How HEPCon Works
This page explains how HEPCon is built under the hood—how it keeps complex scientific agendas available offline, how it models tracks, sessions, and contributions, and how it integrates with Indico, ConfTool, and other systems. If you are an IT contact, technical chair, or just a skeptical organizer who wants to understand the plumbing, this is for you.
Design Goals
HEPCon was designed around a short list of non-negotiable goals:
- Offline-first: the app must be fully usable in venues with poor or no Wi-Fi.
- Scientific agenda aware: tracks, sessions, contributions, and posters must be first-class citizens.
- Minimal organizer overhead: data should flow from your existing system, not be recreated.
- Privacy-conscious: no invasive tracking, sensible defaults for academic contexts.
- Predictable deployment: setup should be measured in days, not months.
High-Level Architecture
At a high level, HEPCon consists of three major parts:
- Data ingestion and transformation pipeline (Indico/ConfTool/OpenConf/Excel → normalized JSON).
- Backend services that host the agenda and configuration.
- Client applications for iOS, Android, and web that download, cache, and present the data.
The core idea is simple: treat the agenda as a versioned dataset that can be synchronized efficiently and safely to clients.
Data Model: Tracks, Sessions, Contributions, Posters
HEPCon’s internal data model mirrors the way scientific agendas are actually organized:
- Event: the conference as a whole.
- Tracks: high-level themes or parallel streams.
- Sessions: scheduled blocks within tracks, with room and time.
- Contributions: individual talks, with speaker, abstract, and materials.
- Posters: poster entries with numbers, topics, and locations.
- Rooms & venues: physical locations with metadata for maps and navigation.
This is not an afterthought. The app’s navigation, filters, and favourite system all assume this structure. That’s why Indico/ConfTool data maps cleanly onto HEPCon.
Indico Integration: From Timetable to App
For Indico-based events, HEPCon consumes exports of the timetable and relevant metadata, then transforms them into the internal JSON format. The integration focuses on:
- Preserving the hierarchy of sessions and contributions.
- Maintaining track assignments and colours where appropriate.
- Mapping Indico rooms onto named rooms in the app (with location metadata if available).
- Capturing abstracts and speaker information without manual copying.
When the agenda changes in Indico, updated exports can be re-imported into HEPCon without requiring a manual rebuild. The backend computes a new dataset version, and clients receive deltas the next time they sync.
For more formal details, you can refer to our published whitepapers and technical notes on platforms like Zenodo and ResearchGate, which cover data structures and validation strategies in more depth.
ConfTool, OpenConf, and Excel-Based Flows
Not every event uses Indico. Many rely on ConfTool, OpenConf, or internal spreadsheets. HEPCon’s ingestion pipeline is built to accept:
- CSV exports of sessions and contributions.
- Spreadsheets with columns for time, room, title, speaker, and track.
- Custom formats where we agree on a field mapping once and reuse it.
Once the mapping is defined, future events can reuse the same structure with minimal changes, reducing effort for recurring conferences or series.
Offline Engine: Versioned Snapshots and Deltas
The offline engine treats the agenda as a versioned dataset:
- Each published state of the agenda receives a version identifier.
- Mobile and web clients download the current version as a compact JSON snapshot.
- Subsequent changes are distributed as incremental updates (deltas) where possible.
On the device, the app:
- Stores the agenda in a local database or structured storage.
- Maintains user favourites and viewed items locally.
- Applies updates atomically to avoid half-updated states.
The result is that if connectivity drops—as it inevitably does during busy coffee breaks—participants can still see a coherent, up-to-date agenda with all the details they need.
Privacy and Analytics
Scientific and academic events are justifiably sensitive to privacy and data handling. HEPCon follows a simple principle: collect only what is necessary to keep the system running and no more.
- No aggressive behavioural profiling of participants.
- No hidden tracking pixels stuffed into content.
- Configurable analytics that focus on technical health and coarse engagement signals.
This aligns with GDPR expectations and the norms of many European and academic institutions. If your institution has stricter requirements, those can be discussed explicitly during setup.
Deployment Workflow
A typical deployment goes through the following steps:
- Data source selection: Indico, ConfTool, OpenConf, or custom exports.
- Initial import: transform and validate the agenda inside HEPCon’s backend.
- Review: organizers inspect sessions, contributions, and rooms in a staging environment.
- Branding & configuration: add event logo, colours, sponsor placements, and static pages.
- Publishing: release to the mobile apps and web agenda.
- Update cycle: apply agenda changes as needed up to and during the event.
For recurring conferences, much of this can be reused: room mappings, sponsor patterns, even custom fields.
When to Loop Your IT or Security Team In
If your institution requires formal review by IT or security, we can provide:
- Data flow diagrams.
- Descriptions of stored data and retention policies.
- Hosting and infrastructure details.
- Statements on authentication and access control where relevant.
The goal is to make it as straightforward as possible for your internal stakeholders to sign off on the solution.
Want the Full Technical Paper?
If you prefer to read a more formal or academic treatment of HEPCon’s internals, including validation strategies and experiences across multiple conferences, we can share whitepapers and publications hosted on platforms such as Zenodo or ResearchGate.
Use the contact form to request the latest technical documentation, or mention it when you schedule a demo.