Stephen,
if the keyboard is bad, don't use it. If the Interface is bad, don't use the application either. But perhaps there is a difference between how something feels and how something really is. Even if a tool does not at first "feel" right, it is not necessarily bad. It maybe you (or me) that is the to blame, not the tool.

In other words, sometimes you need to change the tool and sometimes you need to change yourself. The latter is harder, I know. But this fact does not mean that it might not be worth it.

Life's full of choices like that. Far be it from me to suggest that everyone should follow my example. It's just that a do have a definite view on some things (including on the usefulness and functionality of ConnectedText), which you might (or might not) want to take into consideration.
Manfred

Gee, when I did my Ph.D. dissertation (mathematics) there was no software to use. I did 99% of my research while lying in bed until 4am each night. I also typed most of the draft in bed using an Olivetti light-weight typewriter with two exchangeable keys (for the math symbols). I guess I'd do the same today, but with a light-weight laptop in place of the Olivetti.
-c

Cassius,

As I said: "What does aesthetics have to do with researching and writing? When I wrote my dissertation I really wanted an IBM Selectric, but not because it was available in different (attractive) colors, but because its correction system was better than anything else available."
Manfred

P.S.: Perhaps there should be a thread on the "Aesthetics of Outliners and Pims

Hi,

I'm also a PhD student, with a similar interest in designing a work flow that is supported by appropriate software tools. Unfortunately I'm still far from a fully integrated and streamlined work flow and each time I try, I do it slightly differently. However, so far I have assembled the following tools to support this process:

- EndNote to store and categorise references and link to PDFs;
- Adobe Acrobat and now PDF-Xchange Viewer (thanks guys!) for annotating PDFs;
- Surfulater for capturing website content and organising it into categories;
- Atlas.ti to analyse data (I haven't got to that stage yet);
- Whizfolder for organising ideas and reading notes into hierarchical categories, capturing snippets from PDFs (using the excellent "Watching Clipboard" function), linking to PDFs and any other file with research data, and any ad hoc writing and outlining;
- CmapTools for quick conceptual mapping;
- Natara Bonsai for developing hierarchical outlines to be used for writing the final draft (building on material captured and organised in Whizfolder and the conceptual maps in CmapTools);
- MS Word for writing.

Out of all of these, I am really happy with and couldn't or wouldn't want to live without Surfulater, Whizfolder, CmapTools and Bonsai. I'm hoping that Whizfolder will become the nerve centre of the entire dissertation, although I haven't fully utilised it yet, as I'm only just moving into the data analysis stage of my PhD. Still, for some unexplainable reason, I find working with Whizfolder strangely satisfying...

I've tried out too many mindmapping, notetaking and wiki tools to remember them all. The above are what I ended up with after 3 years of searching and experimentation. I have to say I'm not very patient, so if a software doesn't immediately make sense, I quickly move on. I'm intrigued for instance by Ultra Recall but I just couldn't figure out what it's really for and how I could use it as part of the research process.

Anyway, I'd be also interested to hear about how others map their qualitative research analysis and writing processes.

Hi Peter,

Your work flow descriptions are so helpful to my Word addin project! Thank you so much!

Maybe this is a dump question, but may I ask, by "creating new drafts every week or so and end up with loads of file versions scattered across multiple folders", do you mean that it is a consequence of that fact that you have to save previous versions the files? Thank you.

--
Writing Outliner - Word addin for writers.
http://WritingOutliner.com

No worries Edwin, and happy to hear that you find it useful.

To answer your (not dumb) question: In the process of brainstorming and catching ideas I generally open a new doc and try to stick with it. However, sometimes the ideas get too cluttered, or they branch off and start to develop into something distinctly different. That is when I like to have the experience of a clean slate. Sometimes I even create one or two quick notepad txt files because it's so much simpler than starting up Word. Initially I save these new files on the desktop rather down in some folder under MyDocs because I'm not sure how they will all fit together or how I want to categorize them. Then, over the course of several days or sometimes weeks, I go back through these docs and 'weed' them and cleaning them off my desktop. Sometimes this involves cutting and pasting from several into one and then deleting the originals. Sometimes this involves collecting several doc/txt files together under a new theme and then creating a new folder with that theme's name. Sometimes I try to put the desktop files into preexisting folders under MyDocs. Sometimes I make the new folders a subfolder of a preexisting folder under MyDoc. The upside of this workflow is the level of spontaneity it allows and the on-the-fly recording of ideas. The downside is the clutter and often disorder.

I hope this answers your (not dumb) question and good luck with your project! ;)

Edwin wrote:

Hi Peter,

Your work flow descriptions are so helpful to my Word addin project! Thank
you so much!

Maybe this is a dump question, but may I ask, by "creating new drafts
every week or so and end up with loads of file versions scattered across multiple
folders", do you mean that it is a consequence of that fact that you have to save
previous versions the files? Thank you.

--
Writing Outliner - Word addin for
writers.
http://WritingOutliner.com

Hi Dr Andus!

Thanks for your comments on work flow and I think it is a very useful discussion (although it perhaps should be broken out like the 'aesthetics' one). A few comments / questions...

I'm wondering if you've even compared Surfulater with Zotero's ability to grab web material and turn it into bib source refs? (I think Evernote and/or Mendeley do this as well but I forget which one just now.)

Also, I've never used Bonsai or Whizfolder but they look like identical outliners to me. I'm curious why you start with Whizfolder and then move everything over to Bonsai. Why not just stick with one or the other?


Dr Andus wrote:

Hi,

I'm also a PhD student, with a similar interest in designing a work flow that is
supported by appropriate software tools...

I'm wondering if you've even compared Surfulater with
Zotero's ability to grab web material and turn it into bib source refs? (I think
Evernote and/or Mendeley do this as well but I forget which one just now.)

I've only just recently discovered that Zotero can take screenshots of websites, however that doesn't want me to switch over to Zotero. For my research I need to capture hundreds of websites and organise them into categories. Sometimes I need a snapshot of the whole website, sometimes just an image or a snippet of text, and Surfulater can handle that very well. I like the fact that my websites (which for me are my primary research data) are separated from my academic references (secondary research data) in EndNote. The only thing I use Zotero for is to capture bibliographic reference off websites like Amazon, which I then transfer into EndNote. Which is not to say that I'm not impressed by Zotero, it's just that I've been locked into EndNote before Zotero came about and the current set-up works for me.

Also, I've
never used Bonsai or Whizfolder but they look like identical outliners to me. I'm
curious why you start with Whizfolder and then move everything over to Bonsai. Why not
just stick with one or the other?

Bonsai is more focused on the task of outlining, which it does beautifully, I absolutely love it. It is excellent for constructing the overall outline of a text, as it allows you to create hierarchies, colour-code the various levels, collapse them, move them around, zoom in and out of branches. So one can construct really complex outlines and then follow them while writing. One could use Bonsai at any stage of the research process but I prefer to use it at the very end, to construct the very final clean outline, which I then use for writing the paper in Word. So I don't move over everything from Whizfolder. Rather, I use Bonsai for extracting the clean ideas from the jungle that is stored in Whizfolder.

Whizfolder has different functionalities when it comes to outlining. The key difference is that Whizfolder allows you to develop each topic (item) within an outline into a piece of text, while there is no convenient way to do that in Bonsai (which is not a gripe, I like Bonsai's simplicity as it is). In Bonsai, you just have the topics. Bonsai is good for creating the skeleton, while Whizfolder allows you to add the meat. I happen to use the two 'inductively', i.e. gather the data and analyse it by developing text in Whizfolder and then extract the outline at the end into Bonsai.

But this combination could be used 'deductively' as well, i.e. by developing an outline in Bonsai and then adding the meat in Whizfolder, although it would make a bit less sense, as you could probably do all of that in Whizfolder alone. As I'm doing qualitative research, the 'inductive' approach works for me better. While I love Whizfolder as well for what it can do (great for capturing and organising ad hoc writing snippets and quotes from PDFs and as a general conceptual database of the whole project), the content becomes pretty complex. So it helps to extract the core ideas into Bonsai, which is just so simple and it's so quick and easy to manipulate the elements of the final outline (though I still have version 4).

doctorandus

I am at the final writing stages of my MBA dissertation and I must say that for it, like for several professional projects in the past, information management software has been worth my investment; I don't know how I could have organised my material any other way.

That said, I started with a relatively complex setup, but in the end only the essential tools survived. Nowadays, my setup is similar but simpler than Dr Andus':

- Surfulater for gathering all relevant information; I found most journal articles in HTML form as well as PDF, so I would usually grab the HTML and link to the PDF in my hard drive.

- Brainstorm as my writing environment; I find it indispensable, as others do GrandView, but I must say that at this size and level of complexity of texts, the lack of editable outline view becomes an issue. In any case, I have found nothing better in helping me to focus on the actual writing.

Some notes on workflow:

- I used Evernote to capture material whenever I was away from my main PC; when back, if it was relevant, it would go into Surfulater.

- I built a common outline of keywords (tags) in all three programs, which I found very useful in organising the material consistently. Maintaining this structure across the programs has been the greatest hussle.

- I did purchase an academic license for Endnote when I began my studies, but ultimately I did not use it or any other bibliography manager. I learnt Harvard referencing well and find it much more convenient to keep my references (as titles) along with the rest of the material in Surfulater and Brainstorm. Both programs can sort titles for my reference list; they also support cloning, which is great for keeping references in context in more than one places.

- I find Brainstorm's "namesakes" (the program automatically recognises identical text entries and makes clones out of them) extremely useful to maintain flow in my writing. If something is missing while writing, I can type "citation needed" and go on writing; later I will navigate 'sideways' (left/right arrow jumps through namesakes) through all the "citation needed" entries. I can do the same with keywords. It's very similar to referencing stuff in a wiki, but without the markup.

- Similarly, I find some of the smart touches of Surfulater's latest versions very useful; e.g. highlight of titles is visible in the outline, so I can mark and quickly see what material I've been through.

- For some time I tried UltraRecall because it could index PDFs but found it much less convenient than Surfulater for the full range of material. PDF-Xchange is great for annotating PDFs but mostly I make the annotation in the HTML version.

That's it form my part; I'll post one more comment in the aesthetics thread and the go back to my writing :-)

Dr Andus,

Thanks for describing your setup. The distinction between creating the meat and creating the outline is useful. I actually find it easiest to generate text hierarchically to begin with (in a single-pane outliner like Ecco Pro, OmniOutliner, or Microsoft Word), but the resulting text often becomes unmanageable. I end up generating lots of complex hierarchies of text with various themes reiterated throughout, and ultimately it becomes extremely time-consuming to organize all the text, even though I produced it very quickly. Now that I have begun using ConnectedText, my workflow has changed. Instead of generating text in a hierarchy, I create a new "topic" for each significant idea (like in Ideamason). I can still indicate all the hierarchical relationships among my ideas using the built-in functionality for assigning topic relationships and categories (and this actually allows for richer webs of connection than a straight hierarchy does). The initial task of creating text takes longer, for me, with this method, but the overall time for a writing project becomes significantly shorter, because the text I generate is much more manageable and better organized. (And the dividends pay off even more when one is faced with a new writing project on similar themes, and it turns out half the work is already done.) The next stage with this method is very similar to what you describe doing with Bonsai, only it does not requite a separate program. ConnectedText has a built in Outline module that is designed for arranging existing topics into an outline for export (to be worked on, for instance, in a word processor). Anyway, it seems that the workflow here is somewhat similar to what you accomplish with WhizFolders and Bonsai.

So, what I am learning is that while I love the freedom of generating text in a single-pane outliner, I get better results when I am forced to separate the text I generate into discrete topics from the get-go. And this becomes even more palatable when I still have the freedom to assign complex hierarchical relationships (including bi-directional ones), as in CT. When I used outliners, I was always in search of robust "cloning". But with a wiki like CT, I suddenly have infinite cloning. I can make any topic a parent and/or child of any other topic.

Lucas

Hi Peter,

Thank you for the detailed explanation, I'm glad that I asked this question, now I know the overall process of how you capture, refine, organize and finalize your ideas, It helps a lot to my project indeed :)

--
Edwin Yip
Writing Outliner - Turn Microsoft Word into an all-in-one writing software.
http://WritingOutliner.com

Peter wrote:

No worries Edwin, and happy to hear that you find it useful.

To answer your (not dumb)
question: In the process of brainstorming and catching ideas I generally open a new
doc and try to stick with it. However, sometimes the ideas get too cluttered, or they
branch off and start to develop into something distinctly different. That is when I
like to have the experience of a clean slate. Sometimes I even create one or two quick
notepad txt files because it's so much simpler than starting up Word. Initially I save
these new files on the desktop rather down in some folder under MyDocs because I'm not
sure how they will all fit together or how I want to categorize them. Then, over the
course of several days or sometimes weeks, I go back through these docs and 'weed' them
and cleaning them off my desktop. Sometimes this involves cutting and pasting from
several into one and then deleting the originals. Sometimes this involves
collecting several doc/txt files together under a new theme and then creating a new
folder with that theme's name. Sometimes I try to put the desktop files into
preexisting folders under MyDocs. Sometimes I make the new folders a subfolder of a
preexisting folder under MyDoc. The upside of this workflow is the level of
spontaneity it allows and the on-the-fly recording of ideas. The downside is the
clutter and often disorder.

I hope this answers your (not dumb) question and good
luck with your project! ;)

Hi Lucas,

Many thanks for describing your work process vis-a-vis ConnectedText, it makes a lot of sense. I was hoping to find a tool that can do something like that. I trialled a whole range of wiki tools but in the end none of them ended up as user-friendly as I wished. In fact I did have a look at CT a couple of years ago but eventually I went for Whizfolders. I don't quite remember what it was exactly that swayed my mind. I suspect I found Whizfolders just more intuitive to use (not as steep a learning curve).

Funnily there may have been an aesthetic aspect to it as well. I still can't quite explain it but I get a strange satisfaction out of working with Whizfolders, which seems to be simultaneously an aesthetic and cognitive experience, in the sense that I simply enjoy using it for demanding cognitive work. There is an element of play to it. This is despite the fact that Whizfolders' interface is really way to busy, with too many different coloured buttons, to the extent that I wish I could just get rid of them with one click and just work with a white interface when writing.

I should have also mentioned regarding my work flow that when I write I use two screens. I have Word open on my laptop, while right above that on a stand I have a 22inch screen where I can display my Natara Bonsai outline (or Whizfolder) alongside my EndNote references (and toggle it with a PDF reader whenever I need to go into a referenced text). So there is an advantage sometimes to working with several software, as one can open them on different screens).

doctorandus

Hi Dr Andus,

Windows management is on the windows level, but not the program level, that means standalone windows of the same program can be placed on different screens.

Dr Andus wrote:

So there is an advantage sometimes to working with several
software, as one can open them on different screens).

doctorandus

Believe it or not, Excel and an add-in called Excel Notes works great. Essentially, Excel Notes allows you to create, store, view and search an (unlimited) Word or .rtf document in each cell. This gives you the ability to create an outliner any way that you want, creating columns that you need with unlimited text and graphics for the item. It's easy to set up an outline with columns, if that's what you need


Excel Notes is shareware though giveawayoftheday.com offered it free awhile back and probably will again.

Let me second the endorsement of XLNotes, which is a few days away from version 3.0. Check it out at http://xlnotes.com/eng/ It overcomes the limitations and cumbersomeness of the Comments feature of MS Excel. A very handy software program, indeed.

Regarding ConnectedText I would like to add:

* CT changed a lot in those years. Just this year 7 releases were made, each one adding new features.
* One of the most important features of CT is its query capabilities. It is instantaneous and enables very complex queries. A query can be embedded in anywhere in a topic. There is a learning curve but that is not so steep as it looks.

Lucas wrote:

Dr Andus,

Thanks for describing your setup. The distinction between creating the
meat and creating the outline is useful. I actually find it easiest to generate text
hierarchically to begin with (in a single-pane outliner like Ecco Pro,
OmniOutliner, or Microsoft Word), but the resulting text often becomes
unmanageable. I end up generating lots of complex hierarchies of text with various
themes reiterated throughout, and ultimately it becomes extremely time-consuming
to organize all the text, even though I produced it very quickly. Now that I have begun
using ConnectedText, my workflow has changed. Instead of generating text in a
hierarchy, I create a new "topic" for each significant idea (like in Ideamason). I can
still indicate all the hierarchical relationships among my ideas using the built-in
functionality for assigning topic relationships and categories (and this actually
allows for richer webs of connection than a straight hierarchy does). The initial
task of creating text takes longer, for me, with this method, but the overall time for a
writing project becomes significantly shorter, because the text I generate is much
more manageable and better organized. (And the dividends pay off even more when one is
faced with a new writing project on similar themes, and it turns out half the work is
already done.) The next stage with this method is very similar to what you describe
doing with Bonsai, only it does not requite a separate program. ConnectedText has a
built in Outline module that is designed for arranging existing topics into an
outline for export (to be worked on, for instance, in a word processor). Anyway, it
seems that the workflow here is somewhat similar to what you accomplish with
WhizFolders and Bonsai.

So, what I am learning is that while I love the freedom of
generating text in a single-pane outliner, I get better results when I am forced to
separate the text I generate into discrete topics from the get-go. And this becomes
even more palatable when I still have the freedom to assign complex hierarchical
relationships (including bi-directional ones), as in CT. When I used outliners, I
was always in search of robust "cloning". But with a wiki like CT, I suddenly have
infinite cloning. I can make any topic a parent and/or child of any other
topic.

Lucas

Lucas,
I wish I had paid more attention to what you were saying in this post 3 years ago. I have ended up doing exactly what you describe, for the reasons you describe, but unfortunately it took me 3 years to get to that point. Argh! Anyway, did you stick with CT to the bitter end of your dissertation? I'm curious about your experience.

Dr Andus wrote:

Lucas,
I wish I had paid more attention to what you were
saying in this post 3 years ago. I have ended up doing exactly what you describe, for the
reasons you describe, but unfortunately it took me 3 years to get to that point. Argh!
Anyway, did you stick with CT to the bitter end of your dissertation? I'm curious about
your experience.

Dr Andus,

Thanks for this post. I'm glad to hear that your experience corroborates what I had written 2.5 ago. The funny thing is, I wish *I* had paid more attention to what I had written back then, because I didn't really stick to my own advice. Now I seem to be arriving again at the same conclusion, but in the meantime, I have played around with Infoqube (a little), OneNote (a bit more), and Tinderbox (a lot). All these forays have had merit, I think, and I will continue to use these and other tools for a variety of purposes, but when it comes to the basic task of maintaining my academic knowledge base, I seem to keep coming back to ConnectedText. (Recently I even found a way to convert several months' worth of Tinderbox journaling into CT date topics, and I imagine i'll continue to move data back and forth between these programs to take advantage of their unique capabilities.)

Anyway, I'm currently still doing fieldwork, so I haven't even started writing the actual dissertation yet. I'll keep you posted, and I look forward to continuing to follow your own outlining adventures as well.

Dr Andus wrote:

Lucas,
I wish I had paid more attention to what you were
saying in this post 3 years ago. I have ended up doing exactly what you describe, for the
reasons you describe, but unfortunately it took me 3 years to get to that point. Argh!
Anyway, did you stick with CT to the bitter end of your dissertation? I'm curious about
your experience.

Dr Andus,

Thanks for this post. I'm glad to hear that your experience corroborates what I had written 2.5 ago. The funny thing is, I wish *I* had paid more attention to what I had written back then, because I didn't really stick to my own advice. Now I seem to be arriving again at the same conclusion, but in the meantime, I have played around with Infoqube (a little), OneNote (a bit more), and Tinderbox (a lot). All these forays have had merit, I think, and I will continue to use these and other tools for a variety of purposes, but when it comes to the basic task of maintaining my academic knowledge base, I seem to keep coming back to ConnectedText. (Recently I even found a way to convert several months' worth of Tinderbox journaling into CT date topics, and I imagine i'll continue to move data back and forth between these programs to take advantage of their unique capabilities.)

Anyway, I'm currently still doing fieldwork, so I haven't even started writing the actual dissertation yet. I'll keep you posted, and I look forward to continuing to follow your own outlining adventures as well.