Is emulateapj worth the trouble?

by saurav on September 19, 2012

Do you use the emulateapj style file when submitting to astro-ph or circulating drafts to your co-authors? Or do you find it to be a nuisance?

I have been informally asking this question at conferences (and otherwise) for a few months now and have encountered a mixed response.  A plurality seems to prefer papers in ApJ format, with the main reasons being saving paper and easier reading due to not having to flip pages (or scroll down) to the figures. This was especially true in the case of lengthier papers; astronomers (who participated in my poll) were significantly less likely to print a long paper that is in manuscript format (either double-spaced or single-spaced).

People who read mostly on their laptops/iPads tended not to care either way as they could zoom in and out when needed. However,  as most PDF readers don’t provide the “Back” function, their lives were simplified when the figures and small tables were embedded in the text.

When reading drafts of papers on which they are a co-author, more people seemed to prefer manuscript format: it is much easier to mark up. There were many who did not care.

Proponents of emulatepj also liked to argue that journal-formatted papers also look more professional and were the worth the time investment. My own experience was about a 2–3 hour learning period for my first paper, which was 20+ pages, and very little for subsequent papers. In other words, I am claiming that if you know LaTeX, switching to emulateapj is rather easy (thanks to Alexey Vikhlinin at the CfA) and well worth the trouble.

I happen to be among the group that prints papers that they want to digest. Reading on the computer screen still tends to be for perusal.  So I am fine with killing some finite number of trees, but I do not want to print 30+ pages of a single paper. That is especially true if the paper is in manuscript format. I like to think I should read it on the computer or once the journal version is available. (Unfortunately, neither happens as often as I would like.)

It is safe to assume that long papers in emulateapj format are also less likely to be read. My own magical cutoff seems to be 15–20 pages in journal format, above which I either skim a lot or read only if particularly relevant.

This poll has been pretty unscientific; my own biases might have determined which responses I remembered. So my (informal) survey continues here, in a more scientific manner: What do you do when you submit to astro-ph? As a reader, does the tiny font size of the ApJ format make it harder to read the paper or do you appreciate the time the author invested into formatting the paper?

Does the formatting affect whether or not you read a paper from astro-ph?

View Results

Loading ... Loading ...

{ 22 comments… read them below or add one }

1 Karen Masters September 19, 2012 at 6:07 am

I’m much more likely to read a paper off astroph if it’s nicely formatted.


2 Médéric Boquien September 19, 2012 at 7:02 am

I am in favour of emulateapj or whatever 2-column format used in major journals. There are several reasons for that:
* I find it very annoying when I have to scroll repetitively to another page to look at a figure mentioned in the text. I very much prefer to have the figure and the text side by side so I have all the information directly accessible. This is especially useful when the figure is described/interpreted in the text rather than in the caption.
* I never felt the need to have a double-spaced format to write comments that will probably be typed on the computer at some point.
* As mentioned, the referee, doubled-space format is quite wasteful in terms of pages. Beyond the ecological argument, there is also a financial one. Printing is not that cheap, so if we can save a few pages here and there, the savings may be quite significant at the bottom line.

In the end, I have always put my astro-ph papers in a 2-column format. For a handful of particularly large papers in referee format I have even gone as far as downloading the source and recompiling them in a 2-column format to make them easier to read.


3 mihos September 19, 2012 at 7:49 am

Definitely emulateapj


4 Victor September 19, 2012 at 7:51 am

I agree completely with Médéric on the issue, and for the papers that I absolutely need to print and are in referee format, I download the source and change to emulateapj. And really, it is not so difficult to format a paper more or less decently, so I wish everyone uploaded to astro-ph in two-column formats…


5 Adam Ginsburg September 19, 2012 at 9:50 am

Emulateapj, for all the reasons above.

Saurav, you should make a genuine poll (e.g., use Google docs forms).


6 saurav September 19, 2012 at 2:17 pm

Done. There is a Wordpress plugin that made my job easier. Thanks for the help, Kelle. Let the voting commence.

7 Pauline Barmby September 19, 2012 at 10:55 am

I rarely read articles on paper. I agree that having figures near the relevant text is preferable, but I really dislike the back-and-forth scrolling necessary when reading online in two-column format. So my preference is single-spaced “preprint” format, with thoughtful figure placement.


8 Ian Crossfield September 19, 2012 at 11:04 am

My two cents: use preprint formatting when posting a paper to arXiv that has not yet been accepted to a journal (see discussion) and use official journal formatting when posting the revised, accepted manuscript.


9 TMB September 19, 2012 at 12:50 pm

Definitely emulateapj. I can’t stand single-column format for anything other than early drafts.

But it would be nice if there were an official style file instead of emulateapj.


10 mihos September 20, 2012 at 7:03 am

Actually it looks like ApJ and AJ now accept papers submitted in emulateapj format with in-line figures This is good; I always work in emulateapj from the start of a draft, and hated having to take the time to degrade my papers to the lousy “preprint format” before submission. Now if only we could submit our figures as pdfs instead of eps, life would be good!


11 Jessica Lu September 20, 2012 at 6:37 pm

Excellent! I didn’t realize this. Anyone encountered any problems with submitting using emulateapj?

12 saurav September 20, 2012 at 10:19 pm

@Jessica, my last paper went through in the emulateapj format. I did not realize the I had forgotten to *degrade* the paper till afterwards. I had not realized that was now acceptable either.

13 Grant September 20, 2012 at 9:28 am

I definitely prefer two-column format with in-line figures placed near the relevant text. I really dislike reading 50 page single-column manuscripts with figures placed at the end.


14 Chun September 20, 2012 at 9:45 am

I always write papers in emulateapj. It is much easier to read. I always have figures within text, and do not like it when figures are at the end. I even take it up one notch and have the hyperref package in my preamble so people can click on links to figures, tables, and references.


15 Adam Solomon September 20, 2012 at 8:22 pm

As a frequent computer reader, I’d echo Pauline’s comment that it’s a minor annoyance to have to scroll up and back down every page in two-column format. However, far more insidious is a single-column paper with figures and tables placed at the end. Slight edge to single-column in my book, but massive one to any format which puts figures and tables roughly where they appear in the text!

And I’d also echo Saurav’s (implied) cries for all PDF readers to install back buttons!


16 Ben September 20, 2012 at 11:25 pm

I refuse to read a preprint unless it is at least 10 pages long (shorter papers are always irrelevant) and typeset in Comic Sans.


17 Steve Croft September 21, 2012 at 4:33 am

You can specify single column in emulateapj to avoid the scrolling up and down. I’ve had problems with emulateapj breaking cross referencing though (\ref links to \label marks showing up as ??).

I had no idea people were so vehemently opposed to reading papers in preprint format. Then again I had no idea so many people still printed out papers to read them.


18 Ned Wright October 5, 2012 at 9:17 pm

I think referees should refuse to review papers in the manuscript format. The text, references, figures, tables order is entirely useless. So just send it back, and ask the editor to provide it with the figures and tables properly placed. That said, I don’t much care whether it is two column, or one column single spaced, or the double-spaced one column manuscript format. Trees care, but I don’t.


19 saurav October 10, 2012 at 12:47 pm

From the AAS Electronic Announcement #237 – October 10, 2012:


It is not necessary for figures to be grouped at the end of
manuscripts submitted to the AAS journals. In fact, most referees
prefer that figures be embedded in the manuscript. In consideration of
this preference, we ask you to help the referees of your papers by
compiling your manuscripts with the figures in place when they are
first referenced.

20 Alex Conley October 9, 2012 at 11:09 am

I am all for posting papers to emulateapj when possible. But I’ve had a few papers that simply will not compile with emulateapj even locally. When you couple that with the fact that arXiv seems to use a ‘unique’ version of emulateapj, there have been times when, after a few hours of trying, I have given up in disgust and posted in aastex form.

It would be really nice if aastex had a more compact supported form that we could use instead.


21 Don Terndrup July 14, 2013 at 8:37 am

Somebody ought to write a style file that creates good formats for iPads and other e-readers. Then people should be able to post in more than one format -or- arXiv should be able to generate multiple formats from source. The double-column format is a royal pain on some readers.


22 nicholas suntzeff July 14, 2013 at 4:14 pm

I frankly don’t care. Having the figures and tables at the end is not convenient, but we do have modern computers these days. I merely open up a second copy of the paper in preprint format and have that copy on the figure sections as I am reading. These are minor inconveniences!


{ 2 trackbacks }

Leave a Comment

Previous post:

Next post: