Slashdot Mirror
Fidus Writer: Open Source Collaborative Editor For Non-Geek Academics
johanneswilm writes "While writing my Ph.D in anthropology I found out it's almost impossible to get non-geeks to help me with editing my thesis because it was written in Latex. Lyx is almost there, but as it's not web based, it's difficult to use for online collaboration. Writelatex.com is online, but typing LaTeX code is a no-go for non-geeks. Google Docs is web based and near-WYSIWYG, but lacks support for professional print formats such as Latex. The Ph.D took longer than expected, so before finishing me and three others were able to code an entirely new editor: Fidus Writer: web based, open source (AGPL), almost-WYSIWYG and with tools for academics such as citation management and formula support and output formats PDF, Epub, Latex, HTML."
Do your job or give up on it, jesus christ.
Are you calling for the apocalypse, or is there something I don't know about the Slashdot "editorial" staff?
Why not just create your content in something that all of your non-geek friends are happy to use (Word with track changes, for instance) and then spend a short time formatting it when you're done writing?
"I zero-index my hamsters" - Willtor (147206)
In the late 90s (yeah might as well be 12,000 years ago) i did my PhD in Latex; but, two of my Profs *insisted* on editing bits (in retrospect, mostly adding pointless elaboration) with MSWord. (one committee member didn't even understand how any document wasn't the same as a "word file"). So, i learned to use rtf2latex (and similar tools for bibliography and index), back and forth; email RTF, convert it back, (if the addition was over a three sentences). It wasn't bad at all when one learned maintain all tables and figures as merely included files. It was all worth it when the thesis committees, who check tedium like margins to within a millimeter and equation formatting, passed my thesis right off; while my grad-school mates that used MSWord all had their format go wrong when the "approved font" was applied.
I gave it a try, installed the source, followed the instructions, opened the page in a browser and... was greeted by an error message telling me it needed a 'Facebook app' or some other social drivel.
Facebook? Are they serious? I opened a bug report:
I thought to give Fidus a try by installing the source and following the directions. When I tried to log in to my freshly minted server it told me I couldn't because I had not configured a 'Facebook app'. Looking through the Django config page I noticed it only gives the options of using Facebook, Twitter and Google. Neither of these are acceptable in any environment which has even the slightest respect for an author's privacy and confidentiality.
Please make it possible to use Fidus using either a 'private' 'social' 'app' (lots of quotes there, for different reasons) or by foregoing on the social fad entirely. Since Fidus seems to be about getting work done I don't see the need for more 'social' distraction anyway.
--frank[at]unternet.org
A book I published almost two years ago underwent a similar, painful process. I (as the coordinator) had it all typeset in LaTeX. It was not perfect, but it was beautiful already. The university (social science research institute) sent it (the PDF I gave them, as per their request) to the style corrector. I got back... An ugly MS Word document with some corrections included in it (but not version-controlled or anything like that, not even MS Word's sorry replacement for a real version correction). Merging that back into the original was way beyond painful.
The second round of style correction was, fortunately, done different. I was able to work through the process with our editor, fixing some details out of common aggreement (and not only accepting their changes as during the first round, where I even spotted places where the style corrector misunderstood and even reversed the meaning of some fo the sentences). The second revision was not a piece of cake, but it made me learn quite a bit about editorial reasons and aesthetics, and had me way happier at the end. And the editor even learnt a bit about LaTeX as well.
The Latex-lab guys have LaTeX for Google Docs working very well. Our maths faculty use it all the time. See http://docs.latexlab.org/docs Source http://code.google.com/p/latex-lab/
>"The Ph.D took longer than expected, so before finishing *me* and three others were able to code an entirely new editor"
A Ph.D that makes a mistake that glaring is a sad thing.... we are talking very basic English.
I vaguely recall that early on in my philosophy career, I produced a lovely manuscript in LaTeX. The journal insisted that I convert it to Word. I put the effort in to do that. Finally, I get the galleys, with my Word file typeset by the journal's typesetters in India. And it was obviously LaTeX that they used at the typesetting end! I was annoyed. But eventually, I just switched to using Word for most things.
However, more recently I've gone back to using LaTeX for a fair amount of my philosophical writing, partly as my writing has got more technical. I've noticed that many journals accept LaTeX as is (I don't know if that's a new thing). Some do require Word. But my thinking is that I typically don't know which journal the paper will end up accepted by, LaTeX is more fun to write in, the paper is easier to adapt into a Beamer presentation (I've found Powerpoint too difficult and cumbersome), the manuscript will look prettier to referees for whatever that may be worth, and if I need to do one final conversion after acceptance, that's annoying (it can take a while, as I have to go through the text sentence by sentence to make sure nothing was screwed up) but not a very big deal.
And perhaps most importantly, if I use LaTeX, my content and style aren't biased by the limitations of Word. For instance, it would be a big nuisance to include a Fitch-style formal logic proof in a paper in Word. So I probably wouldn't bother, even if doing so would help the reader. Likewise, perhaps throwing in a formula or some symbols with subscripts would be stylistically optimal, but because these things are harder to type in Word than in LaTeX, I might not bother ($x_2$ is more natural for me to type than ctrl-i x ctrl-i ctrl-= 2 ctrl-=, and with LaTeX you don't have the problem that if in later editing you later try to insert a comma after it, Word wants to subscript the comma).
Moreover, for collaboration, plain text formats work very well with svn (there are no do doubt better rcs's, but svn is what I'm used to) as I and my coauthor can easily view the latest diffs, either from the commandline or the web. I suppose Google Docs has nice good collaboration features, too, but they aren't an option for me as Google Docs doesn't have the automatic cross-referencing features that Word and LaTeX have and that I tend to rely heavily on (I just tried Fidus and couldn't find cross-referencing for numbered lists, nor a way to make the numbering resume after an interruption of a numbered list).
So, yes, even in the humanities it can be worth using LaTeX, though admittedly much of my work is on the technical end of the humanities (e.g., I prove not entirely trivial theorems).
Yes, you have quarks to work with, but they are well known.
And, luckily, there are only six known types.
Right tool for the job, stop pounding nails with a hammer, etc.
Latex is typesetting tool. It's designed to take final materials and make them look good in print publications.
Word is an editing tool. Word "processing", as they say. It sucks for any sort of high-quality publication, but you can get what you see on your local inkjet printer.
Word -> Tex is used by real publishing houses as a final step. Unlike the typical PhD student, they have mastered the concept of "division of labor" and don't typeset a document until it is finalized.
The problem is Word is not the lack of "print ready". It is that Word's collaborative editing model is firmly rooted in the 1990s model of sneakerware and emailing documents. And kludging SharePoint on the side doesn't really change the nature of the beast.
The endgame here is a collaborative web application, which provides Word-style collaborative editing capability while able to export structured markup to a Latex-style typesetting program. A healthy dose of user-friendly Git-like functionality is needed too. Google Docs is like WordPad.exe compared what could be done in this space.
Hello, I've been writing books for close to 20 years. In addition, for most of the last decade, I've also been a co-maintainer of ~30,000 pages of technical documentation for a well-known family of Open Source software products (one of which is used on Slashdot's backend). This documentation is updated and re-published in toto on a daily basis, in about a dozen end-user formats.
If you're an author, then you're supposed to be writing meaningful content. This means that you should be concentrating on data and semantics.
Presentation and layout should not be your concern--leave this to the professionals (editors and layout people).
Otherwise, use DocBook XML and MathML to author your content, then transform to PDF, RTF, Word, HTML, or whatever end-user format(s) are required using the appropriate toolchain and transforms. There are heaps and heaps of XSLT stylesheets out there for this purpose. You can tweak these as desired/necessary, and it's at this stage--and not before--that you should be even the slightest bit worried about how things look.
If there is one thing that many years in this game have taught me, it's that futzing with presentation issues while you're trying to write merely serves as a huge distraction. And that it is counterproductive to reinvent the wheel for every writing project, which is what formats that munge together content and presentation at the expense of semantics invariably force you to do.
I know it's fashionable around here to disparage XML, but text + semantic markup + styles/transforms works very, very well for producing dense technical material that preserves semantics while providing an easy way to publish something that's pleasing to the eye. For the last 10 years or so, I've refused to use anything else for this purpose. I strongly encourage anyone who's planning to write anything over a few paragraphs in length to check it out.
As for collaboration--why do you even have to ask? Pick a revision control system and use it. Depending on the project and who I'm working with, this would be SVN or BZR for me, but there are many choices. Choose one of them.
Il n'y a pas de Planet B.