. Revision 04 of the IETF GREEN power-and-energy YANG draft says its component reference is bound to a globally unique UUID. The module printed a few pages later…
Story
RIPE NCC’s K-root Refresh Needs Three Acceptance Records, Not One Batch Status
Amsterdam can be refreshed while the three-site programme remains in progress. That is not a contradiction; it is a warning about the unit of evidence. RIPE NCC has named three K-root core sites whose seven-year-old routers and servers were due for replacement. The useful public…
IETF
DKIM2 Says Dual-Sign Until It Is Ubiquitous. Revision 01 Does Not Define the Exit
Migration advice becomes governance when it tells operators to run two systems until a condition is met. Revision 01 of the DKIM2 Best Practices draft names that condition—DKIM2 should be “effectively ubiquitous”—but supplies no common denominator, observation window or decision…
IETF
The manifest followed the telemetry. The verdict did not
Revision 14 gives collected network data a companion record of its platform, schema and collection conditions. That makes a datapoint easier to read later; it does not make the datapoint complete, the clock trustworthy or the resulting decision correct.
IETF
The log line parsed. The authority to read the connection did not: RFC 9850
RFC 9850 gives diagnostic tools a common way to read TLS connection secrets. The grammar is small; the authority it can transfer is not. A usable key-log record proves neither permission nor safe closure.
CASE FILE
The Cookie Returned. The Service Had Not: RFC 9853
RFC 9853 lets a DTLS endpoint test a peer address after a Connection ID recovers an existing security context from a datagram whose source has changed. The returned cookie can justify one binding decision at one moment. It cannot explain the change, promise a stable path, or…
IETF
The verifier subscribed. The evidence chain still had seven tests
An IETF draft turns device attestation from a polling exercise into a stream. That is a useful change in tempo, but a signed notification is still only one link between a measurement and a decision.
CASE FILE
The Paths Were Paired. The Round Trip Was Not Proven: RFC 9854
RFC 9854 gives a low-power network two related route discoveries when one radio direction cannot stand in for the other. That pairing prevents control state from being confused; it does not turn two temporary graphs into evidence that a request went out, a response came back, and…