1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
|
[[!meta title="From FRDCSA to FLP2: Building AI-Powered Life Planning Systems in Emacs - A Journey from Research to Real-World Impact"]]
[[!meta copyright="Copyright © 2025 Andrew John Dougherty"]]
[[!inline pages="internal(2025/info/life-nav)" raw="yes"]]
<!-- Initially generated with emacsconf-publish-talk-page and then left alone for manual editing -->
<!-- You can manually edit this file to update the abstract, add links, etc. --->
# From FRDCSA to FLP2: Building AI-Powered Life Planning Systems in Emacs - A Journey from Research to Real-World Impact
Andrew John Dougherty (he/him) - Pronunciation: Andrew DOW-er-tee, IRC: aindilis, Website: <https://frdcsa.org> GitHub: <https://github.com/aindilis>, <mailto:adougher9@gmail.com>
[[!inline pages="internal(2025/info/life-before)" raw="yes"]]
Six years after my EmacsConf 2019 presentation on the Formalized Research
Database (FRDCSA), I return to share the evolution of this ambitious
project into the Free Life Planner (FLP) - a comprehensive AI-powered
system designed to help people manage the complex logistics of daily life,
from basic survival to thriving in challenging circumstances.
This talk chronicles the development journey from FRDCSA's initial vision
of collecting and integrating all free software AI systems, to FLP's
practical focus on solving real-world problems like homelessness, medical
crises, and financial hardship. I'll demonstrate how Emacs serves as both
the development environment and user interface for a system that now
encompasses over 1000 subsystems including meal planning, financial
forecasting, medical management, and emergency preparedness.
Key highlights include:
Technical Architecture: How we've built FLP using SWI-Prolog as the core
reasoning engine, with Emacs providing the primary interface through 4680+
custom ELisp functions. I'll show the integration of modern LLMs (Llama3,
DeepSeek-Prover-V2, etc) with classical AI planning systems (PDDL planners,
behavior trees, agent-based systems) all orchestrated through Emacs.
Security Evolution: The transition from the original internet-connected
FRDCSA/FLP to air-gapped systems, and the development of FRKCSA/FLP2 as
open-source, internet-facing reboots. How we balance powerful AI
capabilities with privacy and security concerns.
Real-World Impact: Using generative AI and our PresGen presentation system,
I'll showcase documented case studies where FLP has helped users navigate
homelessness, medical crises, and benefit loss scenarios. These aren't
hypothetical use cases - they represent the system's practical application
to urgent human needs.
Emacs Integration: Deep dive into our Emacs-based development workflow,
including the CLEAR book reader for academic research, Verber planning
domain editor, UniLang integration for multi-system communication, and our
Academician mode for synchronized document processing. How Emacs serves as
mission control for a complex AI ecosystem.
AI Planning Systems: Demonstration of how we've integrated automated
planning, temporal reasoning, and contingency planning into daily life
management. Users can ask "How do I get through the next month if I lose my
job?" and receive detailed, executable plans.
Community Building: How FLP facilitates mutual aid networks and community
resilience, turning individual survival into collective thriving. The
system doesn't just help people - it helps them help each other.
Future Directions: The roadmap for FLP2, including improved security
models, enhanced AI integration, and broader accessibility. How we're
preparing for a world where AI-powered life assistance could become a
virtual social safety net.
This presentation will use our PresGen system to generate slides
dynamically, incorporating the latest AI developments and real user
feedback. Attendees will see both the technical implementation and the
humanitarian vision that drives this project - demonstrating how Emacs can
be the foundation for transformative social technology.
About the speaker:
Andrew Dougherty is the creator of the FRDCSA (Formalized Research
Database: Cluster, Study and Apply) and Free Life Planner projects. For
over 25 years, he has worked on integrating AI systems to solve real-world
problems, with a focus on helping vulnerable populations navigate complex
life challenges. His work spans automated planning, knowledge
representation, and practical AI applications. Andrew presented FRDCSA at
EmacsConf 2019 and has continued developing these systems with Emacs as the
primary development and deployment environment. He believes in using AI
technology to strengthen communities and provide practical assistance to
those who need it most.
[[!inline pages="internal(2025/info/life-after)" raw="yes"]]
[[!inline pages="internal(2025/info/life-nav)" raw="yes"]]
|
