Hi all,

I am interested in your opinions on what the simplest and most efficient set of tools/apps is for a writer's/researcher's workflow. I am sure you get this sort of query all the time but my head is just swimming with all the choices out there. This is a desperate attempt at some final clarity! Here it goes.

I have consulted several excellent blogs on this already, including:

http://www.organizingcreativity.com
http://morepork.home.xs4all.nl/software.html
http://www.gradhacker.org/2011/08/12/going-paperless-from-day-one-digital-academic-workflow-for-grads/
http://jenniferclaro.wordpress.com/2011/08/12/a-simple-academic-workflow/
http://blog.macademic.org/2012/04/22/three-stages-of-the-academic-workflow-and-mac-software/
http://fletcherpenney.net/2012/05/workflow

Here are a few tentative conclusions and articulated needs and questions...

Tagging
To begin, I have a messy system and folder hierarchy that has grown like a cancer over the past several years. Therefore I'm interested in graduating to a more extensive use of doc tags (codes/aliases) to make use of the mac's spotlight search. I have seen Tags ($30) and Tagger (open source) recommended. Does anyone have experience with one or both of these? Also, suggestions for feature/app that could help ease/automate the transition from folders to tags is appreciated. Do either of these apps do this? I have seen Hazel recommended for file org but somehow that just seems like overkill. I need a way to make sense from the mess but still live with it.

Reference Management and PDF annotation
Jennifer Claro makes a good claim for Bibdesk and Skim for bib/pdf management/annotation (see link above). Currently I use the Zotero standalone which opens the pdf in Preview to allow annotation. However Zotero doesn't seem to search/index these annotations. She makes Bibdesk and Skim look pretty good together. Does anyone have experience with these or other set-ups for working with research pdf articles?

Export/Import/Sharing PDFs across apps
It is also important for me that the pdf database "talks" with other apps like DevonThink. See next...

Collecting and Outlining
While I have more experience using bib/citation/ref software I am less versed in outliners. I have never really used more than a simple mindmapper link Freemind, along with Word/Pages and Scrivener. However several of bloggers above make a strong case for Circus Ponies Notebook (CPN). Nevertheless, I am tempted to throw everything (mainly notes, docs, annotations, etc) into Scrivener or Word/Pages and create one big doc, perhaps with import/export to a mind mapper like MindNode for visual outlining. I think I understand the advantage of using a separate outliner like Notebook but at the same time I'm worried that too many apps will slow my flow and my concentration! Do people really think Notebook (and other dedicated outliners) really have a clear advantage? Are there any all-in-one or at least streamlined solutions out there?

If I start using DevonThink, perhaps I could just throw everything onto one of its rich text docs? It is difficult to know how well the latest DevonThink works since many of the reviews are based on earlier versions - pre 2.4 (see below).

Data Management and Idea Development (DevonThink)
While I am still on the fence about CPN for outlining, I am more convinced by DevonThink (I will probably go with Pro or Office). I am concerned however about the pdf reference/citation/annotation issue that people have posted about. There seem to be scripts that allow DevonThink to work with pdfs in an external folder (like Dropbox). I wonder if this works with Zotero/ Bibdesk? I'm not sure I would use dropbox much since I don't work away from my mac for writing (e.g. on an iPad) so I assume a normal folder might also work? I just not clear yet about how to get DevonThink and a ref manager like Zotero to use the same pdf database and also search the annotations. (A screencast tutorial would be great!) Also, what happens if I accidentally more or change the folder? I guess DT wouldn't know where they went?

For the actual writing I will stick to Scrivener/Word/Pages.

Bottom line is that I'm looking for a straight-forward and efficient paperless workflow to jumpstart my thesis writing! It's just a jungle for a novice like me to figure out the best and simplest combination of tools for research and writing! I therefore certainly will appreciate any tips people have to offer! I am looking for an integrated solution, all the way from data collection and analysis to outlining and writing.

Kind regards,
Peter

PS Here are some additional apps that I came across in my search that might interest you, but then I suppose this community already know about them. Do any of these offer serious advantages over a system like Tags/Bibdesk/Skim/CPN/DT/Scrivener/Word/Pages?

Curio www.zengobi.com/products/curio/ - Wessels reviews it
Inspiration 9 http://www.inspiration.com/Inspiration - combines mindmaps, outlines and presentations
The Brain http://www.thebrain.com - a bit clunky and "Windowish" but somewhat innovative
Mindjet's Mind Manager http://www.mindjet.com - expensive but this guy uses it with DT http://vimeo.com/33645437 - any success with other (cheaper) mind-mapping apps?
Qiqqa http://www.qiqqa.com - ref manager for windows on steroids and an intelligent search tool that seems to try to rival DT!
Docear - http://www.docear.org/ - Wessels reviews it

Peter, for what it's worth, I use Mendeley, a free program, to manage my extensive PDF collection. It also stores them in the cloud, giving my backup redundancy to Carbonite and Dropbox, both of which contain my work documents.

Mendeley automatically adds any news PDFs OI have on my system, allows for tagging, etc.

A lot of material I acquire off the web. Non PDF goes into Evernote - which also stores my data in the cloud. EN is good for gathering material, but not much of an info manager apart from its tagging system. It does not even have highlighting of text, which for me is pretty basic. However, I work around it by colouring text which I would otherwise highlight.

From a writer's perspective EN has one feature which for me is a deal maker - and that is that it allows me to have multiple open windows so I can scan docs I'm using for my writing.

I tried Scrivener for Windows but did not find it helpful.

My writing is either in EN, or in MyInfo, which provides me with columns for metadata, etc. MI also allows for multiple open windows. Unfortunately, it only allows for editing in one.

You may wish to look at a Swiss program called Citavi, which looks after references, etc, and may be one of the best of its kind.

Daly

Peter, I should add that Mendeley and Evernote will work on Mac. I am not about Citavi.

Daly

Hi, Peter,

I can't comment much on reference management or PDF annotating, as I don't do much of that (I've been using Papers for that, but I haven't leaned too heavily on it, so don't want to make a recommendation).

I'm also not familiar with tagging software, so my comment may be off base. DevonThink can catalog all the documents in folders on your Mac, and you can use DT for tagging. I guess my question (for others to answer) is why do you need both tagging software AND DevonThink?

As for CircusPonies Notebook... I am not sold on that application as a great place to store lots of information. The notebook metaphor is intriguing, but it gets clumsy with a lot of information. At least that's my experience. I also don't think it is a very good outliner -- serviceable, yes. But not the best option. My favorite outliner is Tinderbox, and it's a great note application; however, you may not want to devote the time and effort into getting comfortable with it. If you want some insight into Tinderbox, I have a series of posts on my blog:

http://welcometosherwood.wordpress.com/tinderbox/

And a Mac Appstorm review:

http://mac.appstorm.net/reviews/productivity-review/taking-the-information-plunge-with-tinderbox/

If you don't anticipate relying heavily on a complex outline, then you should be fine with the outline capabilities of Scrivener.

As for some of the other applications you mention:

TheBrain is a very nice application (but, as you observe, sort of Windows-oriented). It works fine on a Mac. It excels at organizing various pieces of information related to single topics. I think of it more as a super-charged Finder. I write about TheBrain on my blog too. I also wrote a review for MacAppstorm that you might find useful:

http://mac.appstorm.net/reviews/productivity-review/personalbrain-a-gps-system-for-your-information/

Curio is a nice "notebook" type of application. It is kind of a Swiss Army Knife. You can "explode" PDFs... that is, create a page for each page of a PDF and take notes relating to that page. You can build mind maps, do some basic outlining. If you like fiddling with your information, it could be a nice choice, but its very versatility is a bit of its undoing, so it doesn't do any of those things as well as dedicated applications. I'm also not sure about how cohesive its exports are.

Anyway, I'm sure I have only added to your confusion with the above. If I were you and I were already comfortable with Scrivener, I would just get DevonThink and start using those two tools. They may be all you need.

Steve Z.

Thank you for your comments. A few replies:

Daly, I will have a look at MyInfo and Citavi - thanks for the suggestions. I've tried both Mendeley and Evernote but a) I prefer not to work in the cloud (I think you can use both offline) but more importantly b) i just haven't been satisfied with them. I think they're loaded with too many bells and whistles for my taste. Zotero is really great, it just doesn't do searchable annotation/tagging very well (at least I haven't been able to get it working). I might try Sente now because it is a full-fledged ref manager but also incorporates OpenMeta tagging and the possibility to set up an external pdf database (both used with DevonThink for instance) http://code.google.com/p/openmeta/wiki/OpenMetaApplications

Steve, the point about tagging outside of DevonThink is that DT only searches within it's database, not the drive(s). With a simple tagger app Spotlight (like Windows Search) can be used to narrow searches outside of DT. Tagger (and I suppose Tags) also permits Boolean searches. I guess it's a bit like advanced googling for your computer.

Apropos Tinderbox, I've been reading through this forum and know it has some strong supporters;) I will check out your blog. BTW I found a great 30 min video of a researcher using Tinderbox for data analysis here: http://vimeo.com/8772338 The way he uses it reminds me more of something like Atlas.ti (which I find curious). My only concern is that it might take some heavy lifting to before I can learn it well enough. I will certainly keep it in mind. I guess it can be "programmed" to do some pretty powerful stuff but not sure it competes with DT's more automated intelligent searching capabilities, the writing capability of Scrivener, or the slicker mind-mapping tools (many for free).

In the meantime I will explore Scrivener further, perhaps in conjunction with a mind-mapping software that uses OPML. I like to move between the more graphic and textual to keep the writing alive. That way I think I can make Scrivener work better for the more fine-grained "content outlining" - something like this: http://www.christopherspenn.com/2012/05/how-to-turn-mind-maps-into-ebooks-and-presentations-instantly/

Hopefully I'm getting things narrowed down a bit! More comments are greatly appreciated!

Peter

Peter wrote:

Steve, the point about tagging outside of DevonThink is that DT only searches within it’s database, not the drive(s). With a simple tagger app Spotlight (like Windows Search) can be used to narrow searches outside of DT. Tagger (and I suppose Tags) also permits Boolean searches. I guess it’s a bit like advanced googling for your computer.

You can "index" files using DevonThink, which means DT will have an avatar like item in its database that is a link to the actual file outside that database. You can then create tags and add Spotlight comments right within DT. Indexing is a piece of cake. Just use the Index command from under the File menu and then select the folders you want to Index.

I just Indexed a folder of Scrivener projects as a test. Here's a screenshot of what I got:

https://www.dropbox.com/s/pk9ft4it6qze4sw/Indexed-Files-in-DevonThink.png

There's a tutorial for doing this, which you can access under the "tutorials" listing of the Help menu. The tutorial is called Importing Data.

Anyway, I'm not necessarily saying this will be more effective than having a separate tagging program, but if it works to your satisfaction it will have to be a simpler solution if you're already using DT for other information chores.

Steve Z.

Hi Steve,

Thanks! Looks like it must be possible. I must admit I haven't started using DT yet but fascinated by what I hear/see. Unfortunately there are very few screencasts of it being used in action, a bit strange given it's popularity. The DT tutorials are okay but the narrator has an annoying "marketing' kind of tone that puts me off. About the only thing I've found, along these lines, is this: http://www.youtube.com/watch?v=UH2L6WSA0UY - but it's from 2009. If you ever felt like doing a screencast of how you use it I would be an eager audience! (Here I can recommend the free app Jing http://www.techsmith.com/jing.html

One question I have is how you distinguish/integrate your workflow across different apps like DT and other things like Tinderbox, Scrivener. In other words, the importing and exporting and flow of information (keeping in mind the different file formats). Do you start with TBX first or DT or perhaps S or a mind mapping? Does this different for the kind of project you have in mind (like a thesis hint hint)?

I would be interested in posts that detail the challenges and promises as one navigates, imports-exports (and when) across such tools in their workflow!

Cheers,
Peter

Peter:
It's been many years since I wrote my dissertation. So many that none of the programs that you mention were available. I suspect that the people who wrote them may not even have been born when I was in Grad School. Indeed, I was one of the first people in my department to write my dissertation on a computer. I used Microsoft Word. Not the Windows version or the Mac version-the DOS version.
However, I have stayed in the business since then. First as a Post-Doc, and then as a researcher at a pharmaceutical company. Now I am back in academia.
In each of those different jobs I have had a similar set of tasks to the ones that you have. Assemble information. Put it someplace I can find it when I need it. Come up with a way to write about it. Come up with a convenient way to cite it when I write. This is an unending set of tasks that will always be there.
You will probably find that any system that you come up with is unique to you. Try a few things. If they help, keep using them. If they don't, dump them. Don't worry about what has worked for someone else. We all think and work differently. That being said, it is always good to have someone say "I tried program X and it worked to do Y." If you need to do Y, then give it a try. But just because someone else found it useful, that doesn't mean that you will. You found your way to a good place here. The people here have tried a lot of different things and write about them. Good place to find stuff to use.
What I have found works for me is four things. Devonthink Pro Office. Papers. Evernote. Endnote.
I use Devonthink for keeping a lot of the junk that I accumulate in a kind of order. I have a series of folders that Devonthink watches over and from which it sucks files. I have sets of smart collections and sets of manual collections that things get put into as well as some databases that just have stuff in it. Papers is invaluable to me for dealing with PDFs. Up until version 2 it wasn't that useful. Now it does most of the things I need. Highly recommended. Evernote is what I use for sharing stuff that I find with others. I could use Devonthink for this, but Evernote is easier and cross-platform. Finally Endnote. I don't really like Endnote, but it is the standard when it comes to sharing bibliographies and when you are writing.
Oh, I should also add, I use a Mac. Not that I have any sort of dislike of Windows machines. The software above runs on Macs, so I use a Mac. Also, I do my first drafts with pen and paper, then move them to the computer.
Finally a comment about writing-Just write. I found that the hardest part of the dissertation was simply sitting down and entering the words. I'll leave you with a quote from E. L. Doctorow: Planning to write is not writing. Outlining, researching, talking to people about what you're doing, none of that is writing. Writing is writing.

Good luck.
Jim

Jim, thanks for those words of wisdom. It's easy to get carried away with all the new tech out there. By and large I've felt my current workflow is sufficient for smaller projects (mainly Word and textedit) but the thesis has me a bit concerned. So many good ideas written down, but where did I put them? With my dissertation deadline approaching I've frozen a bit, like a dear in the headlights. I thought I would take another dip into the sea of apps after a few years of abstinence.The pencil and paper are under-estimated, so thanks for the reminder. My problem is that I tend to loose them.

A question about your use of DT: do you export anything when writing or just leave it all your DT databases for inspiration? (You didn't mention what word processor you use.) I wonder because I'm still a little unclear how best to integrate it into my process. I'm pretty clear now on the reference management, outlining/word-processing bits.

BTW I was a diehard Windows guy up to this past year and then finally broke down and got a Mac. I have Parallels with Win 7 installed as but its really more of a security blanket than anything else as I make the transition.

Peter,

Have you read Steven Berlin Johnson's article about how he uses DevonThink? It's insightful:

http://www.nytimes.com/2005/01/30/books/review/30JOHNSON.html?_r=1&oref=login&pagewanted=print&position=

If you haven't explored its export capabilities, you'll find that it is extremely easy to export data from DT for use in Scrivener.

Regarding Tinderbox, I find it a great resource when I feel a little lost in my data. You can use it to create a road map of your project, and not even worry about putting any of your research in it. (Note: I've never had a project as research-heavy or rigorous as a Ph.D. thesis, so I can't comment about how all these tools work when you delve into a job like that.) Tinderbox can now import a Scrivener project directly, which is nice; but it will be especially nice when and if the promised export TO a Scrivener project is ready. That will make for Tinderbox/Scrivener nirvana.

Steve Z.

Thanks Steve, this is very helpful! The export feature you mention for Scrivener would certainly give it an advantage.

Just a quick update on how I see my workflow coming together, and a few questions:

Reference management/Annotation of PDFs
For now I will settle for Zotero but I am looking for reference manager that can sync with a shared folder such as dropbox. I believe Sente does this but I've been unable to find a way to export/sync my reference/PDFs from Zotero. Any advice here is helpful.

Data Management
Here I have settled on DevonThink, although Tinderbox is a strong contender.

Note-taking/Outlining/Writing
Rather than using Circus Ponies Notebook, I am going to stick with Scrivener and a mind-mapping program. For notes I think I will try Notational Velocity or just create them inside DT.

On the Back Burner
Tinderbox
CPN

Peter wrote:

Thanks Steve, this is very helpful! The export feature you mention for Scrivener would certainly give it an advantage.

This Tinderbox feature was hinted at on the Scrivener forum (I believe). There seems to be a lot of communication and cooperation between the two developers, so I am hopeful this is indeed in the works.

For me, one of the biggest drawbacks of Tinderbox remains its arcane (to me, anyway) export functionality. I've even had a communication with Mark Bernstein (the developer) about it and he assures me it is easy, just a matter of creating some export templates. Still, I can't quite get my head around it. Fortunately, it is pretty easy to export to plain text and I can usually make do with this. Still, a dedicated Scrivener export would be brilliant and make the two applications a perfect match.

Steve Z.

A word about Tinderbox: although Tbx can be used for data management, such a use wouldn't be playing to its strengths. In a workflow, I think it's best used for either "creating by jotting down random notes to see what they add up to" or "analysis of what already exists to get a firmer handle on it". But I think it does already successfully export to Scrivener - at least there's a template in its file menu to allow it to do so, and I seem to remember a recent dialogue in the Tinderbox forums on this subject (do a Search on the term "Scrivener").

Incidentally, Steve Z.'s blogs on Tinderbox provide the quickest route I know (and I've done all the the tutorials, and I mean all!) to getting a decent grounding in the app.

A relatively cheap and simple mind-mapping programme for the Mac is MindNode Pro: http://mindnode.com/#!/mac It exports via OPML to Scrivener.

A word about DevonThink. It's a magnificent application, but you have to get used to its quirks. Personally, I wouldn't tag anything that I index or import into it; tagging complicates what you can do, and in the DevonThink context it has a slightly special meaning and takes a while to learn and understand. The other methods of finding documents in DT, or filing them away in the first place are sufficiently sophisticated that you don't really require tagging (although plenty of people would disagree with that). Personally also I would start off by importing rather than indexing, again for reasons of keeping everything simple. With indexing, you have two databases to worry about - the Finder's and DevonThink's own; with importing, effectively only one. You can always re-export what you've imported, and then index, when you've got the hang of the application. (Again, others may disagree.) And I certainly wouldn't mix both importing and indexing, at least to begin with.

I imagine that DT Pro would be enough for your purposes, unless you plan to scan material in, when DTPO would useful.

Generally Scrivener, Tinderbox and DevonThink do all play very well together. Good luck!

Peter wrote:

I would be interested in
posts that detail the challenges and promises as one navigates, imports-exports
(and when) across such tools in their workflow!

Cheers,
Peter

I've never written a thesis as such, but I have written various long-form factual pieces. This is the sort of workflow I recommend (distilled from the advice of many others):

Structuring > Drafting > Formatting/polishing, plus Data Managing and Storage, Citing and On-the-fly Note-taking:

- Structuring: Tinderbox, OmniOutliner Pro, Neo, or a mind-map programme (apart from MindNode, there are several others including NovaMind, MindManager for the Mac and the new/old champ, Inspiration). Curio is excellent as a virtual whiteboard, but better for fiction than fact (as are SuperNoteCard and Blake Snyder's Save the Cat software - though I'd love to try using it for a dissertation!). Scrivener's outliner plus corkboard is also a candidate. Export via OPML to...

- Drafting: Scrivener. Export via Scrivener's .rtf or .docx exporters to...

- Formatting/polishing: MS Word (better on Windows than the Mac), or, on the Mac, Nisus Writer Pro or Mellel (both of them favourites with academics, so I'm told). I don't think Apple's Pages is regarded as appropriate for academic use.

- Data Management and Storage: DevonThink (best for large volumes of data, and for filing, searching and simply getting the data in), Eaglefiler or Together. DT seems to be the choice of academics.

- Citing: Bookends, Zotero, EndNotes, Sente etc. It's well-worth consulting the Scrivener forum for this: there are several threads on this topic and posters there who've recently finished their dissertations, and as far as I can see Bookends and Endnotes are their favourites. Whatever you use, it needs to play well with Scrivener. I should also experiment.

- On-the-fly Note-taking: my favourite is nvAlt too (although there's a Scrivener scratchpad).

There's a running debate about how, in the middle of drafting a thesis with Scrivener, you send individual chapters to your supervisor, and then receive back comments and embody any amendments demanded. It's well written up in several threads on the Scrivener forums. Some people evidently round-trip Scrivener > Word > Scrivener, and some even apparently manage to "sell" Scrivener to their supervisors. But others make major supervisor-induced changes in Scrivener using its dual-editor feature, and then move over entirely to Word (or equivalent word processor) for final detail and polishing. For these purposes, MS Word's "Track Changes" and "Comments" are powerful features, although Scrivener has two types of comments and footnotes and I've read that Word "Comments" can be imported into Scrivener; personally I've never tried to do this.

I hope all this is useful.
H

Peter:

In looking at your later postings it looks like your approach is coalescing around a useful set of tools.
I don’t use Zotero, but I tried it out. It is a great piece of software. If I wasn’t running Papers 2, I would probably use it. Indeed, before the guys at Mekentosj released v 2 of Papers, I was looking at dropping the program and moving to Zotero.
Devonthink is also a good choice. You asked in an earlier posting how I use it, whether I export stuff from Devonthink, or just work within the program. I have tried exporting, but only to see what it can do. I use Devonthink on its own. I use it mostly for storing stuff I don’t want to lose. My databases have all kinds of stuff in them. Snippets from the web. Notes I have written. Copies of emails. I use it as an all purpose storage bin. I also use the collections functions to organize stuff. The easiest way is with a smart collection. Setup the smart collection, and whenever something goes into Devonthink that meets the criteria, it goes in there. Also being able to index individual folders is nice. Dump stuff into specific folder, run the index in Devonthink and presto-material is loaded in.
I don’t do much outlining(kind of an odd admission for someone who reads this group every day). I do use an outline if I am writing something like a paper or grant. In that case I usually do it with pen and paper first, then enter it either into Word, or sometimes I’ve used Notetaker. I am trying to get more in the swing of using Notetaker routinely.
As I said, I use Word when I write. This is mostly for two reasons. First, I have used Word since it was a DOS program, and I am comfortable with it. It’s big and bloated, but I know how it works. Second, it’s the standard. Journals want manuscripts submitted in Word format. Granting agencies want grants submitted in Word format. Exchanging documents with my colleagues I know they all have Word-either the Windows version or the Mac version. My university supplies either the Windows version or the Mac version for a substantial discount.
Still, the first draft of pretty much everything is done with pen and paper. My post-doc boss told me that I would never get anywhere until I learned how to compose on the computer. Well, I never managed it. And I have still been able to have a career in science, both industrial and academic.
Good luck.
Jim

Thank you both for your input. Hugh, in your first post you suggest that TBX can be used both for note-taking and analysis but then in the second post you propose that TBX is mainly for outlining/structuring so I'm a bit confused. Perhaps you could clarify a little more. I'm still trying to figure out if I can do without a dedicated outliner and only use scrivener combined with (cheap) mind-mapping app link MindNode.

I also appreciate your workflow model, however I would have to modify it somewhat (essentially flip it) to adopt it more to the "grounded" approach I use in my work. So rather than moving Structuring > Drafting > Formatting/polishing, plus Data Managing and Storage, Citing and On-the-fly Note-taking, mine looks something like this:

A: "Primary Data Management/Analysis" - in my case "qualitative" e.g. interview transcripts and field notes. Here I have something like Atlas.ti in mind however I would like to move over to a mac app and come up with a way to incorporate DT.

B: "Secondary Data and Literature Management/Analysis" ("Citing") - including amassing references (citations) and pdfs but one could include other secondary source materials such as photos, video, audio, etc. While I think a ref manager app remains useful I also hope to get DT involved in this mode too.

C: "Notes/Annotations" ("On-the-fly Note-taking") - usually during reading the above but also out walking, lying in bed, etc. So I find note-taking application independent (incl. pen and paper). However this eventually creates a problem when searching across-applications/computers to find connections in multiple places. We might call this the "googling inside out" problem. Notational Velocity is a nifty little app but it doesn't address real-life practices of ubiquitous notation. This is where the issue of syncing and phones apps is particularly urgent. I've seen reference here to Omnifocus.

D: "Building Semantic Associations" ("Data-Management and Storage") - which attempts to address the "googling inside out" problem noted above of searching through the hard drive. DT seems like the obvious choice here, but TBX also has Data Management capabilities as far as I understand.

E: "Outlining/Mind-Mapping ("Structuring") - I now see roughly two options here. Either one chooses a dedicated outliner (e.g. OmniOutliner, Circus Ponies Notebook, TBX?) that includes a mind-mapper component, or or one uses the built-in outlining capabilities of a word-processor like Scrivener or Word and add a (cheap) mind-mapping app to work more visually. I'm leaning towards the latter combo but still on the fence. Thanks for the MindNode tip Hugh. I've used the Lite ver but might upgrade to the Pro but both incorporate OPML.

F: "Writing" + "Layout" ("Drafting" + "Formatting/polishing") - Conceptually I think these are pretty close but I do recognize the advantage of drafting in Scrivener and exporting to a word processor, e.g. Word. One could even produce a polished layout in something like InDesign but that's generally over the top for most journal submissions - reference to your post Jim in particular.

The key point through all of this for me is that these modes/phases tend to overlap. For instance, note-taking/annotation is done during data analysis as well as reading, building creative associations, writing (and the other way too). This suggests the need to export/import across various modes/apps (in this case note-taking) as well as "googling inside out" in the apps/computer. So I suggest the more one can integrate this process, thereby limiting the number of apps needed to work the data, the more efficient, productive and creative one can work. This is the role I hope to give DT, if not now then perhaps eventually. But for DT to play "mother-brain" then all of its "child" apps need to export/import/sync with it! Here Dropbox might be an important component as well.

I'll check Scrivener's forum as you suggest Hugh. Hopefully there are some keys there. I recently posted on the Zotero forum and apparently there is a plugin for dropbox: http://forums.zotero.org/discussion/16827?page=1#Item_16

Steve, I forgot to thank you as well. My post was directed to you too (along with the rest of the forum). I'm on my way to check out the Scrivener forum now...

Oh and Jim, I take your point about using a single DT database with a select number of file but without indexing (at least for beginners). However part of my challenge is setting up the darn thing and locating the relevant files on my computer (and eventually an older computer that I haven't used for while). I purchased the DevonSphere Express which searches the entire computer. However you can only select one file at a time to move! I complained to Devon and hopeful they'll improve it to allow multiple file section (or include it in DT but I suppose that's less likely). That way I can start building a well defined DT database.


Stephen Zeoli wrote:

Peter wrote:
> Thanks Steve, this is very helpful! The export feature you mention for
Scrivener would certainly give it an advantage.

This Tinderbox feature was hinted
at on the Scrivener forum (I believe). There seems to be a lot of communication and
cooperation between the two developers, so I am hopeful this is indeed in the
works.

For me, one of the biggest drawbacks of Tinderbox remains its arcane (to me,
anyway) export functionality. I've even had a communication with Mark Bernstein
(the developer) about it and he assures me it is easy, just a matter of creating some
export templates. Still, I can't quite get my head around it. Fortunately, it is
pretty easy to export to plain text and I can usually make do with this. Still, a
dedicated Scrivener export would be brilliant and make the two applications a
perfect match.

Steve Z.

Peter wrote:

Thank you both for your input. Hugh, in your first post you suggest that TBX can be used
both for note-taking and analysis but then in the second post you propose that TBX is
mainly for outlining/structuring so I'm a bit confused. Perhaps you could clarify a
little more.

Sorry, my mistake. Tinderbox is an outliner, although far from the traditional kind. When I wrote that it could be used for “creating by jotting down random notes to see what they add up to”, I was really stressing the second half of that phrase, rather than the note-taking. Tinderbox is an application that excels in the generation, ordering and linking of ideas (that are manifested as notes). There are numerous better note-takers, in the sense of getting notes down.

What I wanted most to do was to differentiate Tbx from data managers and storers like DevonThink; some people do start off by thinking that DT and Tbx are substitutes, but when used for the purposes for which they have been designed, they are in fact complementary.


I'm still trying to figure out if I can do without a dedicated outliner and

only use scrivener combined with (cheap) mind-mapping app link MindNode.

Plenty of people do indeed do without. Tbx, although it has an unusual pricing model, isn't cheap initially, and its learning curve is steep. OmniOutliner Pro is much easier to learn but isn't cheap either. The outliner NeO, when I last looked, was very inexpensive. A less expensive outliner option I should have listed is Opal.

I also
appreciate your workflow model, however I would have to modify it somewhat
(essentially flip it) to adopt it more to the "grounded" approach I use in my work. So
rather than moving Structuring > Drafting > Formatting/polishing, plus Data
Managing and Storage, Citing and On-the-fly Note-taking, mine looks something like
this:

A: "Primary Data Management/Analysis" - in my case "qualitative" e.g.
interview transcripts and field notes. Here I have something like Atlas.ti in mind
however I would like to move over to a mac app and come up with a way to incorporate
DT.

B: "Secondary Data and Literature Management/Analysis" ("Citing") -
including amassing references (citations) and pdfs but one could include other
secondary source materials such as photos, video, audio, etc. While I think a ref
manager app remains useful I also hope to get DT involved in this mode too.

C:
"Notes/Annotations" ("On-the-fly Note-taking") - usually during reading the
above but also out walking, lying in bed, etc. So I find note-taking application
independent (incl. pen and paper). However this eventually creates a problem when
searching across-applications/computers to find connections in multiple places.
We might call this the "googling inside out" problem. Notational Velocity is a nifty
little app but it doesn't address real-life practices of ubiquitous notation. This
is where the issue of syncing and phones apps is particularly urgent. I've seen
reference here to Omnifocus.

Omnifocus is a yet another species of creature in the jungle. It isn't a note-taker; it's primarily a task manager, probably the grand-daddy of Mac task managers. Useful and powerful, but with a moderately steep learning curve. There are several cheaper, or even free alternatives, most of them on the web.

As note-takers, Notational Velocity or nvAlt remain good choices. One reason for this is that they can sync via Simplenote to other note-taking and writing apps on iPhones and iPads. Users seem to rate this functionality highly. The two applications are also free (or donation-ware, I forget).

D: "Building Semantic Associations"
("Data-Management and Storage") - which attempts to address the "googling inside
out" problem noted above of searching through the hard drive. DT seems like the
obvious choice here, but TBX also has Data Management capabilities as far as I
understand.

As I implied above, Tbx has truly excellent data management capabilities in the sense of data manipulation, but not data management in the senses of capture, filing, storage and retrieval, which are what Devonthink is good at.

E: "Outlining/Mind-Mapping ("Structuring") - I now see roughly two
options here. Either one chooses a dedicated outliner (e.g. OmniOutliner, Circus
Ponies Notebook, TBX?) that includes a mind-mapper component, or or one uses the
built-in outlining capabilities of a word-processor like Scrivener or Word and add a
(cheap) mind-mapping app to work more visually. I'm leaning towards the latter combo
but still on the fence. Thanks for the MindNode tip Hugh. I've used the Lite ver but
might upgrade to the Pro but both incorporate OPML.

F: "Writing" + "Layout"
("Drafting" + "Formatting/polishing") - Conceptually I think these are pretty
close but I do recognize the advantage of drafting in Scrivener and exporting to a word
processor, e.g. Word. One could even produce a polished layout in something like
InDesign but that's generally over the top for most journal submissions - reference
to your post Jim in particular.

Conceptually they are close and even overlap, but my experience of the Mac is that its user interface and the ways in which it assists interoperability between programmes encourage one to use the best tool for its role in the workflow, rather than to seek out a "Swiss-army-knife" "jack-of-all-trades" solution. That is certainly one of the assumptions behind the development of Scrivener - that one writes better if one gets the ideas and words right first, with the formatting and polishing left until later, perhaps for another application. For this reason, it exports to many formats, including those used by word processors.

Scrivener declares explicitly that it itself isn't a word processor, in the sense of providing full formatting and layout. Most of its tools are designed to help in getting the ideas and words right, although it can format work simply, and more and more users appear to be carrying out the entire writing and formatting process within it, but mainly for purposes like e-books. The doctorate students and other academics who use it appear to value the qualities of Scrivener in the ideas/words phases of their work, but I doubt very much whether its formatting tools - it has virtually no layout features as such - are up to the requirements of finally outputting a thesis.

The key point through all of this for me is that these
modes/phases tend to overlap. For instance, note-taking/annotation is done during
data analysis as well as reading, building creative associations, writing (and the
other way too).

That is my experience too in writing long-form factual pieces. In fact the process is often one of moving constantly backwards and forwards between different "modes". But that doesn't mean that the tools you use cannot be distinct and appropriate for different processes: the Mac interface encourages this. I often have nvAlt, Devonthink, Scrivener, Tinderbox and MS Word all open on my computer at the same time. As I mentioned in another post, I believe Tinderbox will talk to Scrivener pretty well, Scrivener exports successfully to Word, and DevonThink stores all the files you put in it in their "native", original form, so that they can be successfully retrieved whenever you need them.


This suggests the need to export/import across various modes/apps

(in this case note-taking) as well as "googling inside out" in the apps/computer. So I
suggest the more one can integrate this process, thereby limiting the number of apps
needed to work the data, the more efficient, productive and creative one can work.
This is the role I hope to give DT, if not now then perhaps eventually. But for DT to play
"mother-brain" then all of its "child" apps need to export/import/sync with it! Here
Dropbox might be an important component as well.

I see DT as less a "mother-brain", and more as a "mother-bin". I believe it happily indexes Dropbox folders.

I'll check Scrivener's forum as
you suggest Hugh. Hopefully there are some keys there. I recently posted on the Zotero
forum and apparently there is a plugin for dropbox:
http://forums.zotero.org/discussion/16827?page=1#Item_16

Hope this helps.
H

Peter wrote:

The key point through all of this for me is that these
modes/phases tend to overlap. For instance, note-taking/annotation is done during
data analysis as well as reading, building creative associations, writing (and the
other way too). This suggests the need to export/import across various modes/apps
(in this case note-taking) as well as "googling inside out" in the apps/computer. So I
suggest the more one can integrate this process, thereby limiting the number of apps
needed to work the data, the more efficient, productive and creative one can work.
This is the role I hope to give DT, if not now then perhaps eventually. But for DT to play
"mother-brain" then all of its "child" apps need to export/import/sync with it! Here
Dropbox might be an important component as well.

Like Hugh, I will agree with your conclusion that the various stages tend to overlap; in fact, I would say that the procedure is, to a significant extent, recursive. Sometime ago Dr Andus started a thread on a similar viewpoint http://www.outlinersoftware.com/topics/viewt/3906

With this in mind, I would focus on minimising the number of apps used, and not so much ensuring import/export across a broad range of tools. My experience says that one can spend innumerable hours moving material from one program to another instead of doing real work.

A few years ago I did my MBA dissertation and was initially quite concerned about getting my software setup right and tried quite a few of the tools. Fortunately, quite early in my dissertation and in a rare display of maturity, self-discipline and CRIMP control (rare, and probably the last) I decided to stick with just two programs. What those were is probably irrelevant to you as they are both Windows programs, namely Brainstorm and Surfulater. However, the reasons for my choice might contribute to your own selection:

- I chose two programs that are completely complementary; Surfulater is very good for collecting existing material and organising it with folders, tags and in arbitrary order. As a 'mother bin', it could be the equivalent of DevonThink in your case (though without the 'intelligent' features). Brainstorm, on the other hand, excels in text manipulation and is a minimally styled but powerful writing environment; its text cloning feature means that I can write and rewrite stuff, while still maintaining the link to the original reference.

- I chose programs that I was already familiar with and could use transparently in my workflow. If you are not accustomed to the tools that you intend to use, you have an additional reason to use as few tools as possible.

Just my 2c.

Hi Peter and Steve
I found your posts quite brilliant and helpful. I am also starting up my phd in linguistics, exactly in Peter's situation. I tried almost all the apps you guys mentioned. Here are my observations so far.

1. I use Mendely for some preliminary works; to collect and rename my PDF files into a Dropbox folder. it is not the best reference manager; has a lot of flaws specially for Latex. But, the Watch Folder and Rename features are really helpful to cleanup my computer. I drop all my newly downloaded files into a dedicated folder, then Mendeley automatically sucks the papers in it, recognizes the metadata, rename and put it in another folder in dropbox. All my 2000 articles are beautifully renamed by Author-year-title format; easy to search them in Spotlight (alfred).
2. Bibdesk and Jabref are good enough for bibliography (I also don't like Zotero), for Latex. But I use Papers for its Magic Manuscripts feature (amazing).
3. Devonthink is indispensable for its artificial intelligence; recommends related papers.
-----
4. Tinderbox is too complex. Plus, it has limited capability to import various file formats. I tried to import word files and Scrivener documents. It rips of the images in the files. It is poorly integrated with other applications, generally; seems to work best only with text files. Exporting to other format is also equally messy in Tinderbox.

5. There is one crucial feature that the Circus Ponies Notebook offers that all other note-taking applications, except Microsoft OneNote, don't. The ability to tag (mark) points deep inside the text. When I am writing a long note, I always want some way of tagging some key points that I want to remember latter. All the note-taking tools could tag files, but never bullet points, or paragraphs. Notebook and OneNote can do that. It is called Keywords in the former and tags in the latter applications. Notebook also has a MultiIndex system, another way of finding your crucial points using keywords. If you like tagging your specific points in your extended notes, I think, there is a reason to use Notebook. I still am not satisfied how Notebook syncs with Devonthink. But, I think there is not better note-taking app right now, unless you want to use Scrivener for everything. Here is a good article on the difference between Scrivener and Notebook, http://www.organizingcreativity.com/2012/02/outliner-in-scrivener-vs-outliner-in-cpn-structure-scrivener-vs-content-cpn-outlines/
------
I am not still satisfied about the workflow; the sync between Devonthink and Notebook is so clumsy.

have you guys (specially Peter) developed a good workflow yet? Can you update us if you have come up with better solutions with the general workflow?