Wiegand versus osdp ownership split: A practical account of who owns the reader-to-controller data path when a Wiegand or OSDP lock integration is specified,.

Diterbitkan Pada: Ogs 05, 2024

Who Owns Wiegand vs OSDP: Supplier, Integrator or Client

Who Owns Wiegand vs OSDP: Supplier, Integrator or Client

Most integration disputes about door hardware are not really about the hardware. They happen at the moment a credential is presented at a reader and the system has to decide, and at the moment that decision does not happen and someone is standing in front of a locked door at eight in the morning with a contractor badge that will not read. The wiegand versus osdp ownership split is usually raised only after the second moment, when three parties discover they each hold a partial truth and none of them holds the whole path.

Direct answer. To put it plainly: wiegand versus osdp ownership split has to hold for wiegand versus osdp ownership split. wiegand-vs-osdp has to hold at the same time. The sections below describe the order wiegand versus osdp ownership split is settled in. A wiegand versus osdp ownership split shows what each component must assume about the others. wiegand versus osdp ownership split is the interface that must be satisfied. Then commissioning starts on wiegand versus osdp ownership split.

  • Draw the interface — write down what hands over between lock, door and system at wiegand versus osdp ownership split
  • Exercise the handover — record the sequence that proves wiegand-vs-osdp works end to end
  • Archive the record — store the commissioning evidence with the file for protocol decides what is
  • This is the sequence wiegand versus osdp ownership split has to follow, anchored on wiegand vs osdp.

The argument I want to make is narrow and practical. The ownership split is not decided by which wire protocol ends up on the wall. Wiegand and OSDP change what each side can know about the other, and that changes where the diagnostic burden falls, but neither protocol hands out responsibility. Responsibility is assigned by someone, in writing, before the reader is mounted. This article works through where that assignment usually goes wrong and how to draw it so it survives commissioning, handover and the first bad morning.

Why the ownership question arrives after the reader is already on the wall

Ask an integrator to describe a door that failed and you will usually get three accounts, each internally consistent. The supplier will say the reader is sound because the unit passed a bench test before it was boxed. The installer will say the wiring is sound because continuity and insulation checks were signed off. The client will say the badge is sound because it opened a different door yesterday. Each statement is true about a fragment, and none of them reaches the point where the fragment becomes a door. Specifications at this level come off the smart door locks.

That gap is the ownership split, and it only becomes visible when something fails. Before the reader is installed, the question sounds academic — who owns the interface between the lock's reader and the access controller? After a door fails to open at a time when a delivery, a shift or a fire drill depends on it, the question is the whole of the work.

The reason it is raised late is structural. During evaluation, attention goes to the visible features of a device: the credential types, the finish, the reporting options, the connectivity story. The reader-to-controller boundary is invisible in a catalogue, because it is not a feature of either device. It is a property of the pair and of the building they are installed in.

So the split has to be drawn deliberately, by somebody with authority to draw it. In practice that somebody is the integrator, on behalf of a client, with the supplier answering the part that only the supplier can answer. This article is written from that seat, because the ownership split has more useful texture when it is described by the party that has to hold all three boundaries at once.

What the reader-to-controller data path actually carries

Before ownership can be assigned, the data path has to be described in terms that all three parties recognise.

Wiegand is a legacy wiring convention between a reader and a controller. It carries a data line and usually a power and a ground, it says nothing about encryption, and it carries no device identity. A reader emits a code when a credential is presented; the controller receives the code, looks it up in its own memory, and decides. There is no acknowledgement travelling back to the reader, and there is no way for the controller to ask the reader a question over the same pair of wires.

That single fact has an ownership consequence that people consistently underestimate. On a Wiegand path, neither end can discover that the other end is not there. A reader with a failed output stage, a broken data line, a reader assigned to the wrong controller input, or a controller that never received the code produces exactly the same symptom at the door: nothing happens. The silence is symmetrical. Someone has to own the end-to-end path, because the protocol will not surface the fault for you.

OSDP is a specified, addressed, bidirectional protocol with device addressing and a defined message structure. It is a protocol specification, and whether a given product implements it, and which revision it implements, is a per-SKU question to confirm on the dated product page for the specific SKU.

The direction of travel is the reverse. Because the exchange is addressed and bidirectional, the controller can address a particular reader, and the reader can answer. Faults that Wiegand renders invisible — a reader that is not answering, a reader whose address does not match the configuration, a reader reporting a fault condition — become visible in a status exchange. But visibility is not the same as correctness. An address can be visible and still mismatched. A reader can answer every poll perfectly while carrying firmware whose expectations differ from the controller's configuration. A specification can define a secure channel while a deployment uses a revision that does not include one.

So OSDP changes the shape of the failure and the shape of the paperwork, not the existence of a boundary. It also changes the physical assumptions: the specification is written around a balanced, addressed physical layer with its own rules on cable type, distance and termination, and the wiring that works over a short Wiegand run is not automatically acceptable for the longer runs an addressed bus typically demands. Confirm the physical layer, the distance limits and the termination arrangement for the specific implementation before assuming a cable route is reusable.

Both paths carry the same essential message: the reader reports, the controller decides. The lock reports the state; the system decides what the state means.

The protocol decides what is visible; ownership decides what is owed

It helps to separate two ideas that get bundled together in tender documents. Visibility is what the system can detect about itself. Obligation is who is answerable when something does not work. The protocol sets the visibility. The contract sets the obligation.

Table 1 — the wiegand versus osdp ownership split expressed as the difference between what is visible and what is owed:

AspectWiegand pathOSDP pathOwnership consequence
Reader identity on the wireNone; the controller cannot ask which reader sent a codeAddressed; each device answers for itselfAddressing makes misassignment detectable, but somebody must still read the configuration
Fault detection on the pathSilent; a failure looks the same as a bad credentialStatus exchanges surface a non-answering deviceThe compensating owner shifts from wiring verification to configuration and version matching
Acknowledgement to the readerNot available on the classic wiring conventionBidirectional by specificationA Wiegand reader cannot know whether to retry; the behaviour must be designed around
Credential normalisationReader-side behaviour on unknown codesReader-side behaviour on unknown codes is unchanged in principleThe supplier still owns what the reader does with a code it does not recognise
Physical assumptionsSimple point-to-point runBalanced addressed bus with termination and distance rulesCable reuse between a legacy panel and a new bus is a design question, not a saving
Where the risk concentratesThe wire and the wiring workmanshipThe configuration and the pairing between controller and reader versionDocumentation weight shifts from a wiring schedule to a configuration record

The conclusion is not that one protocol needs less ownership. It is that the same ownership split moves. On Wiegand, the split must compensate for invisibility, which usually means a named party walks the path end to end after installation and again after any change. On OSDP, the split must compensate for visibility, which usually means a named party holds the configuration record and reconciles it against the physical device list when anything changes.

A protocol upgrade on an existing building therefore does not dissolve the boundary. It relocates the boundary, and it relocates it in a way that the people who wrote the original split were probably not watching. That is the reason to write the split in terms of obligations and evidence rather than in terms of wire counts.

The supplier's share of the ownership split

The supplier's responsibility ends at a published, dated interface definition for the specific unit, and starts at the same point. Nothing about that is contentious; what is contentious is how often the boundary is stated vaguely and then argued about later.

What a supplier can genuinely own is the behaviour of its own hardware. That includes the credential formats the reader can parse and the read distance for each format, which are model-level facts to confirm on the dated product page for the specific SKU. It includes what the reader does when it is presented with a code it does not recognise, whether that is a rejection, an ignore or a door-open request forwarded to the controller. It includes the door position contact input and its reporting behaviour, and the request-to-exit input, which signals that the escape side has been operated and is not a release command — release on the escape side is the default behaviour of the exit device itself.

It also includes what the device holds when it cannot reach the server. A lock that keeps an event memory and a lock that does not are different products from the building's point of view, and the difference is where events are lost when the link drops. Whether a given unit keeps an event buffer, what it contains, and how it is reconciled after an outage are questions to confirm on the dated product page for the specific SKU, with a written answer to the specific question, not a general capability statement.

What a supplier cannot own, and should not be asked to own, is equally important. The supplier does not own the field wiring, the controller, the panel configuration, the client's identity source, the network the traffic crosses, or the decision to deploy one reader or five hundred. When a complaint arrives saying the card does not work at a specific door, the useful question is which of those five the failure belongs to.

The evidence a supplier should be able to produce is narrow and specific. A dated product page for the SKU in question. A written response to the exact question asked, under the reference of the enquiry. A statement of what the product does not do, because absence is half of any interface definition.

CONFIRM-PER-SUPPLIER: any capability, protocol or holding claim about a specific unit should be confirmed on the dated product page for the specific SKU before it is written into a scope, a submittal or a contractual position.

The failure mode to watch for is the phrase that sounds like an answer and functions like a deflection — the product is designed to work with standard access control systems. That sentence transfers the boundary to nobody. It is precisely the sentence a supplier should replace with a dated statement of what the reader reports, in what format, and what happens when the controller does not answer.

The integrator's share of the ownership split

The integrator is the only party that sees the whole reader-to-controller path, and that is the source of the role rather than an accident of the construction sequence.

The integrator's share begins at the wiring schedule and the controller configuration and runs through commissioning. It includes the physical termination of the reader, the assignment of that reader to a door, the door record on the controller, and the entry of that door into the client's access system. It includes provisioning — bringing the device into service with its identity, its door and its owner — as a recorded operation, which is a different thing from decommissioning it, and which should be recorded as a separate operation when a device comes out of service.

The integrator also owns the commissioning evidence, which is where most projects quietly lose their ownership split. A bench test proves the reader works. A continuity test proves the wire is continuous. Only a test that presents a credential at the installed door and observes the controller's decision proves the path. Those three are different acts and they belong to different parties, which is exactly why the end-to-end test has to be named as somebody's deliverable. The Wiegand vs OSDP: The Test Plan a Buyer Should Demand covers what that test should contain; the ownership question here is narrower, namely who signs it.

The integrator also holds two other obligations that are easy to forget when the split is drawn. The first is the clock. Events with two clocks cannot be reconciled, so somebody must own time synchronisation at the door, and where the controller and the lock each keep a time of their own, the reconciliation rule belongs to the integrator rather than to whoever notices the discrepancy first.

The second is the first-line triage script. When a reader fails in service, a helpdesk technician decides what to do before anyone from the integration team arrives. That script has to say which observations are decisive and which need escalation. A technician who can check address, controller input state and event history can resolve a surprising share of calls. A technician who does not have that guidance will swap readers, which turns a configuration fault into a hardware fault and destroys the evidence.

None of this makes the integrator the owner of everything. The integrator does not decide who is entitled to open the door, does not decide the fail state of a door on a life-safety route, and cannot answer for the client's identity data. But the integrator is the party that has to be able to say, when a door is reported bad, whether the answer lies in the wire, in the configuration, in the credential or in the policy — and the split is what allows that answer to be given without a three-way phone call.

The client's share of the ownership split

The client's share is the one most often left implicit, because it consists of decisions rather than hardware. It is also the share that determines whether the other two can do their jobs.

The client owns the identity source and the permission to grant. When a person's role changes, someone in the client's organisation decides whether the credential changes, and the Single-Source Identity and Directory Sync considerations are captured differently by every estate, which is why the requirements template in this cluster has to be filled in by the client rather than by the integrator alone.

The client owns the network that carries the traffic and the rules about which segment the readers sit in, and it owns the decision to accept the consequence when that segment is unavailable. It owns the approval authority for fail-safe and fail-secure per door, and that decision belongs to the door's role and to the approval that applies at that address, not to the product.

The client also owns the retention rules and the read access for audit logs and for the event memory that a lock keeps when it cannot reach the server. Those are records of who presented what, where and when, and the questions they raise are jurisdictional. CONFIRM-WITH-LOCAL-PRIVACY-ADVISOR: the retention period, the lawful basis, the read access and any export obligations for access records must be confirmed with your own privacy adviser and your qualified local counsel for the jurisdiction the building sits in, and written into the acceptance record rather than left to the platform default.

Two client decisions are routinely deferred and then discovered to be expensive. The first is whether a critical door gets a redundant path — a spare reader, a spare controller, a documented manual release procedure with an audit trail. The second is the acceptance criteria themselves. If the acceptance criteria are "the badge opens the door", the ownership split is untestable, because that sentence can be satisfied by a door that is genuinely wired wrong but happens to have been tested once when it worked.

The Wiegand vs OSDP Requirements Document Template exists because these client-side decisions are much easier to capture before installation than to reconstruct afterwards.

The three-way seam where a split quietly becomes a gap

Most projects do not fail at the supplier boundary or the client boundary. They fail in the seam where the two meet, which is commissioning and handover.

flowchart TD
    A[Credential issued by the client identity source] --> B[Credential presented at the reader]
    B --> C[Reader reports the code it read]
    C --> D[Controller holds the door state and decides]
    D --> E[Event record written with time and source]
    E --> F[Audit log the client can read]
    C --> G[No acknowledgement returned to the reader]
    G --> H[Only a person at the door discovers the fault]

The seam is where a bench-tested reader meets a continuity-tested cable and a credential issued by somebody else's directory, and where all three are declared working individually. The classic outcome is a door that works when the integrator is present and does not work on a Tuesday morning when the integrator is elsewhere.

The cure is not more testing. It is a named path owner for a defined period. Between the moment the reader is energised and the moment the client accepts the door, one party is responsible for the end-to-end behaviour of the path, and that party is the one with the access to all three layers. In most projects that is the integrator. The arrangement should say so, with a start date, an end date, and the specific evidence that closes the period.

Two habits make the seam worse than it needs to be. The first is declaring the door complete on the basis of a single successful presentation, which tests the credential rather than the path. The second is treating the handover as the moment the ownership ends rather than the moment it transfers, so the acceptance signature is required before anything has been demonstrated under the conditions in which the door will actually run.

The seam also has a documentation dimension that tends to be handled by whoever is most available at the time, which is rarely the same party in both directions. The requirements baseline belongs at the start; the as-built configuration belongs at the end; the commissioning record belongs in between and should show the path being proven in both directions, with the person who presented the credential and the person who observed the decision.

Putting the split into the scope document in words that survive a tender

An ownership split that lives in a meeting is not an ownership split. It needs to be written where it can be tested, and it needs to be written in obligations rather than in intentions.

Table 3 — decision comparison for the reader-to-controller boundary:

Ownership questionDecision sits withExecution sits withEvidence that closes itFailure if left unassigned
What code does the reader report, and what does it do with an unknown codeSupplierSupplierDated answer for the specific SKUBuyer assumes a reader behaviour that was never confirmed
Is the reader physically correct for this door and this distanceIntegratorIntegratorTerminated and tested run with a record of the runFault treated as a credential problem
Which door does this reader belong to on the controllerIntegratorIntegratorConfiguration record reconciled against the physical listDoor opens when it should not, or denies when it should not
What happens to the door when the controller cannot reach the serverClient approval, integrator specification, supplier device behaviourIntegratorWritten degraded-mode statement per doorThe offline case is improvised on the day
What is recorded, where it lives and who can read itClientIntegratorRetention and read-access scheduleRecords are assumed to exist that were never kept
Who fixes it when the door failsPath owner for the stated periodDepends on the failing layerTriage script with escalation triggersThree-way call that lasts a year

Six clauses do most of the work, and each one can be written as a sentence that names a party and an artefact. The supplier clause: the supplier confirms, on the dated product page for the specific SKU, the credential formats read, the behaviour on an unrecognised code, the door position contact and request-to-exit input behaviour, and the event memory behaviour. The integrator clause: the integrator supplies the wiring schedule, the controller configuration record, the commissioning evidence and the triage script. The client clause: the client supplies the identity source, the degraded-mode approval per door, the retention and read-access schedule, and the acceptance criteria.

Then two further clauses make the split real. The path-owner clause names one party for a defined window. The change clause says that a change to the reader, the controller configuration, the credential format or the door furniture triggers an impact statement from the party who initiates it, before the change is applied.

The Wiegand vs OSDP for Smart Lock Projects: What Buyers Must Define First teardown covers the surrounding definition work in more detail; this section is only about the lines that assign the boundary.

Language discipline matters here more than usual. Write who owns what, not who is responsible for coordination. Write what artefact closes the question, not what the outcome will be. Write what happens when the network is unavailable, not that the system will be reliable. These differences are small on the page and large at the door.

The five questions that force the split to become explicit

Every integration has to answer five questions, and each of the five, once framed as an ownership question, produces a different allocation of responsibility.

The first is what each side sends and what happens when it does not arrive. On a Wiegand path the honest answer is that the reader sends a code and receives nothing, so the absence of an answer must be covered by the path owner. On an addressed bidirectional path the exchange normally returns a reply, so the absence of a reply is a diagnosis rather than an assumption — but the party that watches for it and acts still has to be named.

The second is who owns the clock. If the controller time and the lock time drift, the event ordering in the log becomes arguable at exactly the moment someone wants to argue about it. One owner, one stated source of time, one stated tolerance, one stated reconciliation method.

The third is what is written down, where it lives and who can read it. There are at least four distinct record sets in a typical installation: the access system's own audit log, the controller's event list, the lock's event memory for the period it was offline, and the commissioning and configuration record that the integrator holds. These are not the same thing and they are not interchangeable when a question is asked about one of them.

The fourth is what happens when the network, the controller or the server is unavailable, and the answer must be per door rather than per site, because a door's role determines its fail state.

The fifth is which side owns the fix, and this is the question the other four exist to support.

The Security Questions to Ask About Wiegand vs OSDP guide goes deeper on the security-specific variants; the point here is narrower, that an unanswered question in any of the five is a sign the split was never written, not that the parties are unusually forgetful.

Who owns the door when the link, the controller or the server is unavailable

Degraded mode is the part of an integration where ownership most often evaporates, because everyone assumes the question is theoretical until the first outage.

stateDiagram-v2
    [*] --> Normal
    Normal --> Degraded: controller cannot reach the server
    Degraded --> Normal: link restored and events reconciled
    Degraded --> Manual: door released by hand under procedure
    Manual --> Normal: operator records the reason

The first question is which of the three failures has occurred, because each has a different owner. A server outage leaves the controller holding its own decisions and its own stored events. A controller outage leaves the reader reporting into nothing. A network outage leaves both, with the difference that the lock may be able to keep an event memory that the controller cannot.

The second question is what the door does. Fail-safe and fail-secure are decided by the door's role and by the approval that applies at that address, not by the product and not by the protocol. A lock that changes state when power is removed does not thereby have a fail state; it has a power-loss behaviour, and what that means for the door is an approval question. CONFIRM-WITH-LOCAL-FIRE-AUTHORITY: where a door sits on an escape route or on a fire-alarm interface, the release behaviour and its relationship to the alarm must be confirmed with the qualified local fire authority and the client's own fire strategy, and the outcome written into the per-door schedule.

The third question is who reconciles the records after an outage, which is different from who fixes the link. The party that owns the audit trail has to be able to say what happened at a door during a window in which the system was partly blind, and that means somebody owns the event memory and the reconciliation procedure.

The Wiegand vs OSDP Failure Patterns and How to Recover comparison covers the recovery mechanics for specific failure shapes; the ownership point is that the recovery procedure should already name the party who runs it, because it will not be run by the same people on a Sunday as on a Tuesday.

The offline case is the design, not the exception. Writing it while the building is quiet costs an afternoon of argument. Discovering it during an outage costs the building an hour of access and three parties' credibility.

The evidence each owner has to be able to produce

An ownership split is only meaningful if each owner can produce something that a third party can check. This is the practical test of the whole arrangement: if the reader is swapped at eight in the morning, what paper moves?

Table 4 — the evidence register for the reader-to-controller split:

OwnerEvidence heldWhy it matters at the doorWhen it is produced
SupplierDated product page for the specific SKU and a written answer to the exact questionSeparates a hardware question from a configuration questionBefore submittal approval
SupplierStatement of what the unit does not doStops an unconfirmed capability becoming a contractual assumptionBefore submittal approval
IntegratorWiring schedule and a record of the terminated runShows the physical path was considered, not assumedBefore commissioning
IntegratorController configuration record reconciled against the physical reader listShows which reader is on which input and doorAt commissioning and after every change
IntegratorCommissioning record showing a credential presented at the installed doorThe only evidence that proves the path rather than its partsAt commissioning
IntegratorFirst-line triage script and escalation triggersStops hardware being swapped on a configuration faultBefore the door goes live
ClientIdentity source definition and the permission-to-grant ownerIdentifies where a person's access actually comes fromBefore provisioning
ClientPer-door degraded-mode approvalFixes the behaviour when the system is unavailableBefore commissioning
ClientRetention and read-access scheduleMakes the record set answerableBefore commissioning
ClientAcceptance criteria and the signed acceptance against themEnds the path-owner period with a recordAt handover

Two observations follow from this table. The first is that most of the evidence is created before the first badge is presented, which means the cost of a good split is front-loaded and the benefit is back-loaded. The second is that the evidence register is largely independent of the protocol, which is why a building can mix reader types across a portfolio without changing its governance.

The Documenting Wiegand vs OSDP for Handover and Audit guide deals with the format of these records as they are assembled for audit. What matters here is that the register is agreed before commissioning rather than reconstructed from memory afterwards, when several people have left the project.

Ownership under change, re-reading and decommissioning

A split that is written once and never revisited decays quietly. Buildings change: readers are replaced, controllers are upgraded, credentials are re-issued in a new format, doors are re-hung with different furniture, tenants arrive and leave, and the person who knew the original wiring moves on.

The change that most often goes unowned is the one nobody calls a change. A reader is swapped because it looks different. A controller is replaced by a newer model during a building upgrade, and the new controller comes with a fresh configuration that nobody reconciles against the physical list. A credential format changes at the issuing desk, and the readers on the new format are added to a group that also contains old-format readers with different behaviour on an unknown code.

Table 5 — change events and the party who owes the impact statement:

Change eventParty who initiatesParty who owes the impact statementArtefact required before the change
Reader replaced or relocatedWhoever raised the workIntegratorRe-run of the end-to-end path test
Controller replaced or upgradedFacility or IT ownerIntegratorConfiguration reconciliation against the reader list
Credential format added or retiredClient issuing functionSupplier for reader behaviour, integrator for group assignmentWritten statement of behaviour on the new format
Door furniture or frame changedFacilitiesIntegratorConfirmation that the reader and the door record still correspond
Door moved to a different controller inputIntegrator during any workIntegratorUpdated configuration record
Device removed from serviceWhoever decommissionsIntegratorRecorded decommissioning, separate from provisioning
Door role or approval changedClientClient, with integrator applying itUpdated per-door degraded-mode schedule

The rule that makes this workable is simple: the party who initiates a change owes an impact statement before the change is applied. That is a small addition to a work order, and it is the mechanism that keeps the original split valid after the building has moved on.

Decommissioning deserves its own line because it is the operation most often skipped. A reader taken out of service and left in the configuration can remain a live path for a credential that was never revoked. Removal from service is a recorded act with its own evidence, and it belongs to the owner of the path, not to the technician who happened to remove the box.

A workable split on an integrator-led project, step by step

The following is an illustrative sequence for an integrator-led installation of a mixed reader estate in an office building. It contains no product claims and no pricing; it is simply the order in which the ownership questions get answered when somebody insists on answering them in that order.

The first step is the survey. During the Office Smart Lock Site Survey the integrator records, per door, the existing reader type, the cable route and length, the controller and input it lands on, the door's role and approval route, and the client's expectation of what happens when the system is down. That last column is the one most surveys omit and the one that causes the trouble later.

The second step is the interface confirmation. The supplier is asked specific questions about the specific units, and answers come back against a dated page. Questions about behaviour on an unknown code, about the event memory, about the door position contact and request-to-exit inputs. Anything the supplier has not confirmed is written into the scope as unconfirmed, and priced as such.

The third step is the split draft. Before the hardware order, a one-page boundary sheet names the supplier's items, the integrator's items, the client's items, the path owner and the acceptance criteria. This is the artefact that prevents the ownership conversation from happening at the door.

The fourth step is the configuration baseline. As each reader is installed and commissioned, the controller configuration is reconciled against the physical list, and the commissioning record captures the credential presented at the installed door and the decision observed at the controller.

The fifth step is the degraded-mode rehearsal. The client picks a window, the system is deliberately taken offline, and the doors are observed. The purpose is not to prove that the system fails well. It is to find out, before the first real outage, whose decision was never made.

The sixth step is the handover, where the path-owner period closes against written criteria and the register of evidence is handed over as a set, rather than as a folder of photographs.

For estates that repeat this work across buildings, the portfolio framing in Wiegand vs OSDP for a Multifamily or Office Portfolio and the property-specific framing in Wiegand vs OSDP for a Hotel Property both arrive at the same conclusion from different directions: the boundary that is written once and reused is the boundary that survives staff turnover.

What to settle before anything is ordered

Five items belong on the table before the order, and each of them is a question rather than a preference.

First, the credential path. What is presented at the reader, what code reaches the controller, and what happens when the reader meets a code it does not recognise. The supplier owns the answer to the last part.

Second, the addressing and configuration question. Whether the path carries device identity, and if it does, how the address in the controller is reconciled against the physical reader list. This is the point where a switch between reader families has a cost that is not the switch.

Third, the degraded mode, per door, with the approval route named.

Fourth, the record sets, with their retention and their read access, and the party that owns each one.

Fifth, the path owner and the acceptance criteria.

The effort that each of these consumes is worth reading separately from the hardware, because effort is where the differences actually land. The Wiegand vs OSDP Cost and Effort Drivers Without Quoting Prices comparison sets out where the work goes without reducing it to a number, and it makes the same point this article does in a different place: the protocol changes the shape of the work, not whether the work exists.

Where the split is recorded so it survives the handover

The last act of the ownership split is to make it findable. A boundary sheet that lives in one person's inbox is not a boundary, it is a rumour.

The record set should be small, dated and versioned. One document states the split in obligations, with the parties named. One holds the as-built configuration. One holds the commissioning and acceptance evidence. One holds the triage script and the escalation contacts, with a review date. Where the estate has legacy doors and mixed reader types, the retrofit survey referenced in Smart Lock Retrofit Site Survey is the point at which the differences between doors should be captured rather than discovered.

Two habits keep it alive. The first is that every change is logged against the configuration record, so the document describes the building as it is rather than as it was commissioned. The second is that the acceptance criteria are written as observable states rather than intentions, so that a signature means the same thing to all three parties a year later.

And the legal position stays where it belongs. This is an engineering and commercial article, not regulatory advice. Egress, fire-interface, privacy and jurisdiction-specific questions are written here as the question to put to your qualified local authority and your own counsel, and the answer should come back in writing, per building and per door. CONFIRM-WITH-LOCAL-AUTHORITY: which authority signs off on the release behaviour and the record handling at your address, and what evidence they expect, must be confirmed with the qualified local authority rather than inferred from a general description of the system.

It is worth pairing this with the smart lock quotation comparison. That is the same question wiegand versus osdp ownership split has to answer here.If your project has a reader-to-controller boundary that has never been written down, send us the one-page scope note and we will mark up which line of it belongs to the supplier, which belongs to the integrator and which belongs to the client, and which line currently belongs to nobody at all. That marking-up is the cheapest part of the whole job, and it is the part that is hardest to do after the first door fails.

When the file is ready, request a quote against the written requirement.

Troli Belian Anda
Pengalaman anda di laman ini akan Improved dengan membenarkan kuki Dasar Cookie
Chat on WhatsApp