Regarding SEO Note: I took a very brief look at it and was not impressed with the current version (v1.1). It's seems like a stripped down version of Keynote. Furthermore, it doesn't add native table support or true hyperlinks and without those two things you may as well just stick to Keynote as it does everything else extremely well. The one thing SEO Note has going for it is that it is in current development (I believe).
Regarding formats: I hear you here. That's one reason why I'm waiting to see what kind of I/O both of those programs are going to have. Seo Note does HTML, RTF, and text. That doesnt help for conversions!
Regarding Filters etc: I agree that not all users will immediately see the benefit of this. My first taste of this type of thinking was with Opera Mail Client M2, and it makes email management much easier. I think mouser is going to add similar capabilities into FARR so take a look at the next few versions. But, let me try to describe a quick scenario where Keynote (and any other non-relational manager) leave me gasping for air: At work I worked on a motion control project. In the initial phases I was collecting data and didnt know what I knew. Grabbing random bits of info I would put them where I thought they made sense. As my understanding improved I realized I needed to change the structure somewhat. But, what I also realized is that there are different ways of looking at the same data. Sometimes I want to look at things from a vendors point of view (who manufactures what piece?). Sometimes I want to look at it from a system point of view (what piece does what action?). Sometimes I want to look at it from a piece point of view (who makes what piece and what does it do?). In Keynote, I have to make a different tree for each type of view... very time-consuming to enter data initially and update with new info.
But, if I can choose different properties for each piece of information as I put it into the Information Manager and change those properties as I find new relationships, it allows me to back away from the micro-management of the data. If I have a motor from a certain company, I can relate it to that company, the motor's major specs, and the motor's job in the project I am working on. When I want to view the data in different ways, all I need to do is apply a new filter and the data matching that filter comes into view. If I edit a piece of data in one view, it is changed in all other views because it's the same piece of data. As I find new relationships or need to change old ones, I can "reassign" the data new properties. Now, imagine I am done with the current project. Suppose that a year down the road, I need to look at motion control stuff again. I have a repository of information that I can find quickly and remap quickly (some of the data will apply in the new project). I dont have to worry about where I put information, I dont have to worry about changing various instances of the information, and, best of all, I can mold and shape the way I view the data as I learn how it relates.
Keynote can imitate this with Virtual Nodes. While powerful, and still valid for certain situations, they are external to the main file. I like having everything in one place.
Obviously, there are times when it is good (and quite necessary!) to impose a certain rigid structure on data. But, often, I find myself wanting to throw data into the mix and then, once I figure out how it relates to different things (projects, concepts, parts, etc) quickly relate it.
Wikis are weird. This is partially me just not wanting to change!! But, I dont see Wiki's as the answer. Tree structures are great too, but what I would love to see is a program that can combine different ways of looking at the same data with the click of a button. Flowcharts, trees, block diagram, mindmaps... all these should be able to be used to show the same data. I dont want to have to retype or copy/paste data between views.
So... those are some of my thoughts... hopefully I've bored you to death!
Kevin
