Learn how easy it is to sync an existing GitHub or Google Code repo to a SourceForge project! See Demo

Close

Moving beyond pure hierarchy - pros & cons

Ryan
2008-03-18
2013-04-12
1 2 > >> (Page 1 of 2)
  • Ryan
    Ryan
    2008-03-18

    Hi,

    FreeMind is a tool for working with trees. There are others programs for creating graphs, maps, flowcharts, etc. that are excellent. FreeMind should not try and compete with them. They are better in their application field, FreeMind triumphs in its application field.

    There are two possible areas of functionality that we need to consider, that may compromise FreeMind's utility:

    1) Clones
    2) Free floating nodes

    I will add my thoughts in a follow-up post, leaving this post as an invitation to discussion.

     
    • Ryan
      Ryan
      2008-03-18

      Hi,

      My thoughts on this will be in three posts, to make them easier to take in, should anyone want to. My first post will introduce some basic considerations, my second will look at clones, and my third will look at free floating nodes.

      Important considerations:

      1) FreeMind teaches people the power of working with trees by enforcing that limitation.

      2) Features that go beyond pure hierarchy are often requested or have been implemented (clones).

      3) Progress and evolution is inevitable, even if it is not always positive.
       

       
    • Ryan
      Ryan
      2008-03-18

      Clones:

      (-ve) An improper use of clones allows people to be lazy regarding hierarchy.

      (+ve) Clones used properly allow greater power, this comes from the fact that data can occupy more than once place in a hierarchy.

      (-ve) It is worrying that people will not learn how to classify data according to *orders*, that they may not get to understand the principles of taxonomy.

      (neutral) Having clones means the data structure that one is working with becomes a DAG *visually represented as a tree*, this is a big difference to a standard tree.

      (+ve) Leo has had cloning for a while. And there are no signs that it engenders sloppy thinking.

      (+ve) Clones help with workflow and thinking because you don't have to go and find one of your ideas in a distant tree.

      (+ve) I already either desire to, or actually do, duplicate data in different part of the hierarchy. This is because data is relevant in different ways in different contexts.

      My conclusion on clones:

      FreeMind should have clones. But perhaps we should try and balance it with providing some educational materials? I'm currently trying to write a book on all this kind of stuff.

       
      • Seth Johnson
        Seth Johnson
        2008-03-19

        I was asked to repost this here.  -- Seth

        ---

        Several things are not clear:

        - What you mean by "free floating nodes"

        - What you are trying to say whenever you say "hierarchy"

        - What real connection you're drawing, if any, when you say clones allow people to be lazy regarding hierarchy

        I think you associate "auto-positioning" with the issue of "free-floating nodes", as in you seem to feel that free-floating nodes and auto-positioning are somehow mutually exclusive.

        I think you're not just talking about free-floating nodes in the sense that they have no line connecting them to their parents (and children). You mean nodes that users can move around on the map.

        And you say you fear users will turn FreeMind into a graphics drawing and concept-mapping kind of thing (and now you're lumping in DAGs with that).

        It's not clear what you mean by clones allowing people to be lazy regarding hierarchy -- apparently you simply mean *taxonomy*. You'd like FreeMind to be about teaching people to be rigorous in their classification, and clones might get in the way of that design rationale.

        In terms of hierarchy -- just plain hierarchy -- a tree presentation, whether in radial mindmap, FreeMind-stacked or traditional outline form, is not different from any other tree presentation (regardless of whether it's representing a DAG).

        The only sense in which "laziness" gets implied is in relation to the relatively esoteric sense of trying to teach people a specific discipline of classification -- a particular use of hierarchy.

        I think you mean: linear, ordered and taxonomic, both abstractly and visually. For you, free-floating nodes contradict the visual presentation of a vertical list, and clones might take away from making that case.

        When you suggest that "free floating nodes" be treated as "quasi" nodes, what you really mean is to make secondary any non-linear, non-visually aligned/ordered -- and non-taxonomic -- uses. This design constraint is unnecessary. And hierarchy does not imply any of those things.

        Leo fulfills all the pillars of FreeMind you list. The meaning and pertinence of those pillars, is actually not very clear -- they're aspects that people interpret in abstruse ways, not necessarily illuminating and coming to real understanding of what's really in hand with FreeMind.

        A point of clarification regarding the usefulness of clones: clones help with workflow because the ideas that are needed are right in front of you, while their original/alternate locations are also maintained. And as you work on a body of information, the various collections of cloned and arranged set of information serve as documentation that lets you get right back into a task.

        Seth

         
        • Ryan
          Ryan
          2008-03-19

          Ok, Seth. I've answered your points. It looks like a lot of info but I have pasted your comments in too, and answered you point by point. I appreciate your input, maybe we could try and boil this down to a few key issues?

          > Seth's comments
          - my comments

          Part One:

          > What you mean by "free floating nodes"

          - The picture I have in mind is of a box, separate from the main tree -- connected to other nodes, or not. I think when users request this they have various things in mind. Some want to add some annotation to a map, others essentially want labelled arrow links. Some people even want multiple roots, that can be added at will.

          > What you are trying to say whenever you say "hierarchy"?

          - Taxonomy is probably closest to what I mean. But I'm also thinking about how things in general are broken down into classes, subclasses and members. Some Venn Diagrams too, show hierarchy in the way I think about it.

          I'm reaching for an analogy with the term 'order' from Logic or Maths, but it's not coming to me.

          > What real connection you're drawing, if any, when you say clones allow people to be lazy regarding hierarchy.

          - I mean when we are first wrestling with a project, we have to think quite hard about how to break it down and where to put this or that piece, so that it has some kind of logical order.

          - The term 'folksonomy' disparagingly refers to the way people can now classify information by simply adding a 'tag' -- they don't have to think about structure. The use of clones, at its worst, would mean throwing data together in a haphazard fashion with plenty of clones, so everything is easy to find because there are copies all over the map.

          > I think you associate "auto-positioning" with the issue of "free-floating nodes", as in you seem to feel that free-floating nodes and auto-positioning are somehow mutually exclusive.

          - I think that the location of a free floating node is always going to be more ambiguous than a node on a tree. The user has to think about where the best location is, to represent what he/she wants to represent in that particular instance.

          > I think you're not just talking about free-floating nodes in the sense that they have no line connecting them to their parents (and children).  You mean nodes that users can move around on the map.

          - Correct. I mean separate from the main tree, with or without a connector of some sort, and can be moved around freely.

          > And you say you fear users will turn FreeMind into a graphics drawing and concept-mapping kind of thing (and now you're lumping in DAGs with that).

          - I know concept maps aren't the same as DAGs, but I thought Ed Ream describes the structures Leo creates as DAGs. Am I wrong on this?

          >It's not clear what you mean by clones allowing people to be lazy regarding hierarchy -- apparently you simply mean *taxonomy*.  You'd like FreeMind to be about teaching people to be rigorous in their classification, and clones might get in the way of that design rationale.

          - Yes, I mean taxonomy. I don't want to be prescriptive; dictating how people should work or think. But trees, taxonomy, and hierarchy are closely tied together -- and I think it's really good to learn to work this way. And FreeMind is a tree editor.

           
          • Seth Johnson
            Seth Johnson
            2008-03-20

            I'm posting three separate threads below here.  :-)

            Seth

             
            • Seth Johnson
              Seth Johnson
              2008-03-20

              Seth:

              > > What you are trying to say whenever you say "hierarchy"?
              > >
              > > What real connection you're drawing, if any, when you say
              > > clones allow people to be lazy regarding hierarchy.
              > >
              > > It's not clear what you mean by clones allowing people to
              > > be lazy regarding hierarchy -- apparently you simply mean
              > > *taxonomy*. You'd like FreeMind to be about teaching
              > > people to be rigorous in their classification, and clones
              > > might get in the way of that design rationale.

              Ryan:

              > - Yes, I mean taxonomy. I don't want to be prescriptive;
              > dictating how people should work or think. But trees,
              > taxonomy, and hierarchy are closely tied together -- and I
              > think it's really good to learn to work this way. And
              > FreeMind is a tree editor.
              >
              > - I mean when we are first wrestling with a project, we
              > have to think quite hard about how to break it down and
              > where to put this or that piece, so that it has some kind
              > of logical order.
              >
              > - Taxonomy is probably closest to what I mean. But I'm
              > also thinking about how things in general are broken down
              > into classes, subclasses and members. Some Venn Diagrams
              > too, show hierarchy in the way I think about it.

              > I'm reaching for an analogy with the term 'order' from
              > Logic or Maths, but it's not coming to me.

              I don't think you should treat taxonomy (or encouragement of
              any kind of "rigorous grouping") as a design principle for
              FreeMind.  All you want to do is give people a tool that
              lets them branch nodes under nodes to group and connect
              sporadic notions freely, make a branching, tree structured,
              visual layout.

              We don't have to taxonomize rigorously at all.  Especially
              when you're first wrestling with a project.  And in fact,
              that instant is the single most important moment when a tool
              like FreeMind is most in its element.  Encouraging
              taxonomizing isn't the thing that's working at that moment.
              Users group things and  perhaps, only in some places,
              taxonomize as an incidental aspect of what they're doing

              Ryan:

              > The term 'folksonomy' disparagingly refers to the way
              > people can now classify information by simply adding a
              > 'tag' -- they don't have to think about structure. The use
              > of clones, at its worst, would mean throwing data together
              > in a haphazard fashion with plenty of clones, so
              > everything is easy to find because there are copies all
              > over the map.

              Everywhere you put a clone, it's for a purpose; you add
              siblings for that same purpose and you arrange them under a
              header for that purpose.  You then go to that purpose and do
              what you want there.  It's nothing like tags that you
              propagate while dreaming of using them consistently and
              hoping to recall your information by remembering them.

              Seth:

              > > The only sense in which "laziness" gets implied is in
              > > relation to the relatively esoteric sense of trying to
              > > teach people a specific discipline of classification -- a
              > > particular use of hierarchy.

              Ryan:

              > - OK, You caught me. ;-) I agree that there is more than
              > one way to use hierarchy. But surely you will grant there
              > needs to be some underlying understanding for hierarchy to
              > be used most effectively? However one is using it.

              Well I grant that everybody inherently has some kind of
              understanding (though even there, I know I've used
              mindmapping tools to just grab a bunch of unconnected bits
              and fold them away in subsets that are more manageable only
              by dint of their size).  But I think you want to argue for
              more -- probably related to visual design, given how you
              want to treat "free floating nodes" -- by talking about
              hierarchy and some sort of discipline like rigorous
              classification; and I don't see it being necessary as such
              or necessarily useful.

               
            • Seth Johnson
              Seth Johnson
              2008-03-20

              Seth:

              > > And you say you fear users will turn FreeMind into a
              > > graphics drawing and concept-mapping kind of thing (and
              > > now you're lumping in DAGs with that).

              Ryan:

              > I know concept maps aren't the same as DAGs, but I thought
              > Ed Ream describes the structures Leo creates as DAGs. Am I
              > wrong on this?

              They are DAGs.  I'm not interested in concept maps here (or
              at all, really).

              Seth:

              > > In terms of hierarchy -- just plain hierarchy -- a tree
              > > presentation, whether in radial mindmap, FreeMind-stacked
              > > or traditional outline form, is not different from any
              > > other tree presentation (regardless of whether it's
              > > representing a DAG).

              Ryan:

              > I agree with the first part of this comment. But it is my
              > understanding -- and correct me if I am wrong -- that once
              > you add clones, the underlying structure of the data
              > ceases to be wholly hierarchical and becomes DAG-like. The
              > benefit of a tree presentation is then largely cognitive,
              > the actual data structure is more complicated.

              I'm not concerned with the data structure -- I'm concerned
              with the fact that visually you've got manipulable branches
              radiating from a center.  That can include clones without
              any detriment.

               
              • Bernard Lang
                Bernard Lang
                2008-03-20

                Hi Ryan / Seth,

                In case anyone was wondering what a DAG was, you weren't alone. Don't be embarassed (I've been cogitating for 53yrs, my teeth are falling out, and I wrote my first code on punch card 36yrs ago, & I didn't know what it was), it is (apparently) a "Directed Acyclic Graph":

                http://en.wikipedia.org/wiki/Directed_acyclic_graph
                http://en.wikipedia.org/wiki/Topological_sort

                Don't be baffled by bs (occasionally, it's actually good to know), just get wiki it ... ;-)

                Bernard

                 
                • Ryan
                  Ryan
                  2008-03-21

                  Hi Bernard,

                  But you know about a lot of stuff I don't have a clue about. ;-) I can't write even a line of code. I'm thinking maybe that should change, but there are so many things I am trying to learn about that it seems a big commitment.

                  Ryan

                   
            • Seth Johnson
              Seth Johnson
              2008-03-20

              Seth:

              > > What you mean by "free floating nodes"
              > >
              > > I think you're not just talking about free-floating nodes
              > > in the sense that they have no line connecting them to
              > > their parents (and children). You mean nodes that users
              > > can move around on the map.
              > >
              > > I think you associate "auto-positioning" with the issue of
              > > "free-floating nodes", as in you seem to feel that free-
              > > floating nodes and auto-positioning are somehow mutually
              > > exclusive.

              Ryan:

              > - The picture I have in mind is of a box, separate from
              > the main tree -- connected to other nodes, or not. I think
              > when users request this they have various things in mind.
              > Some want to add some annotation to a map, others
              > essentially want labelled arrow links. Some people even
              > want multiple roots, that can be added at will.
              >
              > - Correct. I mean separate from the main tree, with or
              > without a connector of some sort, and can be moved around
              > freely.
              >
              > - I think that the location of a free floating node is
              > always going to be more ambiguous than a node on a tree.
              > The user has to think about where the best location is, to
              > represent what he/she wants to represent in that
              > particular instance.

              A free floating node that's connected to the tree is still
              on the tree.

              A node that is connected to the tree is not separate from
              the tree -- except in the case (and speaking only of the
              visual representation) that the node is connected in a way
              that makes the tree no longer a tree -- as in, it forms a
              loop.  Moving nodes around doesn't change that fact.

              I'm not interested in nodes that aren't connected to the
              tree, visually or otherwise -- except that I do think
              there's no harm whatsoever in having them.  Not that it
              makes a difference, but the functionality of the map comes
              from the map itself, regardless of other features that may
              or may not be employed.  I think multiple roots, visually
              included in the same screen, are off track (though clearly
              having multiple roots represented separately, as in on their
              own screens or separate pages, is perfectly sensible and
              consistent).

              Seth:

              > > I think you mean: linear, ordered and taxonomic, both
              > > abstractly and visually. For you, free-floating nodes
              > > contradict the visual presentation of a vertical list, and
              > > clones might take away from making that case.

              Ryan:

              > - You kind of get what I mean here. I think there are two
              > issues; the visual presentation of some particular data or
              > facts, and the cognitive use of maps/trees to help us
              > understand and take control of our work.
              >
              > - Free floating nodes are good when we want to annotate a
              > map in some way. But when hierarchy is neglected, I find
              > the visual representation less useful for actually working
              > with. For example, I don't find concept mapping software
              > as useful as FreeMind, even though it is more flexible.

              This is where I think you're trying to make a case against
              letting people move nodes around on the map, and perhaps for
              automatically arranging nodes in straight vertical stacked
              lists on either side of the center -- as opposed to non-
              linearly arranging nodes all around.  I don't think the
              consideration of free floating nodes in this way is
              relevant.

              I think anybody who wants to say "the branches of my tree
              are mutually exclusive orders of genera and species" or some
              such, will just do so.  In the meantime, everybody else will
              still use the tool lots of ways.

              Seth:

              > > When you suggest that "free floating nodes" be treated as
              > > "quasi" nodes, what you really mean is to make secondary
              > > any non-linear, non-visually aligned/ordered -- and non-
              > > taxonomic -- uses. This design constraint is unnecessary.
              > > And hierarchy does not imply any of those things.

              Ryan:

              > - Yes, you understand what I mean here. Will you flesh out
              > your position on this? When I think of a tree, I think of
              > hierarchy and I think the information on display should in
              > some way align with its location in the hierarchy. The
              > visual structure; the tree, should be relevant to the way
              > one is 'plotting' one's data.

              I think that free-floating nodes as you're discussing them
              are irrelevant to the question of implementing clones.  I
              think for you they are an opportunity to initiate discussing
              hierarchy in a certain way (which I think is not necessary),
              and you do so in order to argue for the way you want
              FreeMind to look visually -- perhaps in part because clones
              might make it harder to make that case.

              First level nodes out from the center do not need to be
              stacked vertically, and the option for them to arrange
              themselves radially around the center -- and let the user
              move them where they please -- should be there.  I actually
              think it's an odd phenomenon when I hear folks trying to say
              how vertically stacked nodes is somehow intrinsic to why
              FreeMind is so great.

              Now note, a visual tree structure of whatever form --
              outline, radial mindmap, bowtie FreeMind diagram, wacky
              squiggly branches splayed wildly around each node throughout
              the tree -- all inherently display information in a way that
              aligns with its location in the hierarchy.

              But a tree structure does not necessarily imply anything
              linear -- which point mindmapping puts to great use.

              A note more to the side than that: as far as "alignment
              based on location in hierarchy" goes, you're always likely
              to have second level nodes as horizontal lines that would
              reasonably be stacked vertically, just because that's the
              most direct way to do text in most graphics environments,
              and of course that's more readable.

               
            • Ryan
              Ryan
              2008-03-21

              Seth, thanks for your comments,

              I will try and put a reply together in the next few days. I will try be concise and pinpoint our main areas of agreement and disagreement.

              Thanks,

              Ryan

               
        • Ryan
          Ryan
          2008-03-19

          Reply to Seth. Part two (of two):

          >In terms of hierarchy -- just plain hierarchy -- a tree presentation, whether in radial mindmap, FreeMind-stacked or traditional outline form, is not different from any other tree presentation (regardless of whether it's representing a DAG).

          - I agree with the first part of this comment. But it is my understanding -- and correct me if I am wrong -- that once you add clones, the underlying structure of the data ceases to be wholly hierarchical and becomes DAG-like. The benefit of a tree presentation is then largely cognitive, the actual data structure is more complicated.

          > The only sense in which "laziness" gets implied is in relation to the relatively esoteric sense of trying to teach people a specific discipline of classification -- a particular use of hierarchy.

          - OK, You caught me. ;-) I agree that there is more than one way to use hierarchy. But surely you will grant there needs to be some underlying understanding for hierarchy to be used most effectively? However one is using it.

          > I think you mean: linear, ordered and taxonomic, both abstractly and visually. For you, free-floating nodes contradict the visual presentation of a vertical list, and clones might take away from making that case.

          - You kind of get what I mean here. I think there are two issues; the visual presentation of some particular data or facts, and the cognitive use of maps/trees to help us understand and take control of our work.

          - Free floating nodes are good when we want to annotate a map in some way. But when hierarchy is neglected, I find the visual representation less useful for actually working with. For example, I don't find concept mapping software as useful as FreeMind, even though it is more flexible.

          >When you suggest that "free floating nodes" be treated as "quasi" nodes, what you really mean is to make secondary any non-linear, non-visually aligned/ordered -- and non-taxonomic -- uses.  This design constraint is unnecessary.  And hierarchy does not imply any of those things.

          - Yes, you understand what I mean here. Will you flesh out your position on this? When I think of a tree, I think of hierarchy and I think the information on display should in some way align with its location in the hierarchy. The visual structure; the tree, should be relevant to the way one is 'plotting' one's data.

          > Leo fulfills all the pillars of FreeMind you list.  The meaning and pertinence of those pillars, is actually not very clear -- they're aspects that people interpret in abstruse ways, not necessarily illuminating and coming to real understanding of what's really in hand with FreeMind.

          - What I wrote is in line with what Dan, FreeMind's former director, thought of as FreeMind's purpose. How would you sum up 'what's really in hand with FreeMind'?

          - I agree that Leo fulfils my criteria, too. Leo is an excellent tool. One limitation with Leo is that every paragraph is not a node. There is a higher degree of separation between structure and content than with FreeMind. Once you start writing your body text, you can't move things around as easily. And you can't view as much data at once in Leo, even with multiple windows open in the body pain.

          > A point of clarification regarding the usefulness of clones: clones help with workflow because the ideas that are needed are right in front of you, while their original/alternate locations are also maintained.  And as you work on a body of information, the various collections of cloned and arranged set of information serve as documentation that lets you get right back into a task.

          - Yes, I that's exactly it! Do you find that idea useful?

          Thanks for being patient with all these comments.

          Ryan

           
          • Bernard Lang
            Bernard Lang
            2008-03-19

            Hi Ryan/Seth,

            >>linear, ordered and taxonomic

            I think this is a good definition of the "rigorous" view & that "taxonomy" is, indeed, the key word. As Seth points out Clone's still exist within outline/tree structure, but in biological taxanomy a life form cannot be a member of 2 kingdoms(?).

            Bernard

             
      • Rigel
        Rigel
        2008-03-22

        For me, and your response to cloning, I do it this way:

        I use a very simple non-breaking hierarchical method.

        I will create a hierarchy, and any node I need to repeat, I will simply link to instead.

        Since each clone of that node usually has to have special context for me, if I create a non-graphical link INSTEAD and just copy copy the non-graphical link all over teh place, I'm always working with the same hierarchy, and also have the option of changing the contextual text / image / script of any linked node.

        I think this is the way things should be done.

        My two cents.

         
    • Bernard Lang
      Bernard Lang
      2008-03-18

      Hi Ryan,

      Used up my ammunition on "Freemind with  Node Cloning". If Seth posts uri for Leo that might be good object lesson for discussion.

      Off to bed now (too much mignight oil)...

      Bernard

       
      • Seth Johnson
        Seth Johnson
        2008-03-19

        Leo's Sourceforge page:
        > http://sourceforge.net/projects/leo/

        Some screenshots:
        > http://sourceforge.net/project/screenshots.php?group_id=3458&ssid=22298

        A tutorial here:
        > http://www.3dtree.com/ev/e/sbooks/leo/sbframetoc_ie.htm

        Start on this one and go through past the 20th slide.  Then go back out and go into other areas.  Don't start on "derived files."  That's the wrong way in for FreeMind-ers, and might even be a huge hurdle for coders who would appreciate that functionality if they really go into it . . .

        An intro:
        > http://webpages.charter.net/edreamleo/intro.html

        Go straight to the "clones and views" section.

        Skip the "derived files" stuff first time through, though you can let the notion simmer that Leo is used (its original purpose was) to manage a linear codebase.

        Note the importance attributed to outlines and outlining (with clones).  Clones in Leo are branch clones, not independent individual clones.

        There's some good presentation of the nature of Leo here:
        > http://webpages.charter.net/edreamleo/testimonials.html

        Downloads:
        > http://sourceforge.net/project/showfiles.php?group_id=3458&package_id=29106

        Discussion forums recently moved to Google Groups:
        > http://groups.google.com/group/leo-editor

        leo's Wiki:
        > http://leo.zwiki.org/LeoWiki

         
        • Seth Johnson
          Seth Johnson
          2008-03-19

          > A tutorial here:
          > > http://www.3dtree.com/ev/e/sbooks/leo/sbframetoc_ie.htm

          > > Start on this one

          . . . I mean, this particular tutorial, of those listed:

          > http://www.3dtree.com/ev/e/sbooks/leo/Leo/Basics/Books/C4_Outlining/slide01.htm

          > [. . .] and go through past the 20th slide. Then
          > > go back out and go into other areas. Don't start on "derived
          > > files." That's the wrong way in for FreeMind-ers, and might
          > > even be a huge hurdle for coders who would appreciate that
          > > functionality if they really go into it . . .

          Seth

           
          • Bernard Lang
            Bernard Lang
            2008-03-19

            Hi Seth,

            Had a look at the Leo links and

            http://webpages.charter.net/edreamleo/intro.html#clones-views

            seems to give a clear definition of my original expectations re cloning (with edits):

            "A cloned node is a copy of a node that changes when the original changes. Changes to the children, grandchildren, etc. of a node are simultaneously made to the corresponding nodes contained in all cloned nodes. A small red arrow in icon boxes marks clones ... there is no distinction between the original node and any of its clones ... insert a node, say B, as a child of any of the A nodes. All the A nodes now have a B child ... when the second-to-last cloned node is deleted the last cloned node becomes a regular node again."

            Bernard

             
        • chp
          chp
          2008-04-07

          Discussion forums recently moved to Google Groups:
          > http://groups.google.com/group/leo-editor

          Ironically, the very first thread I saw there was about "Dynamic nodes?" :) I know it is not the best place to start to talk about it here, but since another "must have" like "cloning" is being heavily discussed here I've decided to add my 5cents..

          Node cloning (including "children") is great. We really need this feature, since bits of data often apply to diffеrent parts of "tree"..
          But the real "next" step would be - tags and dynamic nodes! If one could to add tags like "importance", "due to" and couple of "subjects", the dynamic nodes could be "scripted" so one could edit single instance of a node in any place, rather than make copies or webs of links that got to be followed each and every time.

          It isn't about PIM-s alone. Everything is "tagged" & "dynamic" these days :)

          What do you think?

           
          • chp
            chp
            2008-04-07

            I mean, cloning - is manual thing.. Tags + scrips = dynamic :)
            It would make Freeming more flexible and easy to use (only if tagging would be implemented as easy to use feature of course :)

             
    • Ryan
      Ryan
      2008-03-18

      Free Floating nodes:

      (-ve) this feature has the biggest potential for abuse. Users will contort  FreeMind into a graph/DAG/concept map/network drawing program. They do this because FreeMind is so much quicker to work with.

      (+ve) trees sometimes do converge, and it can be helpful to have labels, etc. Free floating nodes can show relationships.

      (-ve) as Dan pointed out, the more you allowance you create for a non-hierarchical structure, the more FreeMind's core values are compromised. These are hierarchy, folding, auto-position and extreme ease and speed of use.

      The user has to make decisions, such as where to place this or that free floating node, and the devs need to consider how it effects folding -- if nodes are floating around everywhere, attached to different trees, some folded, some not.

      My solution:

      Developers should keep in mind, the Five Pillars of FreeMind (according to me ;-)):

      Five Pillars of FreeMind:

      1) Primacy of Hierarchy
      2) Primacy of Folding
      3) Auto-position
      4) Extreme ease of use
      5) Extreme speed of use

      My suggestion is that free floated nodes should be implemented as an extension of graphical links/connectors. Free floating nodes should be given the status of second class citizens. They should not be given the the status of nodes within (or outside of) the hierarchy, but should be seen as pseudo-nodes, or quasi nodes. And this status should be reflected in the UI, file format, and implementation.

      Thanks to anyone who has been patient enough to read all my ramblings. I luck forward to reading what others have to say. :-)

       
      • Ryan
        Ryan
        2008-03-18

        Obviously. I meant 'I *look* forward to reading what others have to say.'  Doh! 

         
    • Bernard Lang
      Bernard Lang
      2008-03-20

      Hi ll,

      While I'm logged on, just a quick (heart-felt) thought. Communicating through this forum is pretty danmned awkward. No <html>, no <href="">, no <ul>. Given the smarts of (in my estimation) your average Freeminder, this is very limiting (I like to annotate what I write with - inobtrusive/optional - links to wikipedia, but I can't).

      Recommend either Freemind liberates posting on this forum from the shackles (may not be possible), or migrates to a richer/more user friendly forum environment post haste.

      JMy 2C...

      Bernard

       
      • Ryan
        Ryan
        2008-03-21

        Hi Bernard,

        I don't think there is much chance of migrating the forums. It is true that Sourceforge has some frustrating limitations. But it also provides so much, at no coast at all.

        Ryan

         
1 2 > >> (Page 1 of 2)