A feature to provide more control over text size is coming.

The key thing is when I want to present a chart, I have to blow it up on A0 which costs a lot, and people still cant see it properly.

Q) Does anyone know if there is an option to change the font to make it bigger?

So I put up a post here with a concrete example I’m trying to map:

http://forum.flyinglogic.com/index.php?topic=54.msg157#msg157

I’d love to see each of your thoughts on how that data could best be represented as a DAG or hi-tree.

I hardly think that DAGs are so complex when compared to trees that they can be reasonably likened to Quantum theory when compared with Newtonian mechanics! If that were the case, then I doubt anyone except academics would find Flying Logic useful, and this is manifestly not the case.

In my experience, DAGs are a way of representing knowledge that is closer to how our brains actually represent it, and therefore more natural.

In fact, DAGs are used by every spreadsheet package ever created to represent the flow of data through various formulas: any cell can be used by multiple formulas, and any formula can reference multiple cells, as long as no loops are created. Anyone who has ever used a spreadsheet in any depth understands the power of DAGs, even though the graph is not drawn with boxes and lines. This is part of why I like to call Flying Logic “The spreadsheet for rational thought.”

I also note that Flying Logic’s grouping feature can easily be used to gather related premises as needed for analysis purposes, similar to the Rationale analysis view. If there is sufficient demand, I’m sure we could easily automate such a view as well.

Now, back to the DAGs vs trees debate ;-):

Certainly DAGs are more expressive than trees for the reasons that you state. However, whether they are superior depends on the context.

By analogy: Special Relativity is more accurate than Newtonian mechanics (and includes it as a low-order approximation), but the latter is easier to learn, and easier and faster to apply correctly for complex engineering problems!

In the case of complex argumentation, at Austhink we have found the need for an effective representation of co-premises more urgent than the need for multiple parents.

I agree that Rationale is definitely more specifically tuned for traditional argument mapping. However, I think that DAGs are still superior to trees even for that task, as one assertion may be a sub-argument supporting more than one parent argument, or even supporting part of an argument while simultaneously refuting another.

For a solid introduction to building complex arguments check out these free online tutorials:

http://www.austhink.com/reason/tutorials/

[The diagrams were produced using Rationale.]

Rationale might just look like a graphic outliner when you first play with it, but as you delve deeper you will find that there is more to it.

At present Rationale can be used in three increasingly sophisticated ways:

1. In “grouping mode” it allows the user to do a hierarchical tree-based analysis. This is equivalent to a graphic “outliner”.

2. In “reasoning mode” it overlays the tree with semantics tuned to reasoning. It still looks like a graphic outliner, but it allows you to construct informal arguments where one node supports or objects to its parent.

3. In “analysis mode” Rationale adds hi-trees which allow the user to represent the co-premise structure of an argument. (This can be faked with a tree or DAG, but I contend that they don’t do as good a job).

So: It looks like Rationale and Flying Logic have somewhat overlapping concerns, and distinct strengths:

A. Rationale is tuned for visualizing the logical structure of complex argumentation, although it has other applications.

B. Flying Logic is better for visualizing Cause-and-Effect analysis, and perhaps other things.