Great discussion! I'm also pursuing a PhD and am an inveterate outliner (and "CRIMP-er"), so I can relate to all the points raised so far. I've been thinking about a lot of the same questions, but, not surprisingly, I have yet to come up with a clear solution. A few disjointed thoughts on where I'm at:

-- For references, I keep coming back to Zotero, which I love, even although it falls short in terms of analytics and note management (no hierarchical outlining). Citavi has some great features, but Zotero fits my style better, and I like that it is cross-platform and now fully web-accessible. Also, I have more confidence that Zotero will continue and I'll generally have an easier time getting my references out of it in the future. (Plus more possibilities for sync with hand-helds and the like.)

-- For notes, I seem to keep coming back to Tinderbox, although sometimes I find myself switching to OneNote because I want to draw freehand diagrams, add photos/scans of handwritten notes, and so forth. Tinderbox has powerful analytics capabilities, although once I'm done my research (see below), I may try out some of the dedicated analytics packages as well. My guess is that Tinderbox will do the trick though.

-- Right now, I'm doing anthropological field research, so I've been thinking a lot about various methods of capture. I've tried Dragon for transcription of audio notes and OneNote for recognition of handwriting, and it looks like I'll be receiving a LiveScribe pen over the holidays. Right now everything is still a mess.

Concluding thought: I must confess I love OneNotes intrinsic outlining ("one-pane" style, so to speak) and its powerful incorporation of media, handwriting and drawing. I would love to see some sort of software that had base like OneNote but that also included good reference management and powerful saved-searching and filtering. In the long run, although I see Tinderbox and InfoQube as contenders, I actually see Zotero as the most likely candidate in terms of something that could mature into such an ideal research tool. But as many here have pointed out in the past, a mix of dedicated tools can also be preferable, so I'll just keep bumbling along for now.

(Sorry for the not very well-wrought post.)

Hi again,

@MontrealFan: why are you doing everything through Zotero, instead of using UltraRecall (or similar) and then leveraging the capabilities of Zotero through UR?

@Lucas: if you use Zotero, again UltraRecall is an ideal platform for outlining, since Zotero references can be integrated into Word embedded files inside UR. TreeProjects can probably also do this quite well, although I have not investigated if export of multiple documents to a merged document is possible.

@people taking handwritten notes: I agree! That's why I am using the Livescribe Echo, and really it is fantastic. Don't expect 100% OCR with MyScript, OneNote or Evernote, but it's good enough and then you have the original copy to refer to when in doubt. For me, the main "problem" with handwritten notes is misplacing them or not filing them, as well as double work to scan or enter them into a permanent record. Hence, the brilliance of the Livescribe solution.

@Dr Andus: I forgot another main reason that I think UltraRecall is idea for your setup: they are planning to integrate a Python API, which means that datamining of your database should soon become possible. As an academic, I am actually planning to write some code to do this as soon as they release (stated to be early 2012). Right now, I have templates (forms/attributes) holding the key information I plan to data mine later, so that I can add this capability to exisiting information. Now it is possible to export the data by "Copy grid values", but I prefer the integrated approach.

Best, /Pavi

@Lucas - I'm curious about your set up. Since you use Tinderbox, you have to be using a Mac. Do you use OneNote with VMWare Fusion or Parallels, or do you use different machines? Also, have you taken a look at Curio? It can do a lot of what OneNote can, and some things OneNote can't or doesn't do as well. Specifically, diagramming is more powerful in Curio. The table feature is better and you can create "index cards." The main thing that Curio can't do that OneNote can is OCR text from scans or photos. However, Curio does work very well with Evernote -- you can embed an Evernote note right in Curio -- so there must be a way to leverage Evernote's OCR capability. I'm not urging you to change what works for you, just wondering if you've investigated this.

As for this discussion in general, very interesting. There is clearly no right way. But one thing that we can be sure of, whatever works best today, won't be the best way three or four years from now. I do some non-fiction writing which requires research, but nothing coming close to the magnitude of the scholars here. Even so, I stumble around quite a bit. I keep references in Bookends, and compile a library of the notes from those references in DevonThink. From DT I pull information as needed into Scrivener perhaps stopping first in Tinderbox if I need to really make sense of the information before going to Scrivener for writing. This kind of work is more my hobby, and I'm dealing with two or three dozen references, not thousands like you all.

Steve Z.

Pavi wrote:

@MontrealFan: why are you doing everything through Zotero, instead of
using UltraRecall (or similar) and then leveraging the capabilities of Zotero
through UR?

@Lucas: if you use Zotero, again UltraRecall is an ideal platform for
outlining, since Zotero references can be integrated into Word embedded files
inside UR. TreeProjects can probably also do this quite well, although I have not
investigated if export of multiple documents to a merged document is
possible.

@Dr Andus: I forgot another main
reason that I think UltraRecall is idea for your setup: they are planning to integrate
a Python API, which means that datamining of your database should soon become
possible. As an academic, I am actually planning to write some code to do this as soon as
they release (stated to be early 2012). Right now, I have templates
(forms/attributes) holding the key information I plan to data mine later, so that I
can add this capability to exisiting information. Now it is possible to export the
data by "Copy grid values", but I prefer the integrated approach.

Best, /Pavi

Thanks Pavi, I will take another look at your original UR/Zotero post. Actually one reason why I never got into UR is because I discovered NVivo and that turned out to be a very powerful way of bringing all my research data together and analysing it. So I think NVivo is also a worthy contender for the second stage of analysing and organising reading notes (@MontrealFan).

The problem with NVivo is that it is very expensive. Also, it is a resource-hog, similarly to Dragon, and quite slow. But the analytical capabilities are superb and apparently v. 9 (which I don't have) can handle PDFs and link with EndNote. Also, NVivo have some fabulous tutorial videos, so one can learn it in a day or two, while I admit that I still haven't got my head around UR completely.

So it sounds like to get a handle on dealing with reading notes for a huge project, at the moment one needs a cross-platform solution for capturing notes (e.g. Evernote) + a powerful database and analytical software (UR, Nvivo, MyInfo) + a reference manager (Zotero, EndNote).

Hi,

Regarding NVivo, I am having a hard time understanding how it works, aside from "analyzing patterns in data". Their website is not straightforward, which is unfortunate given the high price of the product.

Regarding Curio, this product looks amazing, which is unfortunate since I am not a Mac guy... I miss the freeform whitespace of OneNote when using UltraRecall, and am hoping that they improve data input (as is on the roadmap).

Best, /Pavi

Pavi wrote:

Regarding NVivo, I am having a hard time understanding how it works, aside from
"analyzing patterns in data". Their website is not straightforward, which is
unfortunate given the high price of the product.

As I said, I found the tutorial videos very helpful. Here are the videos for v. 9 (though I watched the ones for v. 8):
http://www.qsrinternational.com/support_tutorials.aspx?productid=20

@Stephen, thanks very much for your thoughts:

@Lucas - I'm curious about your set up. Since you use Tinderbox, you have to be using a
Mac. Do you use OneNote with VMWare Fusion or Parallels, or do you use different
machines?

I use a MacBook Pro running Parallels. Parallels works brilliantly, but the only issue is that it needs a whole bunch of RAM. I think that's fair enough given all the work that it does, but I find myself wishing I had more than 4 GB RAM in my machine.

On the other hand, my brother uses two machines (one Windows, one Mac) but just one mouse and keyboard via the free Synergy software. (synergy-foss.org)

Also, have you taken a look at Curio? It can do a lot of what OneNote can, and
some things OneNote can't or doesn't do as well. Specifically, diagramming is more
powerful in Curio. The table feature is better and you can create "index cards." The
main thing that Curio can't do that OneNote can is OCR text from scans or photos.
However, Curio does work very well with Evernote -- you can embed an Evernote note
right in Curio -- so there must be a way to leverage Evernote's OCR capability.

Great point --- I like Curio (still haven't bought it), but I had forgotten about the Evernote possibilities. The problem is that I'm a stickler for metadata. OneNote barely cuts it in this regard, but at least it records a specific time stamp for every line of text, so that if I go back and edit some fieldnotes, I can keep track of when I wrote what. And the tagging features of OneNote provide further metadata possibilities --- nothing like Tinderbox or InfoQube, but at least some basics. Whereas Curio seems to be too limited in this regard. (The same goes for the otherwise remarkably capable GrowlyNotes.)

(Interesting that you're using DevonThink and Bookends. They're both high on my list for further trying out. It seems that getting one's data from Zotero to Bookends is a bit of a process, but should be possible either via the free Mendeley and/or the free trial of Sente.)

@Lucas, just for the record, Curio has a fairly powerful meta data scheme, especially for an application like this. You can assign tags, flags, start and due dates and even resources to any individual item (line of text, diagram node, etc...). It does not (at least not that I could see) assign a created date automatically. Another handy feature is that a diagram can be converted instantly to an outline and vice versa.

BTW, the other way that OneNote is better is in its ability to share notebooks and collaborate on entries.

Steve Z.

@Steve, thanks, you're right! I do have an occasional bad habit of making rushed judgments on software and commenting on that basis. Although I only have an expired trial of Curio, I'm now having a look at the saved search functionality, and I like what I see. Perhaps I'll contact Curio support to see about a new trial. Thanks again.

MontrealFan wrote:

All of them have great features, but none of them beat Zotero for collecting sources (including a huge collection of existing pdfs) and importing
the citation information easily. I am hoping that this new way of taking notes will pay off for the ?second phase? that Dr. Andrus was talking about.
I would love to hear more about what workflow others are using for this second phase!

I have thought about this a bit more and I'm coming to the conclusion that probably NVivo 9 is the most sophisticated tool out there for analysing a large amount of reading notes. Apparently it's possible now to import your EndNote or Zotero references with any attached PDFs and notes straight into NVivo. Of course it's also possible to import regular text, doc or RTF files as well. Once in NVivo, there are a host of analytical features to classify and link items together and extract the essential stuff. This video (after the 9 min mark) explains how to do it:
http://www.youtube.com/user/QSRInternational#p/u/0/HDMxcHpMSOs

I have now come up with some answers to my original problems that kicked off this thread. I've chosen ConnectedText as the central database for all of my notes, incl. reading notes. This gave rise to 3 tasks:

1) to consolidate all existing notes scattered in various hard copies and electronic copies in a variety of software by moving them into CT.

2) to set up a cross-platform system for capturing new notes and quotes while I read both electronic and hard copy documents.

3) to set up a cross-platform system for capturing and managing to-do lists for reading.

Here is a concept map I created in Amode to evaluate the various options, in response to the above:
http://dl.dropbox.com/u/428516/Reading_system.png

1) is easy, mostly a copy-and-paste job, though importing can also be done (but I just want to paste in the good stuff, not to import everything).

2) I will read and annotate PDF books and articles in iPad using GoodReader or PDF Expert, which can export highlighted text and notes as a text file into an email, from which I can paste into CT. For hard copy books I will use iPad to type quotes and comments using Nebulous Notes and sync to Dropbox, from where it can be pasted or imported into CT.

3) I will use CarbonFin Outliner for managing the to-do list for the reading, so I can have it on the desktop (in the browser), and in iPad and iPod. It's important that I can freely prioritise this list, move items around and create hierarchies of themes.

As you can see from the concept map, another option for hard copy books is to take handwritten notes and then dictate them into Dragon or to take notes with a digital recorder directly and then transcribe it then Dragon, from which it goes into CT (copy and paste). I might trial it still but my concern is that even at 99% accuracy I can't have any mistakes in quotes, and because of that 1% risk I'd have to compare the notes with the original, which is a waste of time. Plus the transcription process is a bit convoluted and involves some file management, taking too many steps.

I have abandoned the idea of using a scanner pen because it wouldn't scan my own comments, plus the files still need to be transferred manually into the PC, which is time-consuming. I also haven't listed the option of taking photos of quotes and handwritten notes (e.g. with iPod Touch and then adding them into Evernote) because the image doesn't identify clearly enough which part of the text is the focus, and converting it into text for CT would require some additional manipulation.

For 3) I have also considered using iPad's own Notes, Toodledo, and Simplenote but CarbonFin came out on top for what I wanted.

Dr Andus wrote:

3) to set up a
cross-platform system for capturing and managing to-do lists for reading.

For 3) I
have also considered using iPad's own Notes, Toodledo, and Simplenote but CarbonFin
came out on top for what I wanted.

Another nice thing about CarbonFin Outliner that it can now also sync with Dropbox, exporting .OPML files.

Dr Andus wrote:

Here is
a concept map I created in Amode to evaluate the various options, in response to the
above:
http://dl.dropbox.com/u/428516/Reading_system.png

Love it! I especially appreciate the fact that you can both integrate your pevious work, and expand the system with contemporary tools.

Connected Text seems like the ideal repository. Personally, my only complain is the editor and markup, which I don't find conducive to focused writing, but that's more as a subject of personal preference.

Markdown support would be appreciated, but I hope than eventually, such repositories would be modular: I envisage being able to right click on any text within CT and select "Edit with Sense" or "Edit with Brainstorm". I would then have all the benefits of a dedicated writing environment within the repository. For the moment it's copy and paste.

Alexander Deliyannis wrote:

Dr Andus wrote:
>http://dl.dropbox.com/u/428516/Reading_system.png

Love it! I
especially appreciate the fact that you can both integrate your pevious work, and
expand the system with contemporary tools.

Thanks, Alexander, glad to hear that. This exercise has also revealed to me what a huge state of mess my notes had been up to now, as a legacy of CRIMPing and not having a focused enough overall system. E.g. even my to-do notes for reading have been scattered in 3 email accounts, 3 or more to-do apps, calendars and other places. It's a huge relief to be able to consolidate them into a single list in CarbonFin and be able to prioritise them. The same goes for CT. But from now on I will have to be disciplined to stick with this system.

Dr Andus wrote:

But from now on I will have to be disciplined to stick with this system.

We here at CA (CRIMPers Anonymous) are with you, brother, to support you in this virtuous cause.

Alexander Deliyannis wrote:

Dr Andus wrote:
>But from now on I will have to be disciplined to stick with this
system.

We here at CA (CRIMPers Anonymous) are with you, brother, to support you in
this virtuous cause.

Thank you. Yes, I'm a PIMoholic. But this is a strange support group, as I also come here to get my fix. :)

On a different note, I don't know how profound an observation this is but I just realised that organising is about making (recurring) decisions about what to keep together and what to separate (or rather the principles behind the decisions). In this case I have realised I was wrong to separate my data and notes into 20 different PIMs.

On the other hand I have made the conscious decision to keep my academic research project entirely separate from my day job and personal life, so I use different to-do apps and calendars for those (Amode for the former, Google Cal+Toodledo for the latter). However, when it comes to data and notes, even day job and personal info now goes into CT (although into separate "projects,", i.e. databases).

Gary Carson wrote:

I think the fastest and easiest way to take notes from books, newspapers, etc., is to
use a digital voice recorder and then transcribe your audio notes with Dragon
Naturally Speaking.

I've been wanting to do this for a long time but what's been stopping me was that this method would still require me to compare the transcribed text with the original text using the PC, to make sure that quotes are correct, which would be very time-consuming with let's say an hour-long recording of reading notes on a long text (e.g. a book).

However, I've just tried to use the Dragon app on my iPod Touch (4th gen.) again, and to my amazement the latest version worked really well. Here is my workflow:

1. Sit in a quiet room in a comfy armchair with my book and my iPod. When I find a quote I want to keep (or make a comment about the reading), I tap the Dragon Dictation app and dictate my quote/comment. There is a word limit but it can handle about a half a page, which is usually enough.

2. Copy text to the clipboard.

3. Switch to Nebulous Notes and paste it into a doc I created for this book. Now I compare the transcript with the original and correct any mistakes. Accuracy is lower than in the desktop version of Dragon but I'd say it's about 90-95% (as opposed to 99%). I can definitely live with that, as it saves an awful lot of typing.

4. Repeat process as many times as necessary.

5. Sync Nebulous Notes with Dropbox.

6. Import text document into ConnectedText on the PC. Done!

Before this I was doing this at my PC sitting in my office chair and dictating my reading notes and quotes directly into desktop Dragon, which is a lot less comfortable and took a toll on my back (not to mention the temptation to switch to a browser and do something else instead).

Don't be fooled by the low (3-star) rating of the app and the negative reviews. They must be people who have never tried the desktop Dragon and don't realise that this is not the Star Trek Communicator yet, here you need to adapt to the machine and learn how to dictate properly.

Anyway, this is a major development in my own note-taking system for reading hard copy books and articles. For electronic copies it's easy: it's the iPad and GoodReader.

Dr Andus wrote:

Gary Carson wrote:
>I think the fastest and easiest way to take notes from books,
newspapers, etc., is to
>use a digital voice recorder and then transcribe your audio
notes with Dragon
>Naturally Speaking.

1. Sit in a quiet room in a comfy armchair with my
book and my iPod. When I find a quote I want to keep (or make a comment about the reading),
I tap the Dragon Dictation app and dictate my quote/comment. There is a word limit but
it can handle about a half a page, which is usually enough.

I should have mentioned that this iPod Touch system would only work in a room with wifi (unless you have an iPhone), as Dragon Dictation requires an internet connection.

Dr Andus wrote:

Dragon Dictation requires an internet connection.

I wonder why?

I would assume that the recognition happens in the cloud (as OCR does in Evernote), but in that case the success rate shouldn't be lower for the mobile version of the software.

Alexander Deliyannis wrote:

Dr Andus wrote:
>Dragon Dictation requires an internet connection.

I wonder why?

I would assume that the recognition happens in the cloud (as OCR does in Evernote),

Yes.

but in that case the success rate shouldn't be lower for the mobile version of the
software.

I think the reason for that is that you can train the desktop version to recognise your voice (and probably the mic is of higher quality), while there is no such training involved with the app (and the mic is not as good). It's actually amazing it works as well as it does without the training. Thick accents though may get in the way.

But my hope is still to be able to find a scanner app with OCR that would allow me to extract specific lines of an image snapped with the iPod and transfer it easily to a text file.

I've trialled a new system and workflow for taking reading notes today. Props: 1) arm chair + foot rest, 2) a board to be used as a make-shift desk surface to sit on my lap, 3) book stand, 4) 2x extra-large foldback binder clips to secure two sides of open book to book stand, 5) pencil for underlining, 6) iPad with iThoughtsHD and Dragon Dictation apps.

iThoughtsHD (a mind map) is the key player here. It allows me to organise my reading notes into a mind map (i.e. hierarchical outline) while I'm reading, which I can then sync via Dropbox with my PC, where I can open the .mm file directly in Freeplane.

Dragon Dictation is there only for the occasional longer quote, which I can paste in as a note to a node. The pencil is there mostly to underline passages that I want to type up in iThoughts (so my eyes can follow the text - but also useful for future reference).

Previously I took notes with pen and paper in a notebook and underlined quotes in the book, and then I used Dragon on the PC to dictate them, and pasted the text into ConnectedText. Only at the end did I organise the text into a hierarchy by adding headings in CT. Then I might select quotes I want to use and put them into separate CT topics.

This new process reverses the place of hierarchical organisation in the workflow. I find that it works better, as it's easier to discern a logic in the text while I'm reading it. Also, typing on the iPad seems to go faster than writing with pencil and paper (and there are just fewer objects to manage on my "desk" - they used to keep falling off).

My dilemma is how to get the iThoughts data into CT. It is possible to import the mind maps into CT as OPML, .mm or even HTML, either as outlines or as CT topics. The problem is that CT's topic can only render 5 levels of a hierarchy, while the outliner can't import notes attached to nodes.

I may just keep my reading notes hierarchy in Freeplane for now and link to it from CT. Additionally I could paste in any longer quotes into CT from iThought's .docx export and then link the Freeplane nodes to the respective CT topics. I still need to figure this bit out. Perhaps the Freeplane map can be the first stage of filtering out important material (mainly quotes) that need to go into CT.

But I like taking reading notes in mind map form upfront. It seems to help my comprehension. Effectively I'm creating an annotated reverse outline of the book. iThoughtsHD is just excellent for this.

I'm wondering how I could use the upfront mind map reading notes process while reading digital documents.

1) I could have PDF journal article displayed on PC screen, while taking notes in iThoughtsHD on iPad. But retyping text, especially long quotes feels like unnecessary duplication.

2) I could copy and paste notes and quotes straight to a Freeplane mind map.

The problem with both is that I'd need to sit at the desk, which is not comfortable when reading an involved scientific article all day long.

3) I could continue with my current process (reading and annotating PDFs in GoodReader on iPad in armchair). But instead of pasting the annotations and quotes straight into ConnectedText from GoodReader's email export, I could use Docear (which is built on Freeplane) to extract PDF notes and highlights and organise them into a mindmap ASAP, before I forget the hierarchical and logical relationships. Then I can link things to and from CT and add selected quotes. Need to test this.

Just came across Dragon Notes by Nuance for instant transcription of audio notes, which seems to have been specifically designed for Win8 tablets but also works on Win7 PCs. I already have Dragon NaturallySpeaking Premium, so this appears to be redundant.

http://www.nuance.com/dragon/dragon-notes/index.htm

The underlying concept seems interesting but it doesn't sound like they've thought through what happens once you've got masses of notes (according to the review below). For me, speech recognition still hasn't arrived to supplant the keyboard when it comes to quick note-taking (although I do use Dragon NS12 for dictating longer texts or recording reading notes).

http://www.pcadvisor.co.uk/reviews/software/3453257/dragon-notes-review/

TextGrabber on iOS seems to take us a step closer to taking reading notes with a camera:

http://www.macdrifter.com/2013/09/textgrabber-for-ios-update.html

Looks interesting (grrrr - no, not a CRIMPer moment, please, really can't afford any more!). I use Snap2PDF for OCRing photos of docs, which also works pretty well.