Thanks for sharing this! I've often tried to define Leo in some other various ways when writing documentation... so I find this experiment very interesting!
... I think I'm going to try it out right this gemini3 thingy now for myself :) Félix On Saturday, November 22, 2025 at 5:07:13 AM UTC-5 Edward K. Ream wrote: > Here is my prompt to gemini3: > > > Please suggest improvements to Leo's "preface" at > https://leo-editor.github.io/leo-editor/preface.html > > Everything that follows is the response. Looks like I flunked marketing > 101: > > Based on an analysis of the text at the provided URL, the "preface" > (titled "A Brief Summary of Leo") exhibits several rhetorical weaknesses > that undermine its effectiveness for a technical audience. > > The current text relies on subjective claims, dated comparisons, and > insider jargon ("Leonine way"). A skeptical reader will likely view the > broad listing of features (IDE, PIM, Outliner) as a sign of a tool lacking > a defined scope, rather than a versatile one. > > The following improvements address these issues, focusing on precision and > verifiable value propositions. > > 1. Remove the Dated Comparison > > *Current:* *“Word outlines are very useful. But Leo makes Word look like > a clunky toy.”—Joe Orr* > > *Critique:* Comparing a developer tool to Microsoft Word is a category > error. It lowers the ceiling of expectations. Most developers do not use > Word for code or serious data organization; they use VS Code, Vim, Emacs, > or Obsidian. Acknowledging Word implies the tool is aimed at non-technical > office users, which contradicts the later claim that it is a > "fully-featured IDE." > > *Recommendation:* Remove the quote entirely. Start with a direct > definition of the tool's primary differentiator: the integration of > outlining with codebase logic. > > 2. Clarify the "Directed Acyclic Graph" (DAG) Claim > > *Current:* *“Leo's underlying data is a Directed Acyclic Graphs. As a > result, Leo organizes data in completely new ways.”* > > *Critique:* Stating the data structure is a DAG is a factual statement > about implementation, not a benefit. The statement "organizes data in > completely new ways" is vague marketing fluff. The user needs to know > *why* a DAG matters (i.e., zero-cost aliasing/cloning of nodes). > > *Recommendation:* Replace the implementation detail with the functional > consequence. > > - > > *Proposed Revision:* "Unlike standard tree-based editors, Leo allows a > single node to appear in multiple places simultaneously (clones). Editing > a > node in one location updates it everywhere, allowing for multidimensional > organization of source code." > > 3. Define "The Leonine Way" or Remove It > > *Current:* *“Leo's clone find commands enable the Leonine way to refactor > programs.”* > > *Critique:* This is circular terminology. A new user does not know what > "Leonine" means. Using proper nouns to describe a methodology often signals > an insular community rather than a standard tool. > > *Recommendation:* Describe the methodology strictly in operational terms. > > - > > *Proposed Revision:* "Leo enables refactoring by treating code > structure as a manipulatable [sic] outline, allowing you to reorganize > large codebases via drag-and-drop operations rather than text cutting and > pasting." > > 4. Address the "Jack-of-All-Trades" Risk > > *Current:* The list claims Leo is an IDE, an Outliner, a Data Manager, a > Scripting Environment, and a study tool. > > *Critique:* Tools that claim to do everything often trigger skepticism > regarding their depth in any single area. A developer will question how a > single tool can compete with IntelliJ (IDE) and Obsidian (PIM) > simultaneously. > > *Recommendation:* Hierarchy is needed. Define the tool primarily as one > thing (likely a *Literate Programming Editor* or *Scriptable Outliner*) > and list the others as secondary capabilities derived from that core. > > 5. Reframe the "Steep Learning Curve" Warning > > *Current:* *“You won't learn all about Leo in a day or two... Please ask > for help immediately if you get stuck.”* > > *Critique:* While honest, this frames the complexity as a liability > rather than an investment. "Ask for help immediately" suggests the UI is > non-intuitive. > > *Recommendation:* Focus on progressive disclosure. > > - > > *Proposed Revision:* "While Leo's advanced scripting features offer > depth that takes time to master, the basic outlining features provide > immediate utility for navigating complex files." > > 6. Verify the "100% Pure Python" Claim > > Assumption: The claim "Written in 100% pure Python" assumes the user > values interpretability over performance. > > Critique: For a modern IDE, "100% Python" may raise concerns about latency > and UI responsiveness compared to Rust or C++ based editors (like Zed or VS > Code). > > Recommendation: If performance is a concern, qualify this claim by > explaining why it is Python (e.g., "ensuring the entire editor is hackable > [sic] and scriptable by the user at runtime"). > > Summary of Proposed Structural Changes > > > 1. > > *Headline:* Change from "A Brief Summary" to "Technical Overview". > 2. > > *Lead:* Delete the Word quote. State clearly that Leo is a *scriptable > outline editor* for code and data. > 3. > > *Features:* Group features by workflow (e.g., "For Coding," "For > Scripting," "For Knowledge Management") rather than a flat list. > 4. > > *Technical differentiation:* explicitly link "Clones" to the DAG > structure to explain the mechanism of action. > > Edward > > P.S. Here <https://gemini.google.com/share/653d72b46500>is the initial > link to this conversation. > > EKR > > -- You received this message because you are subscribed to the Google Groups "leo-editor" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion visit https://groups.google.com/d/msgid/leo-editor/dbdf765b-2d7f-473f-9f33-62fc9118879cn%40googlegroups.com.
