philippine statistics authority · geospatial data management & engineering
introduction
to the training
bnhr.xyz
--- ## Contents 1. What We Are Building 2. The Open Geospatial Stack 3. How This Training Works 4. Three Personas, One Room 5. The Week Ahead 6. What Is Expected of You ---
what
we
are building
--- ## By Day 5, you will have a better understanding on how to... - Store, share, and manage your geospatial data; - Build applications on top of your data such as dashboards and webmaps; - **ALL USING TOOLS THAT YOU WILL OWN** *We will also look at how we can replace your current ArcGIS Online webmap and migrate from your current proprietary setup to a more open geospatial stack.* **Every exercise this week is a step toward it.** Note: The dashboard isn't a separate capstone task added to the end of the week. Every afternoon lab produces one piece of it. By Day 5, those pieces combine into a working system. Participants who understand the destination are more motivated to complete each intermediate step. --- ## Each day adds one layer — and each layer builds on the one before
PostGIS, GeoServer, GeoNode, Applications, Cloud Native
Note: PostGIS stores it. GeoServer shares it. GeoNode catalogs it. Applications make it usable. Cloud native distributes it. Each day's output is the input to the next day. Participants who fall behind on Day 2 will feel it on Day 3 — the cumulative structure is the main reason to keep up. ---
the
open
geospatial
stack
--- ## Every ArcGIS Online component has an open equivalent — and you will build it
| ArcGIS Online | Open replacement | Day | |---|---|---| | Hosted feature layers | PostGIS + GeoServer WFS | 1–2 | | WMS tile services | GeoServer WMS | 2 | | Portal / data catalog | GeoNode | 3 | | AGOL web map | MapLibre or Leaflet web map | 4 | | AGOL dashboards | Streamlit or Observable | 4 | | Cloud hosting (ESRI-managed) | Self-hosted Docker stack | All | | Partner data distribution | COG, PMTiles, GeoParquet | 5 |
Note: This migration mapping runs through every day. When a new tool is introduced, it always starts from the AGOL component it replaces. The conceptual mapping is 1:1 — only the URLs and the license fees are different. --- ## bring your own data (byod) - We'll provide sample data during the demos and exercises but we encourage you to **bring your own data**. - That way, the data you query, publish, and map this week is the same data your division works with every day. Note: No placeholder data. The exercises feel applicable to daily work because they are. Participants with a specific dataset they want to migrate should name it during the pre-training needs assessment — the facilitator will reference it in demos. ---
how this
training
works
--- ## 30% concepts, 70% hands-on — because you learn tools by using them - Every concept is **followed immediately by hands-on exercises** - **The problem comes first** — the tool is the means to solve it - **Debugging errors is learning time** — built into the schedule, not a sign of failure Note: This ratio is deliberate. You do not learn geospatial software by watching it. The mental model consolidates when you apply it yourself, hit a real error, and work through it. Problem-first sequencing means participants always know why they are learning the tool before being shown how. --- ## Every module follows the same arc — from watching to doing independently
I Do, We Do, You Do
**Do not skip the We Do phase.** It is the bridge that makes the independent lab accessible. Note: The We Do phase is where most in-session learning happens. Fast learners help their neighbours — explaining consolidates their own understanding. Slower learners catch up before the independent lab begins. Jumping from demo straight to independent lab is the most common cause of labs falling apart. --- ## Every morning is a shared plenary — all three groups, together
| Time | Block | What happens | |---|---|---| | 9:00–9:15 | Retrieval practice | Written recall from previous day — no notes | | 9:15–10:15 | Lecture | Concepts and foundation-building | | 10:15–10:30 | Traffic light + break | Check-in & self-assessment | | 10:30–11:10 | Demo (I Do) | Facilitator demonstrates | | 11:10–11:50 | Workshop (We Do) | Guided practice — follow along step by step | | 11:50–12:00 | Debrief + exit ticket | Check-in & self-assessment |
Note: Mornings are shared across all three personas. The traffic light check-in at 10:15 is the facilitator's adjustment mechanism — if the room is mostly red on a concept, the demo pacing changes. The exit ticket is read before the afternoon session begins and shapes how the facilitator opens the debrief. --- ## Afternoons split by persona — same theme, different tasks, one debrief | Group | Afternoon focus | |---|---| | **Builders/Infrastructure** | Deploy, configure, script, and build applications | | **Custodians/Administrators** | Govern and manage data, permissions, and metadata | | **Users/Developers** | Query, consume services, and build maps or dashboards | All groups reconvene at **16:40** to share what they built and what questions came up. *The split is not a separation—**it's a deepening and an opportunity for cross-personal learning***—Builders hear what Users need from the server; Managers hear what Builders are creating; Note: The split is not a separation — it's a deepening. Same morning concepts, applied through each role's lens. The 16:40 reconvergence is often where the most interesting cross-persona learning happens — Builders hear what Users need from the server; Managers hear what metadata Builders are skipping. --- ## Day 2+ starts with retrieval practice — no notes, ten minutes, cold call *One prompt on the board. Write what you can remember. No notes, no slides.* The act of trying to remember—even partially—strengthens memory more than re-reading your notes. Note: Example prompts: "Write the SQL for a spatial join on two PostGIS tables" or "Name the three OGC service types GeoServer publishes." The facilitator corrects and clarifies in 3–4 minutes. Even partial recall during the exercise strengthens memory more than passive review — this is why the exercise runs even when participants feel uncertain. ---
three personas
,
one room
--- ## Builders — deploy, configure, and build the structural foundations of the stack - Deploy PostGIS, GeoServer, and GeoNode - Support Custodians and Users - Publish cloud native formats: PMTiles, GeoParquet, STAC *technical backbone of the team. comfortable with the servers and the command line.* Note: Builders are the technical backbone. Everything Managers and Users rely on in the afternoon labs depends on Builders having deployed and configured the stack correctly. On Day 1, Builders should be the first to confirm the Docker stack is running. --- ## Custodians and Administrators — govern the stack and the data it holds - Design and document data schema and naming conventions - Configure user roles, permissions, and access policies - Manage metadata workflows for ISO-standard catalog records - Draft a cloud native data distribution policy for partner agencies *moderate technical background. comfortable with admin UIs.* Note: Managers are the governance layer. What they configure determines who can publish data, who can access it, and how it's described in the catalog. Their work is invisible when done well and catastrophic when skipped. --- ## Users and developers — connect to the stack and create from it - Connect QGIS to PostGIS and query data directly - Consume WMS and WFS services from GeoServer in existing workflows - Use GeoNode to find, preview, and download datasets - Build a web map or analytical dashboard on the open stack *geospatial-literate. operations, projects, and applications.* Note: Users are the audience that makes the stack worthwhile. Their workflows define what the stack needs to support. On Day 1, Users' questions about QGIS connectivity are often the clearest signal of whether the Builders' PostGIS setup is working. --- ## Each persona owns a different part of the stack in the afternoon
Builders: ●, ●, , , ●, ● Managers: , ●, ●,●,, ● Users: , , , ●, ●, ●
*This shows **focus areas** (i.e. primary responsibilities not exclusive ownership).* Note: The heatmap shows afternoon focus areas — primary responsibility, not exclusive ownership. All personas attend the same morning sessions. Day 5 Distribute column involves all three: Builders generate the formats, Managers set the policy, Users consume the assets. ---
# The Week
Ahead --- ## Five days, five layers — each building on the last
| Day | Morning topic | Afternoon | |---|---|---| | **Day 1** | Storing Spatial Data — PostGIS & Others | Deploy stack · Schema design · QGIS queries | | **Day 2** | Sharing Data — GeoServer and OGC | Configure GeoServer · Permissions · Consume WMS/WFS | | **Day 3** | Geospatial Catalog — GeoNode | Deploy GeoNode · User & data management · Use & consumption | | **Day 4** | Application Layer — Web Maps & Dashboards | **Track A:** web map · **Track B:** dashboard | | **Day 5** | Cloud Native Geospatial | Cloud native publish · Distribution policy · STAC |
Note: Day 4 afternoon is the exception to the persona split — participants choose Track A or B based on what they want to build, not their role. Both tracks are previewed in the Day 4 morning session so the choice is informed. --- ## Day 4 afternoon is a self-selected track — not a persona split
Track A — Geospatial Apps
Build an interactive web map. Tools: MapLibre GL JS · Leaflet · React
Track B — Data Dashboards
Build an analytical dashboard (not necessarily spatial). Tools: Streamlit · Observable Framework · Evidence
Note: Track selection happens at the end of the Day 4 morning session — after both demo tracks are previewed. Participants who want maps AND charts will typically self-select Track A — the mapping is the harder skill to add later. Both tracks produce a capstone output that works with the PSA open stack. ---
expectations
--- ## Three things to verify - **Builders:** Are you familiar with docker? - **Users:** Do you have QGIS and Python installed? - **Everyone:** Know your persona. [**Complete the self-assessment quiz**](../quiz/self-assessment.html). If your setup isn't working, don't hesitate to let us know! --- ## Four norms for the week 1. **Be present in the mornings** — morning concepts power the afternoon labs 2. **Attempt before asking** — read the error, try one fix, then ask for help 3. **Use your neighbour** — think-pair-share moments are deliberate, not optional 4. **Answer the check-ins honestly** — they help shape what comes next Note: Frame these as professional norms, not rules. Exit tickets only work if participants engage honestly — reassure them that "I'm confused" is a valid and valuable answer. ---
let's
build
something.