[PLUG-TALK] Computers for linguistics work

John Jason Jordan johnxj at comcast.net
Tue Apr 8 20:12:18 UTC 2008


On Tue, 8 Apr 2008 11:22:55 -0700
"Rogan Creswick" <creswick at gmail.com> dijo:

> On Tue, Apr 8, 2008 at 10:45 AM, John Jason Jordan <johnxj at comcast.net> wrote:
> >
> >  1) the PSU Department of Applied Linguistics woudn't know what to do
> >  with a (La)TeX document, not that it matters, because work must be
> >  delivered in paper format anyway,
> 
> Generally the LaTeX source is for you and a journal (see brief note
> below about how the separation of content from styling is a Good Thing
> for them too), everyone else gets a pdf / postscript.  They can't
> easily modify it, but they can easily print/read it.

Actually, I have never submitted an article to a journal myself and it
will probably be a long time before I do, if ever. But I have submitted
manuscripts to editors in the past and all they want in the way of
formatting is regular, italic, bold and bold-italic. Everything else
they are going to format with their house styles. Thus, the author has
no choice but to "separate content from styling." When I refer to doing
the formatting as I go along I am talking about stuff that I am going
to produce and print myself.

> >  2) I don't know of any peer reviewed journal of linguistics that
> >  accepts anything other than Word or OOo,
> 
> <shudder>  I don't sleep well if/when I have to submit a document of
> that degree of importance as a word doc.  There are entire companies
> based on the need for tools that go through and find all the "hidden"
> content in word docs (old trac changes stuff, text behind images,
> content that was deleted, but still exists in the file for performance
> reasons... etc.)  Not to mention the general modifiability of the
> content (which is occasionally a requirement, if the journal needs to
> do substantial formatting to paginate the articles--but gee, wouldn't
> that be easier if the *content* was separate from the *styling*?).

Do not fear, I don't even have Word installed on any computer I own.
If, for some bizarre reason, I need to submit something in .doc format
it will be produced in OOo Writer and exported at the last minute to
Word format. Or PDF. Or RTF. Or something else, but there is not going
to be any of the junk in it you are referring to.

> > There is an alternate notation system, like:
> >
> > [CP[TP[NP[D The][AdjP[Adv very]][Adj small]][N boy]][VP [V
> > kissed][NP[D
> > the][N platypus]]]]]
> >
> > But I think y'all can see how much faster and easier it is to grasp the
> > structure of even a very simple one-clause sentence like the above
> > with a visual tree.
> 
> Sure -- but assume (1) you know what content you want to include and
> (2) the notation above was *all* you had to enter to get a properly
> formatted visual tree in your document.  Is it really slower to draw
> out the tree on a whiteboard / paper to figure out the short syntax
> and enter it into your file than it is to try and align everything
> poperly visually, and then deal with generating an image, inserting it
> in the doc, adjusting the sizing and placement, ensuring that it
> doesn't cover up text, or get bumped over another image, etc... ?

No, brains of linguists don't work that way. Consider this: To write
that notation above I had to draw the tree first. Only when I could see
it visually could I produce the notation. OK, that sentence was simple
enough that I might have been able to generate the notation without
drawing the tree first, but it was a lot easier to do it by drawing the
tree. Trees are easier. 

And no, I don't have to do the alignment and stuff manually. That is
what TreeForm does for me. I can see the tree taking shape as I drag
and drop structures from the toolbox onto the tree. The structure just
pops in already formatted. If I drop a structure between other
structures the lines automatically lengthen to make room for it. 

> >  I should also mention that presentations are almost as important these
> >  days as paper documents. For example, this term I have to prepare a
> >  one-hour presentation on the FOXP2 gene, using the electronic equipment
> >  in the classroom. I'll use OOo Impress to create the slide show. Then
> >  I'll export and upload the show as a .ppt file to my H drive at the
> >  university, from where I can run it on the Windows computer in the
> >  classroom.
> 
> The *safest* route I've found for presentations is to create a pdf
> file, and run acrobat in full-screen mode.  It works great on every
> platform I've tried (linux, windows, OS X, and iirc, solaris and
> hpux).  The biggest benefit is not worrying about the minor
> differences between OO.org version X, version Y, version Z and the 3-5
> versions of PowerPoint you're likely to encounter.
> 
> That said, I haven't set up a great workflow for doing presentations
> in LaTeX -- there are ways of doing it, and it's not that hard, but I
> do find a tool like powerpoint to be useful.  (I just print to ps/pdf
> and use that for the actual presentation.)

I have done PDFs for presentations too, and I agree that there are
great advantages. However, as far as I know you can't embed sounds and
movies in a PDF. I cannot do without sounds. And flashy things are
sometimes nice too. Officially we are not supposed to be graded on
anything but the content, but the truth is that nice formatting still
puts the professor in a generous mood.



More information about the PLUG-talk mailing list