Cite while you write

Oct 03 2011 Published by under Science Writing

I will never, ever understand you people who "add the citations later".

Citing is an integral part of writing for me.

Discuss.

43 responses so far

  • Usually I know that a paper exists and recall what's in it but don't remember any author/year info. So I'll just add (REF) and then fill in the blank later on. But maybe that's not what you're getting at...

  • human says:

    I agree. Now, sometimes I'll put in placeholder citations (that thing where that guy said XYZ) and go fix it up with the exact citation later, but if you're not doing at the very least that - and usually more - you're probably just making shit up.

    It puts me in mind of an experience I had in an undergraduate research seminar. The professors said in response to my paper draft, "I wonder where you got this idea about [gross generalization about the secondary literature.] That's very interesting." I quickly weighed my options and sheepishly admitted, "I pulled it out of my ass." The class got a good laugh out of it, and I learned my lesson.....

  • drugmonkey says:

    I mean, click over to the damn EndNote and insert the citation. None of this "now that I'm done writing I'll go back and add in the citations" papfharnoli for me...

  • Morgan Price says:

    I usually agree, but what if you're writing an outline? Or summarizing the subtle implications of a figure? Are you really going to go trawling through your citations looking for the right one to reference (and what if it isn't in your reference database)?

  • drugmonkey says:

    Outline? Meh. If it isn't in your database? Why not? Go get that sucker.

  • queenrandom says:

    Yeah. I typically (roughly) hand outline,write the citations I need on that, then insert them while I type, although it's just a first author-year. I save screwing with endnote for last as I am inevitably stuck with a coauthor who only uses macs and/or a different version of endnote/no endnote and/or a different version of word, resulting in a wall of C&P comments on markup when citations get shifted, making the actual edits/comments harder to find, and a loss of proper citation formatting.

  • Dr Becca says:

    You seem to be implying that those who add citations later aren't already aware of which pubs back up their statements. You're clearly more of a Zen master than I am, but If I had to interrupt the putting down of words to go into Endnote/Mendeley/etc twice a sentence, I'd never get anything written.

    I also NEVER write something that I "think" is true but don't know at least one citation that supports it.

  • jekka says:

    current PI doesn't like EndNote and inserts all refs *manually* at the end of writing. The rest of us are allowed to use it though, although that permission was granted after my arrival.

  • DrugMonkey says:

    People who do not use some type of reference manager need to be kicked.

  • gingerest says:

    I usually write with my desk covered in a thick layer of the papers I'm writing from, and so I just stick in "Fred, 2009" or whatever. No matter my good intentions, I don't seem to have a single up-to-date reference database - someone's always flicking me something or I've got some oddball thing that requires hand-entry because it's not indexed in Medline or whatever. So, no, I finish the sentence and probably the paragraph and then go hunt for it and make sure it's entered and all that happy crappy.

    I don't write things I can't back up with evidence on the spot. (Which is killing me right now. Common knowledge is a terrible thing.)

  • Pinko Punko says:

    Some people don't want to disturb the flow of the masterpiece. Or even when it isn't a masterpiece, going to endnote interrupts the flow. I find I cite better when I am in "cite" mode, not "write" mode. The faster I write, the more often I cite later. When I was very slow at writing, I would do citations while writing because they weren't the rate limiting step. It is easier for me to cite all at once and spend time with the references, and when I do this I also do a better job citing because I spend extra time (that doesn't interrupt the thought process for writing the document).

    In short, I have converted from one way to the other way.

    -quuenrandom- turn off "track changes" when you do anything to references- this will alleviate your tracking problem confusion with refs getting marked up as changes.

  • I usually write in LaTeX, where I can insert cite commands without having to change windows. I have a system for tagging refs that means I'm likely to use the right tag, but need to check against Mendeley later.

    I've gone off cite while you write in W*** because the track comments function messes up the references. Bad Word/EndNote, no biscuit.

  • PerrottiSanchez says:

    "People who do not use some type of reference manager need to be kicked."

    I am stunned by the number of people in my department that do not do this and will not even consider changing their ways. Number 1 excuse: it will take too much time to transfer all of their references into the data base. SRSLY?!?!?!?
    It is after making this statement that I consider kicking the person.

  • Bashir says:

    I've been using a sorta-cite while you write method with Mendeley because I couldn't quite get the plugin working. Though apparently there's a fix somewhere.

  • queenrandom says:

    @ Pinko - thanks, but it comes from other authors not turning off track changes then sending that hot mess back to me 🙂 So I just avoid that by not doing the endnote thing until after I've heard back from coauthors.

  • I'll quite often put (ref) when I don't remember which paper said it best or first, and want to check later. Or ([lab]) when there are several papers by the same lab on the same topic.

  • You know what else I do that people around me seem to think is crazy? I organize and capture my figures first!!!! Then I write the discussion. Then the methods. LAst the introduction!!! MWAAAHAHAHAHAHAHAAAA!!!

  • Namnezia says:

    I usually make the figures first, then the results, then the discussion and then the intro. Abstract and citations go last. Methods and figure captions get written in the little bits when my thoughts aren't sufficiently organized to write the other parts of the paper. I usually insert a citation marker like (ref) or if I know the paper off the top of my head I write a little descriptor (Billybob 06) and then at the end I insert the real cites with EndNote. Otherwise it just simply disrupts the flow of writing. But the KEY is having the final figures done first, and I make everyone in my lab do this before we start writing any paper. I fucking HATE looking at crappy ass draft versions of the figures while I write. Plus the writing is so tied into the figures.

  • DrugMonkey says:

    How can anyone else who reviews your drafts know WTF you mean if you don't provide the reference you have in mind?

  • alethea says:

    Like the eminent Dr. Isis, I always make final figures and legends first. Then I write results. Then I go back and forth with discussion/intro. I leave methods for last, because they're usually easy. I put in (citation) where I know I need one, and often note the citations I need (our last paper) (patel from the '90s) or whatever. But I certainly don't stick in a full citation until I have all the words on paper, especially in the introduction, which is usually citation heavy.

  • DrLizzyMoore says:

    This current grant that I'm working on is the only time that I have written first and added citations later. The reason being is that the citations were not mandatory and part of my 10-pages. So I have to strategically add citations. With the size of computer screens right now, just have End Note (or what have you) open on one half an Word on the other....Simply, easy, done.

  • I'll add that I also cite while I write from endnote. But, multiple monitors on my desk is helpful for that. My shit looks like Minority Report.

  • anon says:

    I told my graduate students that if they didn't learn to use EndNote or some sort of ref manager, they will never graduate. Never. I also ask them to put together a list of figures, or the figures themselves, before they start writing the manuscript. I've even done this early in the process as an exercise for them to figure out what key experiments need to be done to complete a manuscript. It does help to motivate them when the going gets slow.

  • qaz says:

    Given that I rewrite everything at least three times, I don't stress citing while I write. If I know the cite I'll put it in, but going to find it if its not in my database can be time consuming and kick me out of my flow. More importantly, I want to go make sure that the cite really does say what I think it says, and that's better done when I have time to check it (i.e on a rewrite). So, I often put in a to-cite marker and then go back and fill it in and rewrite as I do. This gets (for me) the right balance between big-picture flow and little-picture get-it-right.

    But the key to writing is that YMWV (your mileage will vary). Everyone needs to find the writing path that works best for them.

  • drugmonkey says:

    Let me get this straight...

    Y'all flowbies can't be bothered to check if a paper actually supports what you think it does because it might break your groove. But somehow constructing an argument based on your erroneous memory of the lit is an efficient way to write?

  • Susan says:

    I'm with the majority here; the point is to preserve the efficiency of my writing/thought process while I'm in the groove.

    To answer your question about arguments based on erroneous memory: if you're still working on creating a line of argument by the time you're writing the paper, you're not doing it rite. For grants, OTOH, I do lit searches as I go.

    Once upon a time in the poverty of grad school when I perused Craigslist 'gigs', some guy paid me $800 to teach him how to use EndNote. Srsly!

  • To answer your question about arguments based on erroneous memory: if you're still working on creating a line of argument by the time you're writing the paper, you're not doing it rite.

    This is completely false.

  • DrLizzyMoore says:

    I need another monitor....

  • Dr Becca says:

    Dude, where did this "erroneous" business come from? You are going into this whole argument with the approach that people who don't cite while they write don't know their field. As far as I can tell, there's no evidence for that.

  • Namnezia says:

    That's right, not putting in the fucking citation from endnote doesn't mean you are not citing the paper correctly or are remembering it wrong. In fact I'll often have the paper I'm citing open on my desk or on my computer to make sure I'm referring to it correctly, but don't bother inserting the actual citation later.

  • HennaHonu says:

    I don't understand how citing upsets your writing flow... I finish a phrase/sentence, click the EndNote ribbon, click Insert, type whatever combination of the keyword, author, year, journal I know, click insert and move on. The entire process takes less than 10 seconds. I don't even have to have the EndNote window visible on another screen. If I am not sure about the citation I address it immediately. Anything I would write based on an erroneous memory would be a waste of time anyway. And re-writing? Really? I do a lot of editing, but I never have to completely re-write a paper or report. I build an outline of each section with all my citations and then fill it in to create flow between the ideas.
    I can't imagine not using EndNote to write academically!

  • That might work well for a single-author document, but speaking as someone whose job it is to fix the mess caused by multiple authors using multiple EndNote libraries in conjunction with different versions of Word and EndNote, on Macs and PCs, I much prefer to put them all in in one batch!

  • drugmonkey says:

    Cath, you need to discipline your friggin "multiple authors" into behaving properly on collaborative writing projects. Starting with one base library, which is occasionally renewed by importing new refs from the various participants and then re-shared to start anew is essential behavior. Collaborating PIs need to get on board with this too.

    Having someone like Cath, not as deeply tied into your own science trying to figure out which "Gun, 2011" ref to insert in your part of the text is not where you want to be, my friends.

  • @ drugmonkey

    If I'm saying that genus X has 300 species in it, that information will be in 300 different papers. I'm not going to stop and find the most authoritative paper that would be the correct citation to use.

    If I'm citing Author, 2011 who just did a novel experiment suggesting genus X is in family Y, not family Z as previously thought, that totally changed the game of whatever I'm working on, I'll cite it then and there.

    (And I'll put in a quick plug for Zotero now.)

  • neurowoman says:

    I can know what's true in the field whilst being unable to immediately remember the exact citation. As in, I can picture the exact figure of the experiment that shows something, but was it in Whosit, 2005 or Whosit, 2006? I'll have to go back & check, but in the meantime I insert a text with the rough idea of which paper I have in mind. These are two completely different types of information to remember.

  • Cath, you need to discipline your friggin "multiple authors" into behaving properly on collaborative writing projects.

    I'm lucky if I can get some of them to remember to copy me on everything.

  • DJMH says:

    You know what sucks donkey cites? When you spiffy up the doc with your CWYW and then go to move one paragraph from the middle to the beginning and Endnote crashes itself, your computer, and possibly the confocal microscope next door.

    So I put the refs in later.

    Don't say "Leave it unformatted!" because really, the program should serve me, not the other way around.

  • Matt Wall says:

    "People who do not use some type of reference manager need to be kicked."

    Surely they're causing themselves enough suffering and agony already! Seriously, how do they manage without using one?

    I tend to put in placeholders (REF or XXXX) and go back later - when you're flowing, you gotta flow. Man.

  • Odyssey says:

    You know what sucks donkey cites? When you spiffy up the doc with your CWYW and then go to move one paragraph from the middle to the beginning and Endnote crashes itself, your computer, and possibly the confocal microscope next door.

    And this is why I refuse to use Endnote...

    I CWYW - using Papers - it works for me. But as CPP said, there's no one correct way.

  • brngk says:

    Only one person here uses zotero? Seriously?
    This piece of software has turned me into a
    writing ninja, automatic syncing across firefox,
    scriptable keyboard shortcuts, ability to pull a
    ref from a pdf I tell it to link in. It makes citing
    so painless there's no reason to cite afterwards.

    Tried mendeley, but blech, zotero FTW.

  • Lab Rockstar says:

    I'm with Matt, that sometimes you're flowing and ya just gotta put some kind of placeholder in there. However, that's often when I inadvertently start making shit up so I usually cite while I write. Endnote sucks though. My institution uses RefWorks, which I enjoy because it's web-based and that means I can cite while I write anywhere there is internet. But I'm a grad student. Maybe when I'm faculty and have a fancy office with my own computer I can have real life citation software.

  • Anonon Scientist says:

    We'll my PI just writes and we (lab people) fill in the citations later.

  • drugmonkey says:

    Gaaaaaahhh!!!!!! Cite!!!!

    (ok, I feel better now)

Leave a Reply