Using Compendium as a research tool

The below is adapted from an e-mail exchange* with Helga Kocurek, who is doing a PhD in Philosophy at Massey University in New Zealand. She had read our paper on using Compendium for doctoral research, and had some questions on how to use Compendium for her work.

* * *

Al,

I am a PhD student in philosophy and the only person around interested in using such technology. I have been looking for a long time for a good way to store/integrate/find again all the various notes I have taken. I have read your paper Hypermedia as a Productivity Tool for Doctoral Research and was wondering if you have learned any more about how to use compendium as a research tool for PhD work. (Or if you know web sources specifically for this purpose. All I have found is business oriented.

* * *

Helga,

I use Compendium in many ways, as well as help develop it and do research into its use. Is there some specific area(s) that you're interested in for PhD work?

For example, I use it to
- take notes on my readings and analyses
- keep track of tasks and action items
- store notes and maps from meetings
- prepare and give presentations
- query the database for quotations/citations
- mess around in random ways
- construct more formal maps relating concepts in various ways

Let me know what kinds of things you'd like me to expand on.

* * *

Al,

The two most important things at the moment are

- take notes on my readings and analyses
- prepare and give presentations

I have pieces of notes everywhere and it's absolute chaos.

I have tried various ways to organize to help my memory and to sort through relationships etc. I haven't found a system I am happy with. Each one seems to require me to keep some pieces somewhere else.

Part of my problem seems to be that the readings are very dense but link to all kinds of other ideas, are replies to this, use these assumptions, etc. Whatever I am doing now is completely inefficient. And all help that comes close is aimed at businesses, projects, etc.

I tracked down this site on Argumentation Schemes, but that seems to be focused rather on scientific or popular arguments based on evidence. I need to be able to find the logical structure, hidden assumptions, etc.

So if you have any ideas, or can point me to resources, I'd be glad.

* * *

Helga,

First off, it's great you are looking to use Compendium in this way. I think it's a perfect fit.

Having said that, it's a little tough to give you hard and fast examples and pointers -- kind of like trying to advise someone how to write. Although it provides many aids to constructing maps, Compendium is essentially a free-form creative tool that you can use in any way you want (as opposed to a more structured program that prescribes how to do things).

So the way you use it for the purposes you describe is limited, really, only by your imagination. Personally I do all sorts of things, ranging from very informal to very formal. Some use very careful schemes of tags, link/node types, etc.; some are more stream of consciousness. Some evolve from one form to another, or mix up lots of things.

My preferences for using Compendium are for subject matter that falls into a few broad areas (that often overlap with each other): 1) material where the same ideas/concepts/images etc. might be meaningful in many different contexts; 2) material that I am going to reuse in different ways but that I want to keep 'together' in one place; 3) material that I want to put in a searchable repository even if I am not yet sure of the ways I might want to use it; 4) material that is going to be used or shared by groups of people trying to make sense of a situation; 5) material when I want a visual 'picture' instead of, or in addition to, a textual one. There are more areas but those seem like the most relevant to your query.

You could take a look at the examples linked from this page (which was itself generated from my 'PhD' Compendium database), like the items in the Presentations, Literature Notes, and Compendium Experiments sections. You might also get some ideas from http://knowledgeart.blogspot.com. Possibly following some of the links from Simon's Hypermedia Discourse page, or looking at some of the items in the Compendium Institute Showcase. Still another source for ideas would be the OpenLearn Compendium work.

Some other researchers in the Compendium community, like Simon and Mark Aakhus of Rutgers, could have some more ideas. Some of Simon's non-Compendium (or partially Compendium) work in hypermedia tools for scholarly discourse might also be of interest to you (e.g. Claimmaker and the forthcoming Cohere).

* * *

Al,

You are correct that I will need to find my own ideal way of using it. But I'd like to startup/general structure ideas. I have already used almost 4 full days researching this. And my PhD advisors don't like that sort of thing at all. They want traditional writing (I showed them a Cmap I made - and there were just blank stares). Since they want writing from me soon, I need to start efficiently and then tweak as I go along. So some questions are

1. Do I do one map for every paper I read?

2. If I compare 2 papers, do I first do 2 separate maps?

3. What I adore about Compendium is transclusion since my main research interest is connections between ideas. Is there a way to indicate which one is the original thought? or at least include a ref to the source?

4. The most important question. Once I put the time into this, how do I get a quick write up from it. The expectation is that I have a short write-up on anything I read. In addition, I need to do a MAJOR presentation at the MAIN philosophy conference for philosophy in the southern hemisphere. I have about 2 months to do that.

In other words, what I would really like help on is how to get started as fast and as efficiently as possible, and (as you have provided already) encouragement that it's worth the initial extra effort.

* * *

Helga,

> 1. Do I do one map for every paper I read?

That is generally what I do.


> 2. If I compare 2 papers, do I first do 2 separate maps?

I think that is a good practice. That way you can have maps that talk about any ideas you have about the paper on its own, and a map devoted to connections/contrasts etc. between two (or more). As you note below, with transclusions you can keep track of how, say, a quotation appears in the original paper (and any ideas you link to it there), then how the same quotation relates to others from other papers elsewhere. I do that kind of thing frequently.


> 3. What I adore about Compendium is transclusion since my main research
> interest is connections between ideas. Is there a way to indicate which
> one is the original thought? or at least include a ref to the source?

Transclusions are the heart and soul of Compendium, so you've come to the right place :-)

That is a good question about indicating which one is the original. At present there is no official way to indicate the original appearance of a node. However showing reference is easy, that is one of the main purpose of a reference node (especially, though not only, if you can point to a URL; then you can just click on that node to be taken to the original source appearance, which is quite nice). Alternatively you can put bibliographic detail in the contents (Detail) of the node. If you want to indicate which was the map where you first put a node, what I might do is link a Note node to it saying something like "original appearance of this node" or similar. You may not have noticed (and it isn't the greatest implementation yet) but there is a "Linking Info" button on the Views tab of every node, that shows what is linked to that node in each view it appears in. That way you can see at a glance which view had the "original appearance" item linked to the node of interest.


> 4. The most important question. Once I put the time into this, how do I
> get a quick write up from it. The expectation is that I have a short
> write-up on anything I read. In addition, I need to do a MAJOR
> presentation at the MAIN philosophy conference for philosophy in the
> southern hemisphere. I have about 2 months to do that.

Well. That is more a question of style than anything else -- i.e., how/what you write in a node and how you link them. There is no limit on how much text you put in the Detail pages of a node, so in theory you could write a whole paper in a single node (though without much formatting unless you embed HTML tags in the text, as I sometimes do). This approach might also appease your faculty and other interlocutors -- you can point to the "real" paper but also the extras that normal papers don't have -- your notes, connections to other ideas not in the particular paper, etc.

I also will structure some maps so that they follow a conventional outline format, then use them as the seed for a paper/write-up in Word or similar. For me personally, I do most of my writing first in longhand (I find I think better that way), so I don't really use Compendium as a writing source per se. Presentations are a different matter. For those, when I use Compendium as my presentation vehicle (as in, for example, this presentation), I sometimes do a great deal of visual shaping work within Compendium itself to create good presentation appearance and flow. See Maarten Sierhuis's presentation on the Mobile Agents work for another approach, where he uses Powerpoint slides as the background for Compendium maps in his presentations, so that he combines the virtues of both tools.

* * *

Al,

thanks for all your advice. At the moment I have one more clarifying question. When you do the mindmap for each paper, you don't just summarize the paper but already evaluate/interpret it and link it to other ideas?

* * *

Al,

I lied ;) I have more questions. Can I do a quick summary on the map directly or does everything have to be in some node? What's the best way to indicate that several authors are in the same camp, so to speak? Maybe in a separate map for authors? which then links to the maps of their papers?

* * *

Helga,

It depends. Sometimes a map is just a place I throw interesting quotations that I might use later; sometimes I do a summary; sometimes I throw in points of connection to other work or related thoughts.

Here is an example of a map where I actually created a summary of sorts. Here is one that has quotes and some thoughts in a more free-form way.

It really depends on what you are making a particular map for -- yourself, others, etc.

> Can I do a quick summary on the map directly or does everything have to
> be in some node?

I'm not sure what you mean by doing it on the map directly. There isn't a good way of typing onto the map background itself at present (though that's an interesting idea). So for now everything has to be in either the Label and/or Detail of one or more nodes.

> What's the best way to indicate that several authors are in the same
> camp, so to speak? Maybe in a separate map for authors? which then links
> to the maps of their papers?

That sounds good :-)

Not to sound like a broken record, but it really depends on the context and what you are trying to do (as well as your own personal preferences / style, as well as any constraints or expectations from others.

I often have maps that contain just a few nodes from a few sources, that are related to each other for some local reason, but those few nodes are transcluded in other places where they might have all sorts of other links. And there are maps that are just little tables of contents (of sorts) to other maps. For example, I have top-level maps for authors that I might have notes on several different works (Dewey, Schon, Aakhus, etc.), that link to individual maps for the different works.

-----------
* quoted with permission (thanks Helga).