Organizing Scholarship: Technology, Old School Tricks, and Decluttering Research

[overthejordan]

I'm taking this opportunity to start a thread for solutions to organizing research. How do you keep track of citations, statistics, pdfs, revisions, notes, photocopies, etc.? Do you use Endnote, Zotero, Mendeley, something else? Do you use multiple filing cabinets? Anybody still use index cards for managing sources? Do you sync computers with cloud solutions? Dropbox? Box? Something else? Do you use a wiki for research collaboration? Share your tips and tricks for keeping your research organized here.

[polly_mer]

I use BibDesk and Git repos to push between various computers.

It took a while to get used to the workflow with Git (change files, add to the staging area, commit with a comment on what I changed, and then push to the remote repo), but now I love it. 

I have one Git repo that is my Literature Review repo with all my notes, PDFs, etc. that are indexed in one giant .bib file.  The BibDesk interface means I can attach files and click to bring up anything.  I can also search any of the keywords I've entered as well as any of the words I've entered anywhere in a record (can't search the text in the attached PDFs, though).  I can also annotate every entry with relevant notes and even copy abstracts for more useful searches.

I then set up a new Git repo for each individual writing project so that I can use Git's version control so I can look back at any previous iteration and get away from files like http://phdcomics.com/comics.php?f=1531 cluttering up my directory.

[aside]

Almost thirty years ago, in the days before Zotero, etc., or even pdfs, I went to put a photocopied journal article in one of my filing cabinets and discovered it was already there.  So I created a database in something called ClarisWorks and spent a month one summer entering every article, book, and book chapter in my library and filing cabinets, along with a list of keywords that were not in the title of the item.  I have kept up with it since then.  When ClarisWorks went belly up, I ported it to AppleWorks, and when that was no longer supported, I ported it to FileMaker.  It has been a valuable resource for my research.

[fast_and_bulbous]

A grad student introduced me to Paperpile a few years ago and I have used it regularly ever since. It's easy to add papers (PDF) and getting the citations exported is just a click or two. I can run through my university's proxy service when I'm away from the desk to access papers that would otherwise be behind a paywall. I use LaTeX almost exclusively for my work, and Paperpile does great with the Bibtex format. The only downside with Paperpile is it doesn't run on mobile devices, such as an iPad, which I use mostly for just reading and marking up stuff in iBooks.

In my collaborations I always cringe when someone wants to do something with Word. The last time I did a Word paper, I swear I got carpel tunnel from all the mousing! But I am a grizzled Unix/LaTeX kind of guy. I will say, for collaborative editing, I like Google Docs with its "suggestion" mode which is similar to Word track changes (the one thing I do like about Word).

In a proposal I wrote recently I did it in LaTeX, but we did all the collaborative editing in Docs, so I'd have to import the text back into Latex to recompile it, and then export it back using latex2rtf and then reading the RTF with word. Crazy and weird, I know, but I cannot abide Word for doing pubs with figures, citations, formatting etc... I do a lot of stuff with vector graphics and Word seems to always want to turn stuff into bitmapped images. Just give me a LaTeX template and I'm good to go.

[polly_mer]

In my collaborations I always cringe when someone wants to do something with Word. The last time I did a Word paper, I swear I got carpel tunnel from all the mousing! But I am a grizzled Unix/LaTeX kind of guy. I will say, for collaborative editing, I like Google Docs with its "suggestion" mode which is similar to Word track changes (the one thing I do like about Word).

One thing that's a top contender for what I hate about Word is track changes.  I'm involved in a group writing project now that I spent a week converting to LaTeX from Word because that's what the people doing the active writing liked.  The free-to-us tech writers who will come in later to polish only accept Word documents with all the track changes still on so they can see what changed.

My argument that practically every word will change in this process so the sea of red helps no one was dismissed and we will go forward with only Word with track changes on and then someone will have to sort out all those changes at the very end.

[pgher]

I used to use EndNote, but it doesn't get along with cloud services. I'm on Zotero now.

I use LaTeX as well. I'm almost exclusively using Overleaf. It runs in a browser and handles collaboration pretty well. That way I can see and edit my students' work, and they don't have to set up a LaTeX toolchain.

[polly_mer]

I use LaTeX as well. I'm almost exclusively using Overleaf. It runs in a browser and handles collaboration pretty well. That way I can see and edit my students' work, and they don't have to set up a LaTeX toolchain.

Does Overleaf run exclusively in a browser and does that browser have to be able to call home to a server?  I ask because I'm trying to get more of our new folks to convert to LaTeX, but we do have to be able to work with no connection to the internet.

[fast_and_bulbous]

The free-to-us tech writers who will come in later to polish only accept Word documents with all the track changes still on so they can see what changed.

My argument that practically every word will change in this process so the sea of red helps no one was dismissed and we will go forward with only Word with track changes on and then someone will have to sort out all those changes at the very end.

Egads that is dreadful. It also reminds me of a recent article I had to review/approve as editor for a special edition. It was nothing but a sea of red, and I was supposed to review this thing?? I had the (real) editor send me a "accept all the changes" version in PDF so I could actually do my final review.

My liking track changes is predicated upon the fact that, every once in a while, you accept some of the damned changes and then work from that new document. Keeping every bloody change is silly and leads to incomprehensible rubbish.

I can feel my pulse going up just thinking about this haha...

[polly_mer]

The free-to-us tech writers who will come in later to polish only accept Word documents with all the track changes still on so they can see what changed.

My argument that practically every word will change in this process so the sea of red helps no one was dismissed and we will go forward with only Word with track changes on and then someone will have to sort out all those changes at the very end.

Egads that is dreadful. It also reminds me of a recent article I had to review/approve as editor for a special edition. It was nothing but a sea of red, and I was supposed to review this thing?? I had the (real) editor send me a "accept all the changes" version in PDF so I could actually do my final review.

The people who have to vote on the document before it goes public nearly always request a clean copy (no changes marked) for the ballot.  These are volunteer scientists and engineers who want to be sure the document reflects the current best thinking in the field with the open questions clearly marked as still under discussion.  A different subgroup currently has the problem of getting a sea of red back from the technical writers in the form of track changes and one poor sucker pointed out that he is going line-by-line with a comparison between two paper copies to determine where the technical writers changed the meaning by their edits.  We're grateful for the clarity in some cases and sigh heavily about substituting words that may be synonyms according to Roget, but have specific technical meanings that distort a lengthy discussion on why A is not B and vice versa.

[Puget]

Here are some things that work well for me. Context: I'm psych/neuro with a moderate size lab (currently 3 PhD students, lab manager, a bunch of undergrad RAs).

PDF management/citations:
I've been using Papers since grad school, which does the full workflow from searching to PDF management to note taking to citations. They were acquired by ReadCube and moved to a yearly subscription model which is annoying, but its only $60/year so I'll probably stick with it given how huge my library in it has grown (still using the older version I have a perpetual license for for now).

Project management:

We use Asana very heavily for project and lab management, and it works great for us (we use the free version). Each study and paper gets its own project using a template with all the subcategories of tasks already set up (e.g., for new studies there are columns for design and preparation, measures, participant recruitment, data collection, analysis, and papers). There are also sections for on-boarding, admin, important info, technical guides, and inventory etc. Everything gets documented and tracked there, with the relevant files and links attached to each task.

When someone needs me to do something they assign the task to me in Asana and it shows up in my task list. I do whatever it is (comment on the draft etc.) and then assign it back to them. Nothing gets lost in email, and the full version history and all associated comments are preserved. There is also a personal tasks workspace which I use for my to-do list for everything not directly related to the lab (teaching, service, outside collaborations). It integrates well with Slack, which we use exclusively rather than email for in-lab communication for similar reasons.

I do still like track changes in Word when used correctly-- the trick is to make it a rule that the lead author accepts all acceptable tracked changes each time it comes back to them, so only the new changes and continued areas of discussion remain tracked.

Final data files and analysis scripts also get archived on the Open Science Framework, along with our preregistrations.

Outside collaborations:

Most of these still happen by email, but I'm trying mightily to change this. We have tried to move one collaboration to Slack, with links to google docs and a shared Box folder. This has been somewhat successful so far (some folks are more on board with this than others).

[pgher]

I use LaTeX as well. I'm almost exclusively using Overleaf. It runs in a browser and handles collaboration pretty well. That way I can see and edit my students' work, and they don't have to set up a LaTeX toolchain.

Does Overleaf run exclusively in a browser and does that browser have to be able to call home to a server?  I ask because I'm trying to get more of our new folks to convert to LaTeX, but we do have to be able to work with no connection to the internet.

It's a web site. I haven't figured out a good offline solution.

[eigen]

My organization flow is ever-evolving, especially with a recent change of institutions, but for context I'm in a wet-lab research field at a PUI, and am usually managing a group of 3-8 undergraduate research students of all levels of competency.

Overall Organization

• Lab Cloud drive. This was a hosted DropBox, now it is migrating to Google Drive, since that's institutionally supported.
• Within the cloud drive, there are shared sheets for inventory/samples.
• Each student has a folder for raw data, and there are also broader folders for more polished/analyzed data that others might need to use.
• Each student also has a folder shared only between the two of us, where they keep a Research Log and a Literature Journal. The former acts as a rolling log of what they've done with a brief summary on each experiment, as well as a 2-week plan for what they're doing next, as well as any important dates for the two of us to remember (deadlines, conference submissions, etc.)

On a more personal level, I have tried a number of different reference managers, but keep coming back to EndNote- it's the only one on the market that handles some of the more esoteric reference styles in my field (specifically, nested references), but it also has the flexibility to let me add and sort by a lot of custom fields, which I use extensively. I do also maintain a Zotero library that is shared with the lab- this acts as a central repository of papers that gives new students a place to start reading.This is linked to an actual PDF library in the lab google drive.

For writing, LaTeX isn't as friendly for a lot of my (image heavy) work as other software, so I only use it occasionally.

I use a lot of different tools for drafting, increasingly Google Docs, but also Word and Scrivener.

For actual final projects, my preference is InDesign, for complete layout control.

The thing I have a really poor system for is notes and plans for research. A lot of it needs to be drawn out (sketching molecules and schemes for quick use vs. final publication quality figures) and so my tendency is to use whatever notebook or pad of paper happens to be handy, and then try to  deal with it later. Shockingly, this leads to a disorganized system.

Over the last year I've gotten much better about keeping all of my research ideas and notes digitally using an iPad Pro and Apple Pencil, and it's really nice to have them both searchable, all in one place, and backed up regularly.... But the writing work doesn't flow quite as naturally as it does with a pen and paper, at least not yet.

[eigen]

Also, to add a resource, I love Raul Pacheco's blog and some of his organization strategies:

http://www.raulpacheco.org/2017/12/carving-time-to-read-the-aic-and-conceptual-synthesis-excel-dump-combination-method/

[glowdart]

I use a lot of paper. And I use books, with bookmarks.

PDFs get filed electronically in a folder by project and subject.

I also type my own footnotes freehand.

I use track changes when collaborating. My collaborators do, too, because we know how to view simple or full markup and see what we are changing.

I am a dinosaur.