LaTeX Etiquette


Here’s another post that I wished I could flesh out more had time permitted.

A few subtler points of LaTeX etiquette…

  • Quotation marks: use ` and “ instead of ‘ and ” for opening quotations, or else your quotes will look silly.
  • Bras and kets: use \langle, \rangle instead of less-than, greater-than
  • Align your equations! It’s a pain to read a calculation when the equal signs don’t line up. Advanced: if you want to be really proper, use “align” rather than “eqnarray.”
  • Use “\phantom” letters for spacing the indices of giant tensors.
  • Be careful with spaces.  For example, be consistent about how you space your reference numbers: “blah blah[13] .” vs “blah blah [13].” vs “blah blah [13] .”
  • Be careful with capitalization. Do you write “Figure 13” or “figure 13”? Do you write “Equation 7” or “equation 7” or “equation (7)”? Be be consistent.
  • Nothing says “I stole this image off of someone’s conference slides” like a pixellated image. This seems to affect Windows users more since the resolution of an image copied by Adobe Acrobat depends on how zoomed in one is to the pdf. (In Mac a copy-paste of a pdf usually preserves the native resolution.) If you really must, steal the original image from a paper by downloading the paper’s source from the arXiv. Always ask for permission from the authors.
  • Use the “hyperref” package to produce hyperlink-friendly PDFs. This is actually a really, really big help since some PDF viewers like Skim will allow mouse-over previews of an in-document reference. Thus instead of having to go back and forth with the scroll bar whenever I want to look up your references, I can juse put my pointever over “Reference [13]” to get a mini-window that shows the relevant section of your bibliography.


This is an old post that I never got around to fleshing out and finishing. I figured it was worth posting before letting my blog freeze-out (yeah, that was a Boltzmann equation reference).

More than half a decade ago I was sitting in the same course and had a very inspirational TA and now I’ve been toying with the idea of TA’ing the “honors” freshman physics sequence at my new university.

Along those lines, I’ve been thinking about a few ideas that I don’t think are emphasized enough in the typical undergraduate curriculum. These ideas, I think, are important in developing a healthy ‘physics intuition.’

  1. Symmetry. The use of symmetry to solve problems, group theory as the language of symmetry. E.g. how to we `intuitively’ use symmetry to reduce higher-dimensional problems to lower-dimensional problems (e.g. polar/spherical coordinates). Mention symmetry as a `deep physical principle’ in gauge theories, for example.
  2. Geometry. Emphasize the geometric foundation of physics, even in elementary physics. E.g. thinking of cross products as areas with an application to Kepler’s laws and angular momentum. This is a recurring theme that I think needs to be made explicit much more at every level. Lagrangian mechanics courses ought to draw more on the structures of differential geometry.
  3. Dimensional Analysis. Every freshman should understand the power of dimensional analysis and scaling. (I believe there’s a very nice textbook by Barenblatt.) More advanced students should see how dimensional analysis is still a very powerful tool, e.g. Stevenson’s excellent “Dimensional Analysis in Field Theory” review (
  4. “Duality”. E.g. Electromagnetic duality. More loosely, being able to interpret one problem in terms of another problem. e.g. hydrodynamics as E&M.
  5. Back of the Envelope. Learning to make good order of magnitude calculations.

Yeah, I’m supposed to have written my “goodbye” post by now… but things have been really busy. So instead, here’s a link to the PiTP 2008 lectures (recently released):


[Unfortunately things got busy before I had a chance to properly wrap up this blog… I hope to write up a proper `farewell’ post over the next couple of weekends.]

  1. If your paper is more than 30 pages long with multiple sections and subsections, it would be prudent to include a table of contents on the version you submit to the arXiv.
  2. If you are citing just one particular detail in such a paper, it would be very helpful if you included some note about which section (or at least page number) you expect your readers to read. This can be done using the “NOTE” tag in BibTeX.

Doing these things would be a huge help for people reading your papers. They would require a minimum of extra work on the authors’ part and can easily be edited away if a journal doesn’t like it. (Comment out the table of contents line and the journal’s BibTeX style file should already ignore “note” tags if they don’t like them.)

Note that the problem of not having a table of contents is somewhat ameliorated by using a pdf viewer with a “table of contents” panel, though this doesn’t really help if you’re looking at a print out, does it?

Just a random thought for the day:

I spent a couple of years as an undergrad doing research in a cond-mat/materials science building. It was the only place where people would consistently wash their hands before they went to the restroom.

One of my favorite sources of background noise is Chicago Public Radio’s This American Life. The show recently replayed a nice episode called “A Little Bit of Knowledge,” which included a really nice piece on a physics crackpot (without being condescending). The piece starts at around the 31:10 mark. Here’s the blurb:

Bob Berenz had a good job as an electrician. But he wanted to do something bigger. He came up with an idea for an invention. But as he studied physics texts to see if his invention could work, he happened upon the biggest idea of his life: a revelation about physics that would disprove Einstein, and Newton. That is, if Bob’s right. Bob’s friend, Robert Andrew Powell, reports the story.

It’s a cute piece from a 3rd-person’s perspective, but a crackpot is a crackpot in any reference frame except their own.

[Kudos if you got the Schoolhouse Rock reference.] I’ve gotten a few messages now about the Graduate Junction, a social networking site developed to help postgraduate students. After noting it was endorsed by Durham, I decided to give it a try to see what it’s all about.

Update: see Dan’s comment below for further discussion.

The site’s goal is to facilitate academic networking, which is often an under appreciated aspect of academic professions. Users have a CV-like profile page describing their research interests and accomplishments. They are able to join groups reflecting their research and post to forums for advice about all aspects of PhD life. Users can send messages to one another and even (in the future) set up a calendar for conferences.

Does it sound familiar? It should. In many ways it’s just an academic version of Facebook. (“Academic” here means “without embarrassing photos.”) In fact, the whole interface can probably be reproduced on Facebook.

I support the community that Graduate Junction is trying to foster, but I have my reservations about whether it can successfully take off. It requires a large number of users to be useful. This is very much like Wikipedia. Why does Wikipedia have no competitors? Because of economies of scale. (Look it up in an econ textbook.) If there were two Wikipedias, each one would be worse off because they would only have half the capacity to develop articles.

In the same way, a lot of the social features in the Graduate Junction already exist elsewhere for theoretical physics PhDs, making it largely redundant:

Why not Facebook? Call me old fashioned, but I don’t think that searching for collaborators should work the same way that one searches for a new iPod. It should be more like finding the right pair of shoes: while reputation is a start, you actually need to try it out a bit with the guidance of an adviser. This is why we give talks and why conferences have discussion time. There’s much more to a good collaborator than a shiny CV. Research projects need to be taken seriously so that we can’t afford to find collaborators the way one would find an on-line date (which is itself is already somewhat dubious).

So with that I’d have to say that Graduate Junction still needs to develop a larger user base and determine what “new” feature it brings to the table before it can become useful.

For example, they could try to copy Facebook’s “poke” feature, only reimagined as “scoop.” (“You’ve been scooped by ___. Would you like to scoop back?“) 🙂