best software for doing a research paper/thesis- scrivener?

hi all,

a friend recommended scrivener but it is not obvious how i can use it to write a research paper. perhaps some other software would suit my working style better (or improve it). can anyone who knows about these things help me out?

my working method is to take notes from a single article or book into one doc. for each source, i have a single doc. i build up a big folder of note-docs and then write an outline. i go back through my note-docs, pull out quotes and cut/paste them into my outline doc. then i write the paper from the outline.

so i need (1) to have files for each source with my quotes and (2) be able to turn individual quotes into ‘notecards’ to shuffle around in an outline.

it is not obvious to me how i could do this in scrivener (i did some of the tutorial). it looks like i can have cards that are synopses of things like scenes, but that is not quite what i need. i need to have lots of card (quotes) that all belong to the same book and then be able to sort them into an outline (along with other similar cards from other sources).

keith from scrivener contacts said you were all a very friendly bunch and would be willing to help me out. can i use scrivener to do this or should i be looking at different software?

thanks in advance for your forebearance,


Each quote from your source material can be entered onto its own “card.”

Each card can be dragged and dropped into an outline along with other notes and prose you write.

Sections can be grouped in Folders. They also can be dragged and dropped. So you can craft an outline easily. And re-work it easily.

Then, you can, with the press of a button, watch the entire thing assemble in seamless, contiguous prose (the “Scrivenings” function – watch the video to see this).

SUGGESTION: Watch the entire tutorial video to really catch the gestalt of Scriv. Download the trial and play with it. The trial is full-function, and allows a very generous demo period. And the software is inexpensive anyway.

PREDICTION: You’ll be hooked in short order and won’t want to write without Scriv. AND you’ll find many other uses for it.

Good luck finding whatever works well for you!

thanks noah!

if i dont want to retype all my source info every time i make a card
if want to be able to read through my notes from a single source in one place to remind me what the source said

wont that be inconvenient with every quote being its won card?

how could i keep cards from a single source related to each other in a specific order?

kind regards,


Scrivener is a fantastic writing tool, but not quite as suited for research or information management. Otherwise, KB would have named it Researchener instead of Scrivener (sorry if that was a bad pun – my only excuse being that I’m not a native speaker … /err/ … writer).

So, in my opinion, you might indeed need looking for another kind of application. M. Bernstein’s Tinderbox comes to mind, which will do exactly what you described, and much more beyond that. Be warned, though: TB does have a learning curve – it will take you at least a couple of weeks until you really “get” it.

Other options include Journler (can’t wait for Lex, its upcoming, more information management oriented sibling); Together (formerly KIT) from reinvented software; or Eaglefiler.

Then there are, of course, Devon agent, Devon note and Devon think (professional). Extremely powerful – but, alas, not very suited to my work-style, at least.

Personally, I’m very satisfied with the combination of Together, Tinderbox, Scrivener and Pages (plus, for more heavily structured documents, Papyrus). And that may be the main point of this lengthy post: don’t look for one application that fits all your needs; instead try to find the right combination of apps that fit your personal work-style and work flow.

Hope this helps.

The outliner view in Scrivener (as opposed to the corkboard view) will do what you’re asking for here.

If you’re getting your source info (I’m thinking quotes, so paragraph long approx) from other (text/PDF) documents, I think the quickest way would be for you to highlight the text you want to import from that document, and drag it to Scrivener’s dock icon. This will create a ‘clipping’ document with your text as its contents. If you are researching on a particular topic, you could drag all of these snippets to Scrivener as clippings in this way, and then move them from the Clippings folder to a topic folder for working with. You could use Scrivener’s reference pane to keep the original source URL (or file) for citing later.

To view all of the items under a folder in Outline view, select the folder containing your items and hit the Outline button on the toolbar. Then, you’ll be able to expand it out so you can read the quotes, etc, that are in each of the sub-documents, all on one screen. You can sort these as you wish by dragging them around, or sorting ascending/descending as well.

This isn’t to say that any of the other apps that have been mentioned here wouldn’t suit you better, but Scrivener makes a fine tool for academic papers if you want to use it that way. I’ve been using it in conjunction with EndNote to great effect, this semester.

thank you marcus and othello.

you have given me a lot to look into!

othello, can you tell me more of how you work and manage to use all of those apps? it would be interesting to here about your working style.

marcus, do you use the procedure you have described in writing your papers? also, if i am mainly typing notes in from books (ie not pdfs) would your method work as well?

thanks to both of you


after reading marcus and othello’s postings, i spent quite a bit of time with the tutorial and then fooling around with Scriv to see if i could get it to work for me. i tried to follow marcus advice about howe to create quotation cards and re organize them.

these leaves me with some specific questions.

my work method would be to create a folder called Stowers 1994, for example, which would be a source. then i could create cards/text files inside of it, each of which would have a title like Sowers 1994 p20 for an individual quote.

then i make the text in the file into the synopsis so i can see the text when i outline.

i put all of the quotes in to research and then outline in draft. once i have an outline, i shuffle cards around in the draft area.

this leads to my questions:

  1. i work in english greek and hebrew. can synopses be in hebrew? my experience is that hebrew text works fine but the synopsis tunrs to rubbish when i try to make the hebrew text into the synopsis.

  2. can i bulk convert text in to synopses? my experience is this has to be done 1 by 1.

  3. can i export the contents of one folder from research as one doc - where each text file would be concatenated with the others in the folder.

4 can i bulk import quotes from an external doc recognising that a certain kind of break should be the dividing point for a new text file/element?

i could see myself using Scriv if it could do some of these things. the hebrew part is fairly necessary.

thank everyone,

d a friedman

  1. I don’t use non-Latin characters myself so I can just guess why your Hebrew text gets transformed into rubbish in the synopsis:

Can it be the fault of the font (Courier) in the synopsis? Maybe it doesn’t contain Hebrew characters. Try to change the synopsis font in the preference to the same font you’re using in the text. If it works you will loose a little bit of the cool typewriter typing on index cards look but it would be an easy and fast solution.

  1. As far as I know you can’t. But maybe you have noticed that the synopsis of a text chunk stays empty when you type in Scrivener but when you import it the beginning of the text automatically fills the synopsis. Maybe at some times this could be the basis of your handmade synopsis and save at least some time.

  2. First: no. In former versions of Scrivener the feature you’re talking about was indeed called ‘Export’ but nowadays to make it more clear it is ‘Compile Draft’.

With a stress on ‘Draft’, in your case. You can do a lot of things with the texts in the draft folder but not with the stuff in the research folder. The concept of Scrivener is to give you the best possible environment to write your text, including to have all your research material handy, and then to put out THE TEXT. Not the research material.

But as long as you’re talking about text only nothing stands against dragging that particular folder with your research material into the draft folder. Maybe just as an interim solution until you have compiled it. Or you keep it there (maybe at the end of you draft folder) and explicitly exclude it from compiling. The option to easily switch to ‘Compile All NOT Marked’ comes very handy. (And a very easy way to raise your heart rate immediately when you forget to re-set this because this might result in an empty export file. Kicks even better when it happens just minutes before a deadline. And all without illegal substances!)

  1. If I understand you right you are asking for a feature I would like to see in Scrivener too. You import a bigger piece of text and tell Scrivener to automatically chop it at every X – I like to divide my fragments (written in one file) by three asterisks, for example. It would be so great if this was done automatically by Scrivener. Including every n words as the name of the generated text chunk and maybe some trimming of empty lines at the end and the beginning.

Which makes this a ‘no’, I fear.

PS: I recommend Together, too. This is my big notebook for everything and when some of the material starts forming a project I just start one in Scrivener and drag all the stuff in. You might also check out VoodooPad which is a totally different way of organizing an connecting things. Personally, I found that I have no use of it but I can see that this program has it’s point.

You can already do that with Scrivener’s MultiMarkdown import. The trick is to transform your fragments into something that looks like MultiMarkdown headings (e.g. # Heading). Given that your big text file looks like this:[code]text text …

text text text text text text…[/code]
you can let Textwrangler search for (\*\*\*\r*)(^\w*\ *\w*\ *\w*) and replace it with

\#\2\r\r\2 The Textwrangler manual has a short introduction to ‘Searching with grep’, that explains all the magic behind this.

Hi df! It sounds like you and I have similar research/writing styles, so I thought I’d post a screenshot of how I approach my own work with Scrivener.

I created a folder called “reading” and in that I place all of my reading notes - quotations, questions, comments, etc. I used to do what you did and create separate word documents for each article/book, but it got ridiculously cumbersome. When I’m ready to start writing, what I do is split the screen window and use the bottom half for my reading notes and the top half for the writing that I’m doing.

I also used to write an outline in word and then would put the information from my reading notes inside the outline. Instead, what I now do is create a sort of outline tree (as you can see from the list of chapters). Each chapter is broken down into its component parts and I can quickly go from piece to piece. Sometimes one quote fits in chapter 2, so I whiz over and cut and paste. Another quote fits in chapter 5 and I whiz over there. What makes the Scrivener system so user-friendly for me is that I can see the entire structure in the binder to the side. When I was using word and had the entire document open in front of me at once, it was hard for me to maintain an overview. Here, if I want to see the big picture, I glance to the left and see the big picture.

Finally, when I’m writing big, big projects, I love the target feature. Sometimes writing for me is like squeezing blood from a stone and to be able to shoot for having that little red target light turn green is about all I can manage (because thinking of an actual page count would be too daunting).

That’s my two cents worth, df. Good luck in finding a system that works for you.

thanks to suavito, signinstranger and martini for their help. here is some follow-up:


-yes, you are right, changing the font in the index card does make the hebrew visible. sadly, it turns all the roman letters in to rubbish! this might be a limitation of the particular hebrew font that is being used by the software from which i am cutting and pasting the hebrew.

  • yes again, i notice that synopses are already filled in when text is imported. it is too bad that typing text in requires the extra step of pushing the text in to the synopses window, 1 file at a time, when dragging texts in does it for you.

-as for exporting, i realized it could be done (in theory since i havent done it) from the draft folder. it isnt clear to me how i could keep my research together, however if i have it sitting around in the draft folder.

  • regarding importing, signinstranger seems to have a workable suggestion but i t appears to require quite a bit of learning. having looked at the multimarkdown page and having not been able to find a reference to textwrangler, i have to assume it is more high tech than i might be able to work out given my limited time. by all means, tell me i am wrong!


thanks for your posting, it sounds like i am trying to work the way that you used to! yes, it is a drag.

how do you keep your quotes separate and identified by source once you move them around in the draft window.

thanks for your screenshot. i can see how you organised things. do you keep your raw quotes down in the research folder? how do you go about making an outline in Scriv? i have read the turoial and tried doing it but it seems a bit clumsy to me. since we work the same way, i take it you have figered out a good way to make an outline.

if i did have a good outline, i would want to sort my cards (id’ed by sources) into the outline. is that what you do? if i could find a clean way to shuffle cards around while keeping them id’ed, i would like to try it.

what is your note taking process with scriv?

thanks to everyone for their help.

kind regards,


I’m not sure what the accepted etiquette is on this forum when it comes to attachments, so I’ll apologize in advance in case I’m wrongly posting all of these attachments.

If I’m understanding you correctly, df, it seems that you’re asking two questions - one on how to take notes and one on overall organization - so I’ll address them individually.

In the screenshot of my note-taking, you’ll see that the bottom half is the article I’m reading and the top half is where I have my notes. This particular example doesn’t have any comments in the notes section (very dry article), but usually I’ll use bold font for my interpretations and regular font for the direct quotes. You’ll also see that I put the citation information after every quote and footnote it (why it’s highlighted in gray). By keeping that citation information after every quote, I can cut and paste into the body of my text without having to worry about keeping track of who said what and where my thoughts are coming from. In other words, by doing the work of inserting “author (year), page number” after every quote, I can save myself some hassle down the line.

Regarding the whole question of meta-structure and cards and the rest… I’m getting the sense that people use cards differently. One option would be to create a new “card” for each quote. In that case, you might want to create one folder that would hold all of your notes, one underneath that for a particular author, one underneath that for a particular work, and then a series of texts for each quote. I imagine the structure would look something like this (let’s assume the author’s name is Smith): Reading Notes --> Smith --> Smith’s Fabulous Book --> quote 1/2/3/4/etc. That way you could treat each quote as a separate document and move it around your structure freely. The big down side that comes to mind immediately is how in the world would you be able to actually say anything meaningful about that quote in a larger context? If you’re moving quotes around to and fro, would the overall structure get way, way too messy to keep track of? That’s probably something you can work out yourself or someone else might be able to help you with if this is the way they work.

You can see from the cards screenshot that I use the cards/corkboard view only to organize the sort of bigger picture flow of ideas. I don’t make individual items for individual quotes. I use the binder to create top level headings (think the things that are labeled “heading 1” in ms word documents) for chapters. Within each chapter, I break it down into sections (think of “heading 2”). Sometimes those 2,000 word sections get out of hand for me and then I break it into smaller bits. The pieces that need to be moved get dragged to a new home - sometimes in another chapter, sometimes in my “segments” folder until I figure out what to do with them.

I think it might be helpful to point out that the cards are only one way for you to visualize what you’re doing. If you want to see each individual quote and how they relate to each other, by all means put each quote as a separate item. Mess around with how you want to think about them in the corkboard view and, when you’re ready to start writing, merge it all together. If you don’t want to see things at the quote level and are instead interested in the big structural/organizational issues, create broader headings.

Anyway, this is super long-winded and I do apologize for that. I’m trying to balance being thorough with being pedantic and I hope I didn’t err on the side of the latter.

And yes, let’s all note from the screenshot that it’s 9:30 on a Saturday evening and I’m sitting at home (lame, lame, lame).


Since lately I’ve started writing more essays, I’m also started rethinking my working habits - and tools. The two latest essays I wrote (a 40-page dissertation and a long feature article) were all entirely written in Scrivener.

My notes were very confused, unorganized. Obviously, I should have done two things:

  1. Creating a finer system of folders to categorize my notes, writing new notes only in the right folder; and
  2. tag all notes for easier researching (instead of reading everything each time, or relying on the Find function, whose results are never as filtered as tagging would allow).

But I wonder if better reorganizing and tagging are enough. The semantic search system of DevonThink (“See also” function) is surprising me for how cleanly it can produce useful results. But I’m comparing it with my disordered notes, that I could have ordered better.

Another problem is that I’m not always smart enough to put a note in the right folder. For example, I could have two folders like these:

  • Debussy
  • Wagner


  • orchestration
  • music form

What if my note speaks of the relation of Debussy with Wagner’s music? And, what if should speak of orchestral colors in relation to the new idea of time as seen in Debussy’s musical form? Either way, I will put the note in the wrong folder.

So, while I would like to keep everything as much as possible in Scrivener, I’m really feeling tempted by DevonThink. Would the time needed to set everything up with it worth the improved organization system, and a faster/clever searching of related information? I’m trying the demo, but I’m still incapable to decide.


I’m a huge fan of both DevonThink and Scrivener. In my experience, they complement each other very well.

I’ve found DT especially useful as a long term data repository. If you have more than a few projects in the same general topic area, the initial organizational effort pays big dividends over the long term as everything you need is just, almost by magic, there. Scrivener is a far better writing environment, but isn’t really designed for cross-project data storage. DT is.


I’m a novel writer, but sometimes I start with something similar like having notes from one source in one document. Either it’s from research on a subject, or it’s the description of a character.

What I do is:

  • First, I save away a copy of the whole text. Just in case. (Very often, it’s a file written in TextEdit or Sticky Notes or the like anyway).
  • Second (now we’re in the Scrivener project) I give the document every keyword that might be important later. And, important, I click OFF the option “Include in Draft”.
  • Third, I split the document using CMD-K (or OPTION-CMD-K if I want automatic titles) in its tiniest pieces. The point is that the keywords duplicate, so it’s easy to find all these pieces again, if necessary. And all these pieces will be automatically NOT included in the draft because they inherited this attribute as well.
  • Fourth, I move these pieces around. Facts about the airplane my hero flies? Go in chapter 15. Short description of a move the antagonist will do in order to make hero’s life harder? Go in chapter 12. And so on.

This way, I get an outline cluttered all over with a lot of small details, very much as if I would use an index box. When writing the novel (or even when writing a detailed outline), I have everything at hand (using the split screen, of course).

The only issue with this is that it’s difficult to manage from a certain point on, because these “information bits” cannot be distinguished from the actual text - they all look like documents in the binder and in the outline. I try to overcome this by the use of colours: All my actual scenes get a colour depending on the POV character, so the “fact cards” stand out because they remain white.

(I suggested in the past that documents in the draft that are NOT to be included appear with a special symbol in the binder - a thick dot or something -, but unfortunately, that was not possible. It would help because one could distinguish different types of background information with colours - research on subject A green, the character development of the hero yellow, research on subject B blue… :frowning: )

Maybe this helps a little bit with writing a research-heavy paper.

Katherine and Andreas, thank you for your answers.

Katherine, you raised an issue that I was forgetting: reusing the same sources for different works. Yes, actually I’ve a few (interconnecting) thematic areas to deal with: theatre, music, philosophy, information technologies. Reference materials can be either specific to an essay or article (documents on a spectacle, notes about a software) or very general (acoustic theoy, history of music, theory of theatre…).

My guess is that it is time I keep a centralized archive. Maybe Spotlight is enough, maybe not. I’m still confused. Probably, the semantic search of Devon (that I’m continuing to explore) is what really makes the difference, and would save a lot of time and prevent missing documents I don’t remember I had.

Probably, I could simply go on and use it. It’s not even a matter of price, since I don’t need the OCR feature (I’m forced to own a license of Adobe Acrobat, and there is a decent OCR feature there*). Not even a matter of time, if I start working on it right now. Being mostly a matter of text data, exchange with Scrivener doesn’t seem a waste of time in no way.

Andreas, thank you very much for the inside view of how a writer works. Your method is really fine, and please excuse me if I will adopt it. I was going round something similar, but you showed how this should be done.

  • Decent? Depending on the file. With a complex photo reproduction, with text not always on focus and not evenly lightned, Acrobat only extracted a few pieces of words, while DevonThing Pro Office extracted nearly everything. Really impressive.


This thread on DT’s forum may be useful to you, in case you haven’t seen it:


Come on. I will be pleased - and be it just to know my posting was worth the effort. After all, this is what forums like this are about, aren’t they? :wink:

Hugh, thank you for pointing me to that thread. I re-read it more carefully, and I think that what I should do is:

  • Put a copy of all my reference materials in DevonThink.
  • Create an outline of my project in Scrivener. Jot down ideas in both programs, but always have a searchable copy in DT.
  • Do my researches in DT, to have the broader view on all materials I’ve in my disks.
  • Create a dedicated space for my project in DT, and copy there all the distilled infos.
  • Move the distilled infos in Scrivener, and write down the essay.

This should be a comfortable workflow.



My pleasure!