Thursday, September 11, 2025

How Can You Find the Kirketrator of Trump's Friend?

Officer Sean Collier - one of the victims related to the Boston Marathon bombing - during the manhunt for the perps

Stand up unified command. Establish incident command with campus police + city PD + sheriff + state police + FBI/ATF as needed; assign leads for scene, canvass, digital, ballistics, and victim/family liaison.
Make the scene safe & freeze it. Triage victims, neutralize hazards, and immediately set hot/warm/cold perimeters. Log every entry/exit; no one walks unescorted. (Preservation is critical to avoid loss/contamination.) National Institute of Justice+1
Prevent evidence loss right away. Cover weather-exposed areas; shield footprints and bloodstains; stop cleaning crews; halt sprinkler cycles; stop trash removal in the footprint and adjacent buildings. National Institute of Justice

Immediate digital preservation holds. Issue written “preserve now” requests for all CCTV/NVR systems on campus and neighboring businesses, plus body-worn cameras, dash cams, and stage/AV feeds. Request original exports with metadata, logs, and players; don’t accept re-encoded clips. NIST Publications+2Regional Computer Forensics Laboratory+2

360° scene documentation. Before moving anything: wide-to-tight stills, video walk-through, overheads (ladder lift / fixed-wing / UAS if authorized), a measured sketch, and 3D scan if available (e.g., total station/FARO). National Institute of Justice

Trajectory & impact mapping. Identify entry/exit, potential muzzle locations, impact spall, ricochet, and backstop. Use rods/lasers and note vertical/horizontal angles, with photographs and measurements for reconstruction. National Institute of Justice

Locate and collect ballistics. Grid search for cartridge case(s), bullet, and fragments along likely flight paths and impact sites. Photograph in place, mark, and package correctly (separate, rigid containers; no metal-to-metal). Maintain airtight chain of custody. National Institute of Justice

Victim & medical coordination. Coordinate with the ME/coroner for full autopsy, radiography, wound path documentation, and recovery of any projectiles/fragments; preserve the victim’s clothing for trace/soot/stippling analysis. National Institute of Justice

GSR/trace where lawful and appropriate. Consider prompt collection of GSR/trace from persons/areas consistent with policy and legal standards; prioritize swabbing of likely firing positions (railings, doorknobs, roof access points). (Follow your jurisdiction’s scientific and legal guidance.) National Institute of Justice

Identify, separate, and interview witnesses. Rapidly identify closest observers (VIP detail, stage crew, media camera ops, rally marshals). Obtain original phone videos/photos (not just shares) and contact info; conduct cognitive-style interviews; avoid co-mingling witnesses. National Institute of Justice

Video canvass & triage. Map nearby cameras (campus, streets, dorms, stores, parking, transit). Pull native exports with hashes/signatures and clock offsets, plus NVR event logs. Build a synchronized multi-camera timeline for the minute before/after the shot and the ingress/egress windows. NIST Publications+1

Public evidence portal. Stand up an upload link/QR for attendee media; publicize via PIO without revealing sensitive forensics. De-duplicate, hash, and index submissions; tag by vantage point and timestamp.

Acoustic/technology sources. If available, retrieve acoustic gunshot detection, radio logs, emergency call timing, and stage mic recordings to refine shot timing and potential muzzle vector.

Develop suspect path(s). From video + witness accounts, extract stills of any person displaying pre-assault indicators (scouting, rooftop access, concealment, post-shot flight). Note clothing, bag, gait, vehicles, direction of travel. Distribute BOLOs to patrols and neighboring jurisdictions. Regional Computer Forensics Laboratory

Vehicles & movement. Check parking cameras, campus gate logs, rideshare/taxi pickup zones, and traffic cams; where authorized, query ALPR in/out corridors around the event time to identify candidate vehicles. (Follow local policies and legal standards.)

Ballistic intelligence (NIBIN). Submit cartridge cases (and test-fires from any recovered gun) to NIBIN for potential correlations to other shootings; fast-track any “NIBIN lead” for investigative follow-up and seek confirmation by a firearms examiner. ATF+2ATF+2

Firearm tracing (ATF eTrace). If a firearm is recovered, initiate eTrace through the National Tracing Center to identify first retail purchaser and trafficking patterns; integrate with interviews, pawn/range/rental records, and FFL theft reports. ATF+2ATF+2

Forensic comparisons. Have a qualified examiner compare bullets/casings to any seized firearm(s) under a comparison microscope; document class/individual characteristics and limitations; request muzzle-to-target distance estimations if applicable. National Institute of Justice

Access control & rooftop audit. Pull building access logs (card swipes, keys, maintenance tickets) and CCTV for rooftops, garages, dorm windows, and vantage points; interview facilities staff about abnormal access or propped doors.

Digital investigations (with warrants as required). Work with prosecutors to seek narrowly tailored legal process for:
• CSLI or geofence data to place a known suspect device at the scene (must comply with Carpenter v. United States and local precedent).
• Account/content records for threat posts or planning communications.
• Cloud/device extractions from arrested suspects. Oyez+2Justia Law+2

Person-of-interest development. Cross-reference prior threats to the VIP/campus, restraining orders, extremist grievances, stalking reports, weapons arrests, recent rooftop trespass calls, and students/staff with relevant histories.

Photo arrays & identifications. Use double-blind, sequential photo arrays; document admonitions; avoid suggestive procedures; record confidence statements contemporaneously.

Warrants/searches & arrests. When probable cause exists, execute searches on residences/vehicles/lockers for the firearm, matching ammo, distinctive clothing, range receipts, maps, access keys, and digital media; preserve logs and video of the search.

Laboratory queueing. Prioritize testing with clear questions (e.g., “Does bullet A match firearm B?” “Is this the same gun as prior incident C from NIBIN lead?”). Stagger requests to avoid bottlenecks; share status with prosecutors.

Analytic timeline. Build a minute-by-minute composite: suspect approach → pre-shot behavior → shot timing/trajectory → egress route → post-incident movements, linking sources (video, witness, ALPR, digital, lab) with clock-drift corrections.

Comms discipline & PIO. Centralize public messaging; solicit tips and media but do not release technical details (e.g., exact trajectory, unconfirmed NIBIN hits) that could taint witnesses or aid evasion.

Victim/family care & notifications. Assign a family liaison; coordinate with the ME, prosecutors, and protective services; manage property return and court updates.

Interagency case conference. Regularly brief DA/USAO on probable charges, discovery posture (Brady/Giglio), digital returns, and lab timelines; identify remaining gaps and assign owners.

Quality control & legal review. Audit chain-of-custody, warrant scope, minimization steps, retention limits, and disclosure readiness; fix documentation gaps early.

Operational follow-ons. If a network is suspected, coordinate threat mitigation (additional security for related events, protective details, campus alerts), and consider parallel financial/communications analysis.

After-action. Post-case, review CCTV coverage, rooftop controls, access policies, radio interoperability, and public-upload workflows; update SOPs and training.

Key references for best practices (selection): NIJ/FBI crime-scene guidance on scene security/documentation and evidence handling; NIST/FBI guidance on native CCTV exports with hashes/metadata; ATF’s NIBIN for ballistic links and eTrace for firearm tracing; and Carpenter v. United States (2018) on warrant requirements for historical cell-site data. Justia Law+9National Institute of Justice+9Federal Bureau of Investigation+9

What about search engine logs - the shooter knew this person was going to be at this college campus. What about pre-crime preparation - like getting ammo, etc - so looking through purchase records?

1. Digital footprint (searches, online research)

Search engine/server logs: With proper legal authority (typically a search warrant or subpoena), investigators can request logs from Google, Bing, etc., for queries such as the VIP’s campus visit, rally date/time, directions, or seating maps.
Web browsing history: For a suspect’s seized devices, digital forensics (via write-blocked imaging) can recover deleted or cached search terms, visited pages, and metadata.
Social media & forums: Monitoring posts, likes, or group memberships that indicate planning, grievances, or interest in attack tactics.

Ammunition sales: Depending on jurisdiction, ammo may require ID; retailers may retain transaction logs or surveillance. Investigators would check recent purchases against suspect’s ID, debit/credit card, or surveillance video.
Firearm purchase/transfer records: In the U.S., every licensed dealer sale goes through Form 4473 + NICS. Trace requests via ATF’s eTrace reveal first retail purchaser, which helps link a gun to the suspect.
Explosives/precursor materials (if relevant): Purchases of black powder, smokeless powder, or large chemical orders can be cross-checked through vendor records.

3. Financial transactions

Bank/credit card records: Look for unusual purchases (range time, high-capacity magazines, optics, tactical clothing, travel expenses for reconnaissance).
Cash withdrawals: Sometimes suspects switch to cash — but ATM video can tie person + time.

4. Logistics & planning behavior

Travel records: Airline, bus, train, or car rental records to reach the campus area.
Hotel/Airbnb receipts: To see staging location near the rally.
Campus access prep: Checking whether suspect scouted rooftops or practiced entry into restricted areas.

5. Associates & communication

Messaging apps / email: With warrants, examine whether the suspect coordinated with anyone else, sought encouragement, or acquired gear via private sales.
Phone metadata: Call detail records may show outreach to suppliers, online sellers, or co-conspirators.

6. OSINT & tip fusion

Open-source indicators: Blog posts, manifestos, or online handles could reveal motive and pre-attack rehearsal.
Community reports: Gun range staff, store clerks, or campus peers might recall unusual purchases or conversations, which should be folded into lead triage.

👉 In short: authorities don’t stop at the bullet and cameras. They work backward from the crime to reconstruct the suspect’s intent, planning, acquisition, and rehearsals, drawing on digital forensics, purchase/financial records, and witness accounts — always under the framework of legal process (warrants, subpoenas, chain-of-custody).

Wednesday, September 10, 2025

Charlie Kirk - What We Know So Far

Apparently, we don't have an app that lets a drone hover over you and keep an eye out, and rich people don't care enough to insist on one. Do you smell opportunity? This would be really low cost too, yeah?

NY Times - Lazaro Gamio and Daniel Wood

I estimate the shooter's distance at 100 yards. For comparison, Thomas Matthew Crooks was 148 yards away when he tried to take out Trump with a DPMS Panther Arms DR-15.

TSTS distance estimate : 300 ft

Charlie must have lived it up to look wise beyond his 30 years. RIP.

You have to enable Globe view in Google maps to be able to rotate the map, FYI

I would like to build a gadget that uses a drone with camera(s) to keep an eye out for potential assassins while I am giving a speech at a rally. If it detects something suspicious, I will get an alert - maybe something similar to what the emergency alert system.

How difficult is this to do? Can you give me a roadmap to pull this off?

From the air, if one observes someone horizontal and, on a rooftop, that's suspicious. If they have their arms out, like they're aiming a gun, that's suspicious.

What can computer vision reliably recognize today, and what can we use to make this happen?

1) Define the mission & constraints

Primary goal: early warning of likely threats seen from above (e.g., person prone on a rooftop, arms extended in a firing posture).
Environment: dense crowd, variable lighting, lots of false-positive look-alikes (photographers kneeling, people on balconies, HVAC techs on roofs, etc.).
Responsibility: the system should assist trained security; it must not be your only line of defense.

Operations over people: Flying over crowds is tightly regulated. Part 107 allows “operations over people” only if the aircraft meets Category 1–4 requirements (energy limits, declared compliance, etc.) or via specific waivers—and you’ll still need airspace authorizations where applicable. Plan this first with a Part 107 certificated pilot/ops team. Federal Aviation Administration+1
Remote ID is now generally required for registered drones; compliance is enforced. Ag & Natural Resources College+3Federal Aviation Administration+3Federal Aviation Administration+3
Local law & privacy: State and city rules vary; event surveillance can trigger additional restrictions and disclosure duties. Coordinate with venue, local authorities, and counsel. The Washington Post+3Facit Data Systems+3Brookings+3
Ops reality: Many public-safety teams use drones, but policies, geofencing behavior, and public expectations keep shifting—plan for scrutiny and transparency. The Washington Post+1

3) What CV can (and can’t) do today

Firearm/long-gun detection: Modern one-stage detectors (e.g., YOLO-family) can flag weapons in clear, near-field views with decent precision, but accuracy drops at long range, low resolution, odd angles, occlusions, and harsh lighting. Peer-reviewed work continues to report promising lab/curated results, but “in the wild,” real-time performance often degrades. Expect both false negatives and false positives. arXiv+2Nature+2
Pose cues: Combining person detection + pose estimation (arms extended toward a line-of-sight) can cut false alarms compared to weapon-only detection, but remains fragile when the subject is small in frame or partially hidden. Use pose as a signal, not sole proof. arXiv+2ScienceDirect+2
Prone/rooftop heuristics: Recognizing “prone on rooftop” is usually a scene understanding problem: detect people, segment rooftops/edges, infer posture. It’s feasible, but you need careful thresholds and human review.

4) System architecture (safe, high-level)

A. Airframe & optics

Multirotor with Category 1–4 compliance (as applicable), low mass (safer over people), redundant link, and a gimbal.
Optics: stabilized zoom (20–40× equivalent) for distant roofs; consider dual-sensor (RGB + thermal) to help at dusk/night and to pick out prone bodies against hot rooftops. (Thermal reduces but doesn’t eliminate false alarms.)

B. Compute & link

Edge compute (e.g., compact GPU module on the drone or tethered ground unit) to run CV at ≥20–30 FPS on 1080p crops.
Downlink: robust, low-latency digital video to a ground safety cell (monitors + human operator) who confirm/triage alerts.

C. Perception pipeline

Stage 1: Region-of-interest (ROI) focus (rooftops, balconies, elevated perimeters) via geofenced polygons on a pre-surveyed map.
Stage 2: Person detection (find humans).
Stage 3: Posture/pose cues (prone, arms extended).
Stage 4: Object cues (long-gun/handgun–like shapes).
Stage 5: Multi-frame confirmation (require consistent evidence over N frames + small camera re-angle to reduce single-frame artifacts).
Stage 6: Human-in-the-loop validation before alarming the principal.

D. Alerting

Integrate with a vetted haptics pattern on a secure device (distinct long buzz + 3 beeps is fine), and a separate channel to the protective detail (radio/dispatch). The alert should include a thumbnail + map bearing for immediate action.

5) Reliability engineering & UX

Pre-event site survey: map vantage points, list legitimate rooftop workers/positions, and mark friendly cameras/press stands to reduce false positives.
Test under event-like conditions: distance, glare, crowd clutter, similar objects (tripods, monopods, umbrellas, microphones).
Latency budget: detection → human check → alert must be seconds, not tens of seconds.
Fail-safe behavior: What happens if the drone loses link, drifts, or needs an emergency descent? You need standard operating procedures.
Data policy: retention limits, on-device blurring outside ROIs if possible; coordinate with counsel before recording attendees at scale. Facit Data Systems+1

6) Team & process

People: Part 107 RPIC + visual observers; a CV engineer; security lead to own rules of engagement; legal/compliance owner.
Training: operators practice manual confirm—e.g., auto-bookmark suspected frame, yaw/zoom for a second angle before escalating.
Coordination: notify venue and local PD in advance; integrate with their comms plan. Public-safety drone programs are common now, but transparency matters. The Washington Post

7) Build phases (suggested)

Paper design & legal check: pick a drone class that can legally operate near/over people under Part 107 categories or secure a plan for protected perimeters that avoid over-people flight; confirm Remote ID and airspace. Federal Aviation Administration+1
Bench CV prototype: run person+pose+weapon detectors on archived rooftop footage to gauge real-world precision/recall at your expected distances. Expect to tune thresholds heavily. arXiv+1
Dry-run with actors on an empty venue (actors mimic benign vs. suspicious postures). Measure latency and false alarms.
Supervised pilot at a small event with clear perimeters, robust signage, and explicit notification to attendees.
Scale cautiously: add thermal, better zoom, improved tracking, and logging for post-event model tuning (within your privacy policy).

8) What to expect difficulty-wise

Hard: sustained accuracy at long range on moving platforms; permissions to fly near/over crowds; keeping false alerts low enough that you still trust the system; staffing a human-in-the-loop 100% of the time.
Moderate: getting a stable airborne platform with zoom and a good downlink; building a usable alerting UX.
Straightforward: making the phone buzz the way you want once an operator confirms an alert.

Bottom line

Technically plausible as an assistive layer, not a silver bullet.
Treat it as decision support for trained security, with rigorous legal compliance and privacy safeguards.
Start with non-over-crowd perimeters and human-verified alerts, then iterate.

Minimal Pilot Plan

1. Objectives

Primary test: Can a drone-mounted camera + CV pipeline consistently flag a prone person or extended-arm posture on a rooftop at realistic distances?
Secondary test: Can alerts reach you (or a designated operator) with low latency and low false-positive rates?
Exclusions: no crowd overflight, no live weapons, no public deployment.

Drone:
- A sub-250 g category quadcopter (e.g., DJI Mini series) – avoids complex “flying over people” restrictions in early tests.
- Gimbal with at least 3× optical zoom (or high-resolution sensor for digital crops).
Compute:
- Laptop or small edge device (Jetson Xavier, Coral Dev Board) for running detection models.
- Ground link (standard controller feed; save video to feed model in real time).
Models:
- Pretrained person detection (e.g., YOLOv8).
- Pose estimation (arms extended, prone).
- Optional: off-the-shelf weapon-detection model (expect limited reliability, but useful for testing).
Alerting:
- Simple phone app or script that buzzes your phone when an operator “confirms” a detection.

3. Test Script

Site survey: pick a safe, empty building or rooftop environment with clear line of sight.
Actors: have volunteers mimic scenarios:
- Standing casually.
- Kneeling (like photographers).
- Lying prone (no weapon).
- Lying prone with a prop (e.g., tripod, broom).
- Standing with arms extended (tripod vs. mock rifle silhouette).
Flight profile:
- Hover 50–100 m away, point camera at roof.
- Record continuous footage, run detection pipeline live.
Data capture:
- Log true positives, false positives, false negatives.
- Time from “action begins” to “alert received.”
Alert loop:
- Operator sees flagged frame, decides if it’s worth escalating, then taps “confirm.”
- You (or test recipient) receive distinct buzz pattern on phone.

4. Evaluation Metrics

Detection accuracy: % of correct detections at >50 m.
False positives: e.g., prone sunbather, tripod mistaken for gun.
Latency: camera → detection → operator confirm → phone buzz (target ≤5 sec).
Operator load: how often they’re overwhelmed by spurious alerts.

5. Phased Progression

Phase 0 (bench): run models on stored rooftop footage.
Phase 1 (field, empty site): actors on roof, drone hovering, no public exposure.
Phase 2 (controlled event): small gathering with perimeter security, drone watches rooftops only, alerts go to operator—not to you directly.
Phase 3 (scaling): if results are promising, consider heavier drone with thermal + zoom, legal approvals for operations near people, and integration into a full security workflow.

6. Safety & Legal Guardrails

Stay sub-250 g for pilot tests to simplify compliance.
Don’t fly over uninvolved people.
No real weapons—use safe props.
Treat video as sensitive: purge after testing unless needed for training.
Always have a licensed Part 107 pilot if scaling beyond hobby/research use.

✅ By the end of this pilot you’ll know:

Whether CV models can actually spot “prone with long object” at useful distances.
How often false alarms overwhelm the system.
Whether the alerting loop is practical under real conditions.

Tuesday, September 09, 2025

Make Brendan Burchard Your High Performance Coach

"flow" came out of the work of Mihaly Csikszentmihalyi

What you need for flow

Challenging but attainable goals
Strong concentration and focused attention required
Activity intrinsically rewarding
Serene loss of self consciousness
Lose track of time
Immediate feedback on performance
Challenge matches skill level
Sense of personal control over situation and outcome
You stop thinking about your physical needs
Ability to focus completely on the activity (now you know why Zuck is your enemy)

High performers aren't just good at a singular task or skill. They usually learn adjacent competencies to complement a particular expertise. They are not a one hit wonder. They have multiple skill sets that allow them to succeed over the long term and, importantly, lead others. They practise meta habits that allow them to succeed in multiple areas of their lives.

TBC..

Monday, September 08, 2025

Charles Darwin and the Evolution of Python - Gen AI Beats Guido's Own Efforts

Guido : https://python-history.blogspot.com/

And the oracle:

Go to the summary

Python 1.x (1991 – mid-90s)

Initial release (1991): Guido van Rossum’s invention. Already had dynamic typing, exceptions, functions, modules, and high-level data types (lists, dicts, strings).
Big idea: readability and simplicity (“executable pseudocode”).
Earth-shaker: The language itself. At a time when scripting was Perl-dominated, Python’s clarity stood out.

Python 2.0 (2000)

List comprehensions → a concise, expressive way to build lists in one line. Hugely influenced by functional programming and later spread widely in other languages.
Garbage collection with cyclic references → no more manual memory juggling with ref cycles.
Earth-shaker: List comprehensions fundamentally changed idiomatic Python, making it more elegant and expressive.

Python 2.2 (2001)

New-style classes (unifying classes and types). Before, class Foo: ... was distinct from built-in types like int. This release made everything an object with a single inheritance model.
Generators (via yield). Revolutionized iteration — let functions produce sequences lazily.
Earth-shaker: Generators. They made Python truly excellent for data streaming and iteration.

Python 2.7 (2010)

Last major 2.x release. Added dictionary comprehensions and set literals.
Earth-shaker: not a feature, but end of life looming. This froze the world in a “2 vs 3” split for nearly a decade.

Python 3.0 (2008)

Unicode everywhere → strings became str (Unicode), with bytes separate. Made Python a first-class citizen for the modern, international internet.
print() as a function instead of statement.
Removed many warts (like <>, implicit int/long distinction).
Earth-shaker: Unicode by default. It was painful, but essential for the web era.

Python 3.3 (2012)

yield from → made composing generators much more natural.
Virtual environments (venv) → lightweight isolation of dependencies.
Earth-shaker: venv. This made dependency management accessible to everyone, not just sysadmins.

Python 3.4 (2014)

asyncio module → foundation for async/await concurrency.
enum, pathlib modern conveniences.
Earth-shaker: asyncio. It opened Python to high-performance networking, microservices, and event-driven programming.

Python 3.5 (2015)

async and await keywords baked into the language. No longer obscure callback soup — asynchronous code looked almost synchronous.
Matrix multiplication operator @ (for NumPy users).
Earth-shaker: async/await. This turned Python into a serious competitor for Node.js in async I/O.

Python 3.6 (2016)

F-strings (f"Hello {name}"). A massive quality-of-life improvement for string formatting. So widely loved that it instantly became idiomatic.
Underscores in numeric literals (1_000_000).
Earth-shaker: f-strings. Conciseness + readability = a feature devs never looked back from.

Python 3.7 (2018)

Data classes (@dataclass). Removed boilerplate for simple classes.
Context variables (for async).
Earth-shaker: dataclasses. They changed how Python code models data, especially in APIs and configs.

Python 3.8 (2019)

Walrus operator (:=). Assignment inside expressions.
Positional-only parameters.
Earth-shaker: walrus operator. Controversial but powerful (e.g. regex matches in if, avoiding duplication).

Python 3.9 (2020)

Dictionary merge/update operators (|, |=).
New parser (PEG-based) → allows more flexible grammar in the future.
Earth-shaker: PEG parser. Invisible to most users, but opened the door for new syntax (like pattern matching).

Python 3.10 (2021)

Structural pattern matching (match ... case). Brought modern algebraic data-type style dispatch to Python.
Earth-shaker: pattern matching. A new paradigm for control flow, making Python more expressive for compilers, AST work, and config parsing.

Python 3.11 (2022)

Major speed boost (10–60% faster) via specializing adaptive interpreter.
Precise error locations in tracebacks.
Earth-shaker: performance. Suddenly Python wasn’t “slow by default” anymore — CPython itself got much faster.

Python 3.12 (2023)

Even more performance boosts, cleaner C API.
Type system improvements (e.g. generic types without from __future__).
Earth-shaker: types go mainstream. Type hints aren’t just annotations now — they’re actively shaping Python’s ecosystem.

Python 3.13 (2024)

No GIL preview builds (still experimental). Removing the Global Interpreter Lock would be epoch-making for multi-core scaling.
Earth-shaker: If no-GIL lands fully, true parallelism in CPython will change Python’s place in HPC and server workloads.

Summary Table of Earth-Shakers

Version	Earth-shaking Feature	Why it mattered
1.0	The language itself	Readability over Perl/C
2.0	List comprehensions	Expressiveness
2.2	Generators	Lazy iteration
3.0	Unicode everywhere	Web-era ready
3.3	venv	Dependency mgmt for all
3.4–3.5	asyncio, async/await	High-perf networking
3.6	f-strings	Readability boost
3.7	Dataclasses	Boilerplate killer
3.8	Walrus operator	Concise expressions
3.9	PEG parser	Enables new syntax
3.10	Pattern matching	New control paradigm
3.11	Speedups	Python isn’t slow anymore
3.13+	No GIL (in progress)	True parallelism