One More Lens by Timothy J. Harland Same Picture, Different Lens · Book Five Piece 3 of 4 — Chapters Five–Six From the Author’s Portal · showyourmap.com Free for weary travelers. E UNUM PLURIBUM. ------------------------------------------------------------ CHAPTER FIVE The Requirement By noon, they had reduced forty-three pages of standards to eleven questions. Jonah looked at the list. “That feels either efficient or dangerous.” “Both can be true.” Lena moved the first requirement to the top of the screen. Regional nodes shall maintain continuity during loss of central coordination. Jonah read it. “We’ve seen that one before.” “Yes.” “What are we asking now?” Lena typed beneath it. What outcome is this requirement intended to protect? Jonah nodded. “Better.” The next requirement concerned synchronization after reconnection. Then response timing. Then reporting. Then fallback communications. The standards had been written as though each requirement existed independently. They did not. Some protected continuity. Some protected visibility. Some protected interoperability. Some protected governance. Some appeared to protect nothing Lena could identify beyond consistency with previous versions. Jonah pointed at one. “Why does that have to be centralized?” “I don’t know.” “Technical reason?” “None stated.” “Safety?” “None stated.” “Regulatory?” “Not in the supporting documentation.” “Historical?” Lena opened the change record. There it was. The requirement had originated in an early procurement package because the first vendor platform used central authorization for configuration changes. Jonah leaned closer. “So the original system needed it.” “Yes.” “Current systems?” “Three do. Four don’t.” “And the requirement stayed.” “Yes.” Jonah smiled. “Legacy fossil.” Lena looked at him. “That is not a classification.” “It should be.” She typed: Inherited implementation constraint - current necessity unverified. Jonah nodded approvingly. “That’s less fun.” “It is more useful.” By 13:20 they had created five categories. Outcome requirement. The thing the system actually had to accomplish. Interface requirement. The boundary necessary for different systems to work together. Safety requirement. The constraint whose failure increased unacceptable risk. Governance requirement. The condition necessary for authority, accountability, or oversight. And finally: Inherited convention. Jonah stared at the fifth category. “You actually kept it.” “It is accurate.” “I’m proud of you.” “I am not interested in your approval.” “Still proud.” They began sorting. The first twenty were easy. Then the arguments started. Jonah moved one requirement toward governance. Lena stopped him. “That is interface.” “It tells people who gets notified.” “It specifies data routing.” “To a person.” “Through an interface.” “Which exists because someone needs to know.” “That does not make the requirement governance.” Jonah leaned back. “So if nobody needed the information, would the interface still exist?” “No.” “Then the human purpose matters.” “The purpose and implementation are separate questions.” “Exactly.” Lena stopped. Jonah smiled. “That was suspiciously easy.” She moved the requirement halfway between the two categories. “Dual purpose.” “Look at us.” “This is not a bonding exercise.” “Everything becomes a bonding exercise if you lower your standards.” At 14:03 they reached a requirement neither could classify. All regional nodes shall maintain continuous readiness to receive central coordination directives. Lena read the supporting notes. Jonah read them too. Nothing. No safety justification. No interoperability justification. No stated operational outcome. Just the sentence. “Where did it come from?” Jonah asked. Lena searched the revision history. Version one did not contain it. Version two did. No explanatory note. “Meeting minutes?” She searched. Found them. A coordination working group had added the requirement after a discussion about inconsistent regional response times. Jonah read the minutes aloud. “Participants expressed concern that regional operators may not maintain adequate attention to centralized coordination channels during routine local operation.” He stopped. “That’s not a technical requirement.” “No.” “That’s a trust problem.” “Possibly.” He looked at her. She sighed. “Fine. It originated as a trust concern.” “Thank you.” Lena continued reading. The proposed mitigation had originally been procedural. Periodic acknowledgment. Mandatory monitoring intervals. Escalation if contact was missed. None had survived. Instead, the final standard required continuous readiness. Jonah frowned. “So a human concern became a technical requirement.” “Yes.” “And then nobody remembers why.” “Apparently.” He tapped the screen. “Does that matter?” Lena did not answer immediately. The requirement looked harmless. Continuous readiness sounded prudent. Responsible. Professional. Exactly the sort of sentence no committee wanted to object to. She opened implementation records. Northern sites had interpreted it as requiring persistent connection and active central monitoring. Southern sites treated it as ensuring the communication channel remained technically available. The western site maintained the channel but did not assign continuous operator attention to it. All three claimed compliance. Jonah watched her face. “This one matters.” “It might.” “Lena.” “It explains one divergence.” “Which is more than we had five minutes ago.” She nodded. “Fair.” He pointed again. “The sentence isn’t precise enough to constrain implementation.” “No.” “But it sounds precise enough that nobody asks what it means.” “Yes.” Jonah leaned back. “There’s our problem.” “Part of it.” “You’re impossible.” “Accurate.” They called the current standards chair. His name was Adrian Bell. He had held the role for six years and had inherited most of the requirements rather than authored them. Lena shared the sentence. Bell read it. “What’s wrong with it?” “We don’t know that anything is wrong with it,” Lena said. Jonah muted himself. Bell continued. “It seems straightforward.” “What does continuous readiness mean operationally?” Bell paused. “That the regions have to be ready to receive directives.” “Yes.” “How?” Another pause. “Communications available.” “Only communications?” “Well, no. Someone has to receive them.” “Continuously?” “That’s what it says.” Jonah unmuted. “Does someone have to be actively watching?” Bell looked at him. “I suppose.” “Every second?” “No.” “How often?” Bell frowned. “That would be an implementation detail.” Lena nodded. “Exactly.” Bell leaned back. “I’m missing something.” Lena showed him the three interpretations. Persistent central monitoring. Technical channel availability. Available channel with local operator discretion. Bell read them. “All of those could satisfy the sentence.” “Yes.” He looked uncomfortable. “So the standard is ambiguous.” “Potentially.” Jonah looked at Lena. She ignored him. Bell asked, “Do they all perform?” “Yes.” “Then maybe the flexibility is intentional.” Lena nodded. “That is possible.” Bell relaxed slightly. Jonah stepped in. “Did you intend it?” Bell stopped relaxing. “I didn’t write it.” “Did your committee ever define the allowed range?” “No.” “So the flexibility may be useful.” “Yes.” “But accidental.” Bell rubbed his forehead. “Yes.” Lena wrote that down. After the call, Jonah stood by the window. It was the first office they had used with one. Rain moved across the city in narrow gray bands. Lena kept sorting requirements. “Do you think ambiguity is the problem?” Jonah asked. “No.” He turned. “That was fast.” “Ambiguity can be intentional.” “Okay.” “It can preserve implementation freedom.” “So what’s the problem?” Lena looked at the requirement list. “Unbounded ambiguity.” Jonah waited. “If a requirement allows multiple implementations but does not identify the outcome or boundary that must remain invariant, then we cannot distinguish valid flexibility from accidental divergence.” Jonah nodded slowly. “So the problem isn’t that everyone interprets it differently.” “No.” “It’s that nobody knows which differences matter.” “Yes.” He returned to the table. “That feels bigger.” “It is.” The next morning they met with three regional teams simultaneously. Not leadership. Operators. Technicians. Supervisors. People who actually used the system. Lena put one requirement on the screen. Maintain continuity during loss of central coordination. “What does this protect?” she asked. The northern supervisor answered first. “Service.” A southern technician said, “Local control.” The western operator said, “Both.” Lena wrote each answer. “Which one is required?” Silence. Jonah watched. Eventually the southern technician spoke. “Does it matter if the outcome is the same?” Lena looked at him. “That is what we’re trying to determine.” They moved to the next requirement. Synchronization after reconnection. “What does this protect?” “Data integrity.” “Shared visibility.” “Scheduling accuracy.” All valid. Not identical. By the seventh requirement, the participants had stopped defending their regional interpretation. They had started comparing them. One operator said, “We do that completely differently.” Another replied, “Yeah, but we’re protecting the same thing.” Someone else said, “We’re not.” The room changed. Lena noticed first. Jonah felt it first. The meeting was no longer about compliance. It was about purpose. By the ninth requirement, the participants were identifying their own inherited conventions. “We only do that because the previous platform required it.” “That was a workaround from before the upgrade.” “That procedure exists because our regulator used to require manual acknowledgment.” “Used to?” “Seven years ago.” Jonah looked at Lena. She did not look back. She was writing too quickly. At lunch, Jonah carried two trays to the table. Lena was still reading notes. “You’re doing it again.” “What?” “Working through food.” “I am eating.” “You have not touched anything.” “I intend to.” “That is not the same thing.” She put the tablet down. Jonah sat. “What did you see?” “Several requirements are preserving obsolete constraints.” “Besides that.” She thought. “The regional teams identified them faster than we did.” “Why?” “They know which procedures create friction.” “That’s one reason.” Lena waited. Jonah continued. “They also weren’t trying to defend the architecture once we stopped asking whether they were compliant.” She nodded. “The question changed.” “Yeah.” She looked across the cafeteria. Groups from several regions were sitting together. People who had spent months explaining why their implementation was correct were now comparing why it existed. Jonah followed her gaze. “They’re building a shared map.” “Not shared.” “No?” “They’re comparing maps.” He smiled. “Right.” That afternoon Lena revised the framework. Not forty-three requirements. Eleven questions. Each requirement now had to answer: What outcome does this protect? What failure does it prevent? What boundary must remain invariant? What implementation freedom is acceptable? What evidence demonstrates success? What assumptions existed when the requirement was written? Do those assumptions still hold? What dependencies does the requirement create? Who experiences the cost of that dependency? What happens when the environment changes? And finally: When should this requirement be reconsidered? Jonah read the last line. “You put revision into the standard.” “I put reconsideration into the review.” “Close enough.” “It is not.” He smiled. “Of course.” At 16:10 the committee reconvened. Lena presented the framework. No conclusions. No recommendation to centralize. No recommendation to decentralize. No attempt to declare one regional architecture correct. The chair listened. Then asked the question Lena had expected. “What does this mean for the implementation problem?” “We’re not ready to answer that,” she said. The technical advisor visibly resisted reacting. Jonah watched him lose. “After three days?” “Yes.” “What exactly have we accomplished?” Lena displayed the original question. Why are regional implementations diverging from the shared architecture? Then she crossed out the last three words. The room became quiet. The new question read: Why are regional implementations diverging? The advisor frowned. “That is broader.” “Yes.” “Less precise.” “No.” He looked at her. Lena continued. “The original question assumed a shared architecture existed in a form that all regions were expected to reproduce.” “And you no longer believe that?” “We no longer have sufficient evidence to treat it as established.” The chair looked at the screen. “So we were investigating deviation from something we had not defined.” “Yes.” Jonah leaned forward. “And treating difference as evidence of failure before determining whether the differences violated anything that mattered.” The standards chair looked down. The governance representative asked, “What do you need next?” Lena changed the slide. Three columns appeared. Required commonality. Permitted variation. Unknown. Forty-three requirements began populating the columns. The third was the largest. The chair stared at it. “That is uncomfortable.” “Yes,” Lena said. Jonah smiled slightly. “That’s probably useful.” No one disagreed. They left the meeting just before sunset. The rain had stopped. Jonah walked beside Lena toward the hotel. “So.” “So?” “We have more unknowns than when we started.” “Yes.” “You’re happy.” “No.” “You are.” “I am satisfied with the direction.” “That is happy with paperwork.” Lena kept walking. Jonah glanced at her. “Does it bother you?” “What?” “That the better model contains more uncertainty.” “No.” He waited. She finally looked at him. “The uncertainty was already there.” Jonah nodded. “We just stopped hiding it.” “Yes.” They crossed the street. Behind them, forty-three requirements still existed. No standard had been removed. No architecture had been changed. No region had been told to operate differently. The system had not become simpler. It had become more accurately incomplete. For the first time, Lena considered that progress. She did not say it aloud. Jonah did not need her to. CHAPTER SIX The Boundary The first disagreement that mattered arrived on Thursday morning. Not between regions. Between Lena and Jonah. It began with a line item. A governance requirement required all regional exceptions to be reported to the central office within twelve hours. Lena classified it as required commonality. Jonah moved it to unknown. She moved it back. He looked at her. “What?” “It belongs there.” “No, it doesn’t.” “It establishes reporting consistency.” “That isn’t the same thing as necessity.” “It creates system visibility.” “Maybe.” Lena stopped. “You’re using my word.” “I’ve been exposed.” She ignored that. “Without common reporting, central governance cannot reliably identify regional deviation.” Jonah folded his arms. “Why twelve hours?” Lena looked at the requirement. “That’s implementation detail.” “Then the requirement isn’t the line you just defended.” She said nothing. Jonah pointed. “You defended visibility.” “Yes.” “This says twelve hours.” “Yes.” “So which part is actually required?” Lena read the sentence again. The requirement bundled outcome and implementation into one line. She knew that. She also knew she had just treated them as the same thing. Jonah waited. “Visibility is required,” she said. “And twelve hours?” “Unverified.” He moved the requirement halfway between the columns. Lena let him. “Progress,” he said. “Do not celebrate.” “I’m marking the occasion internally.” The argument should have ended there. It didn’t. They found the same problem again. And again. Mandatory reporting format. Central approval threshold. Escalation sequence. Regional audit frequency. At first the disagreements were small. Then they became structural. Jonah kept asking who experienced the cost. Lena kept asking what failure the rule prevented. Sometimes they landed in the same place. Sometimes they didn’t. At 10:14 Lena classified a central approval rule as governance. Jonah left it in unknown. She looked at him. “You disagree?” “Yes.” “Why?” “Because everyone keeps saying the rule protects accountability.” “It does.” “No. They say it does.” Lena waited. Jonah slid three interview notes across the table. Regional managers described the rule as accountability. Operators described it as delay. Central staff described it as visibility. A former program manager described it as a response to a procurement scandal nine years earlier. Lena read the notes. “What was the failure?” “One region committed funds outside the agreed scope.” “So governance.” “Then why does the current rule apply to operational changes with no budget impact?” Lena looked again. Jonah leaned back. “The original boundary may have expanded.” “Possibly.” He smiled. “You’re back.” She ignored him. By noon, they had a new category. Not because they wanted one. Because the existing categories could no longer carry the evidence cleanly. Boundary drift. Requirements whose original purpose was valid but whose scope had expanded beyond the failure they were designed to prevent. Lena typed the label. Jonah looked pleased. “That one’s good.” “It is descriptive.” “That’s what makes it good.” They began sorting again. The twelve-hour reporting rule went there. So did central approval for low-risk operational changes. So did a mandatory escalation chain that had originally applied only during cross-region incidents. The pattern was not dramatic. Nothing had been hijacked. No one had quietly seized control. No one had even intended to broaden the rules. The boundary had simply moved one administrative decision at a time. Jonah read through the change history. “Every expansion makes sense in isolation.” “Yes.” “That’s annoying.” “Why?” “Because I’d prefer at least one obviously bad decision.” “That would make the analysis easier.” “And the story better.” Lena looked at him. He grinned. “Kidding.” “Mostly?” “Mostly.” At 13:00 they met with the governance team. Five people. Two lawyers. One policy analyst. One regional liaison. One program director. Lena displayed the new category. Boundary drift. The program director frowned immediately. “That sounds accusatory.” “It is not.” “It sounds like something moved improperly.” “Not necessarily.” The lawyer on the left leaned forward. “Then define it.” Lena did. “A requirement whose current scope exceeds the failure boundary established by its original rationale.” The lawyer looked at the screen. “So the rule may still be valid.” “Yes.” “But broader than necessary.” “Potentially.” The program director crossed his arms. “Who decides what is necessary?” Jonah answered. “That’s what we’re trying to make visible.” The director looked at him. “Meaning?” “You have rules justified by outcomes. Some of the outcomes are still relevant. Some of the rules have grown beyond them. If we don’t separate the two, we can’t tell whether we’re protecting the system or protecting the habit.” The room went quiet. One of the lawyers looked down at her notes. The regional liaison spoke. “That’s fair.” The director looked at her. She continued. “We’ve had teams request waivers for years because local changes technically trigger central review even when the risk is negligible.” “Why not revise the policy?” The liaison gave a small shrug. “Because waivers work.” Lena looked at Jonah. There it was. A local workaround preserving a flawed boundary because the workaround reduced the pressure to fix it. Lena asked, “How many waivers?” The liaison opened her laptop. “Last year?” “Yes.” “One hundred and eighty-seven.” Jonah blinked. “For one rule?” “For that rule.” The director looked uncomfortable. Lena did not react. “How many were denied?” The liaison searched. “Six.” Jonah leaned back. “So the exception process is functioning as the normal process.” No one answered. The meeting lasted two hours. By the end, nobody had agreed to remove a single rule. That was fine. They had agreed to review the relationship between each rule and the failure it was meant to prevent. That was enough. For now. Back in the office, Jonah closed the door. “One hundred and eighty-seven.” “Yes.” “Six denied.” “Yes.” “That’s not exception handling.” “No.” “That’s architecture.” Lena looked at him. He pointed toward the governance chart. “The system says central approval is the default.” “Yes.” “The operational reality says approval is almost always granted.” “Yes.” “So the real operating model is local discretion with ceremonial central confirmation.” Lena paused. “That is not how I would phrase it.” “But?” “It is directionally accurate.” Jonah smiled. “I’ll take it.” Lena opened the waiver data. Average processing time: sixteen hours. Median: eleven. Longest: nine days. She filtered for operational impact. Thirty-four changes had been delayed long enough to affect scheduling. Five had created temporary capacity reductions. None had caused a safety incident. Jonah read over her shoulder. “There’s the cost.” “Yes.” “What’s the benefit?” “Governance visibility.” “Could they get that without approval?” “Possibly.” “Reporting?” “Yes.” “Audit?” “Yes.” “Sampling?” “Yes.” Jonah looked at her. “So?” Lena stared at the data. “We need to separate oversight from permission.” Jonah nodded. “Exactly.” That afternoon they split again. Not geographically. Methodologically. Lena stayed with governance data. Jonah went back to the regional teams. They agreed not to compare until evening. This time the separation was easier. Neither needed to negotiate it. Lena examined decisions. Who approved what. How long it took. What information changed during review. How often approval altered the original proposal. Most did not. Jonah examined behavior. Who waited. Who worked around the rule. Who delayed raising issues because they did not want to trigger central review. Who quietly bundled several small changes together to reduce administrative overhead. At 17:32 they compared. Lena started. “Eighty-one percent of approvals result in no change.” Jonah nodded. “Teams are delaying low-risk changes to avoid triggering review during peak periods.” “Average delay?” “Depends on region. Two to fourteen hours.” “Documented?” “No.” Lena looked at him. “Then how confident are you?” “Moderate.” “Basis?” “Five interviews across three sites. Same behavior, different wording.” She nodded. “Acceptable.” Jonah smiled. “Your turn.” “Thirty-four scheduling impacts last year.” “From review delay?” “Associated with review delay.” He nodded. “Good distinction.” “Thank you.” “What changed after approval?” “Usually nothing.” “So the approval adds delay without adding information.” “Not always.” “How often?” “Approximately nineteen percent of cases produced a revision, request for clarification, or scope change.” Jonah paused. “That’s not trivial.” “No.” “So removing approval entirely would discard useful review in one out of five cases.” “Yes.” He leaned back. “Annoying.” “Reality remains inconsiderate.” They sat quietly. This was the kind of point where a simpler investigation would have produced a recommendation. Remove the approval requirement. Or preserve it. Centralize. Decentralize. Reduce friction. Increase oversight. Neither answer survived the data. Jonah looked at the board. “Maybe we’re asking the wrong binary.” Lena looked up. “What binary?” “Approval or no approval.” She considered it. He continued. “The review has value sometimes.” “Yes.” “The delay has cost often.” “Yes.” “So the question is when the review adds enough value to justify the delay.” Lena turned back to the data. Risk class. Change type. Cross-region effect. Budget impact. Safety impact. Reversibility. She began sorting. Jonah watched. “What?” “You already have it.” “Not yet.” “You have the shape.” Lena created three columns. Notify. Review. Approve. Jonah smiled. “There.” She pointed at him without looking away. “Do not.” “I didn’t say anything.” “You were about to.” “I was going to say that this is why we get along.” “We are not discussing that.” “Another time.” They began mapping decisions. Low-risk, reversible, local changes: Notify. Cross-region changes with moderate impact: Review. High-risk, irreversible, safety-sensitive, or significant governance changes: Approve. Jonah looked at the result. “Same oversight.” “Different intervention threshold.” “Less delay.” “Potentially.” “More local autonomy.” “Where risk permits.” “Still central visibility.” “Yes.” He nodded. “That’s cleaner.” “Maybe.” He smiled. “There it is.” The next morning they presented the model. The program director studied the three columns. “Who sets the thresholds?” “The governance group,” Lena said. “Then this is still centralized.” “For boundary definition.” “And execution?” “Local where the boundary permits.” The director looked at Jonah. “And you agree?” Jonah shook his head slightly. “I agree with the structure.” The director frowned. “That sounds like a distinction.” “It is.” Lena looked at him. Jonah continued. “The thresholds will be wrong at first.” The room shifted. The program director looked offended. The regional liaison smiled. Jonah went on. “You don’t have enough evidence to define them perfectly.” Lena added, “So the thresholds should be reviewed against outcomes.” The lawyer on the left said, “How often?” “We don’t know yet.” The director looked from one to the other. “You’re proposing a governance system you admit will be wrong.” “No,” Lena said. “We’re proposing one that expects revision.” The room went quiet. The liaison looked at the model again. “That may be the first realistic thing anyone has said about this process.” The director sighed. “Comforting.” Jonah smiled. “Wasn’t meant to be.” The committee did not approve the framework. They authorized a pilot. Three regions. Ninety days. Defined risk classes. Tracked approval changes. Tracked delays. Tracked incidents. Tracked reversals. Tracked complaints. Tracked nothing that could not be tied to a question they actually wanted answered. Lena insisted on that last part. At 15:41 they left the building. Jonah carried his jacket over one shoulder. “You realize we just created another model.” “Yes.” “That will eventually drift.” “Probably.” “And someone will have to revisit it.” “Yes.” He looked at her. “You seem surprisingly comfortable with that.” Lena stopped walking. A year earlier, she might not have been. She looked back at the building. The standards. The committees. The diagrams. The inherited requirements. Every one of them had once been an answer to something. Then the environment changed. Or the interpretation changed. Or the reason was forgotten. The problem had never been that the systems were imperfect. The problem was pretending the boundaries would remain correct forever. “We’re not designing permanence,” she said. Jonah waited. “We’re defining a current boundary and making revision part of the boundary.” He nodded. “Approximal.” Lena looked at him. He smiled. She almost objected. Almost. “Close enough.” Jonah stopped. “That might be the nicest thing you’ve ever said to me.” “It wasn’t about you.” “I’m still keeping it.” They continued walking. Behind them, the pilot had already entered scheduling. No grand redesign. No declaration that the old system had failed. Just a narrower question. Where should authority live under these conditions? For now, they had an answer. Not the answer. That difference was becoming easier to live with.