@manooooh the measurement of "efficiency" seems spurious but it is also irrelevant as I do not want a Word document as the end result, so being able to efficiently make such a thing is not at all interesting.
@manooooh I mean start writing the document then find two co-authors to edit the same document, one using a mac, one linux, and share the authoring. that was the first part...
@DavidCarlisle you do not want a document made with Word? Tell it to my teachers, to the 90% of normal people that use a program to make a document in Argentina; as long as they can fulfill what their superiors tell them, it will be all right. I even have colleagues that using Word are not able to render an equation
@DavidCarlisle I do not understand what you mean by "find two co-authors to edit the same document". I start writing a document, then what is the next step?
@manooooh ask me to write the next section, using linux, then ask someone else to write the sentence after that, using a Mac, or preferably have the document in source control such as git or svn so all the people can work on it.
@manooooh -- This article contains many multi-line displays, a figure (simple diagram), several tables, and a number of citations. How about looking at display equation (4.11) through the (unnumbered) display that follows, not entering the number of the \eqref that occurs in the text as a literal value. The simple proof before Lemma 6.2 is also nice, with a cite.
@manooooh Word documents are the problem not the solution, they are fragile binary version-specific blobs that mean that any information in them is really hard to extract.
@DavidCarlisle well, I would write the text and in a new paragraph, with color, I would write what he has to do. Even in Google Drive you can comment on parts of the text without any difficulty; is already established. Can you do that with LaTeX quickly and easily? I do not think so...
@manooooh Still doesn't do typography, is very limited in features, etc. Clearly, for many people that's fine. But it's not enough for good typesetting, or for lots of features.
@manooooh google drive if you prefer (although that doesn't really have full version control) but given that I can't have Word on linux every time I edit the document in libr office or some other clone the entire document will be converted (if possible) and then if you edit it it will convert back.
@manooooh Word has its uses, my Mum uses it to write letters, my Son uses it occasionally to do homework, I can't think of any situation when i have ever wanted to use such a system.
@manooooh -- Another consideration. Write a textbook ... It proves to be popular, but fiver years later there is new information, so a revised edition is in order. Retrieve your old files and use them as a starting point. Three years later, do it again.starting with the files from the second edition. And so on.
@JosephWright again, a mathematician or more generally a scientist should not spend his time in learning formats and good writing techniques, but to produce good mathematics, which is very different
@DavidCarlisle Google Drive is a page and you can enter it even if you are a Linux or Mac user
@manooooh 90% of people are not the subject of that study. the question is what is a suitable system for long term archival digital academic documents. Word is not it.
@manooooh google docs I think you mean, not drive, yes I know what it is, it is completely unsuitable for a mathematical research paper.
@manooooh It does depend what you mean. My thesis was written in Word: I'm a chemist, and the amount of maths is very very limited. And Word gave me problems. And it's given other people I've known problems. Problems that won't occur in LaTeX.
@manooooh 30+ years ago, most theses were written on a typewritter and you'd be right. But computers mean that production standards have changed. What's expected has altered, and that means people spend ages in Word fiddling with appearance.
@DavidCarlisle what about doctors, biologists, police, firemen, distributors, students? They all care nothing that the document looks nice, and rightly so, because they have more important things to do
@manooooh Try writing a 200 page document in Word with cross-references, figures, pastes in from multiple programs, citations, etc. Word really doesn't cope well. That's nothing to do with the appearance.
@manooooh That's simply not true. People do care how things look. And sometimes it affects how easy they are to parse. It's a matter of degree: how important are different factors.
5
@manooooh When someone sits down to write a thesis, they'll do anything to avoid actually writing the text
@DavidCarlisle yes, I mean that, sorry. And yes you are right, but do not pretend that a mathematician should learn format rules because that is not his job
@manooooh If you are writing archival documents that you expect to be used unchanged on 50 or 100 years, that should be enough to think of using a different system.
@manooooh I had to print off a thesis for someone about two years ago. They had some dodgy Word cross-refs that would not work when printing, even though the pages looked fine. It took me a whole day to fix, which in the end meant re-keying the text on those pages by hand. I was not at all please. And the document is clearly hosed.
@manooooh See the comment I've just posted. In a long document, you need cross-refs. It's bad news if they fail
@manooooh A colleague helped write a book a few years ago. It was built up from several authors work, and was about 1000 pages long. One chapter, around 200 pages, used EndNote for references. Copy-pasting some stuff broke all of the references, which had to be fixed by hand. This person now refuses to touch EndNote stuff.
@JosephWright my teacher of Discrete Mathematics did not care if the document contained rendered formulas or not, and that subject is about mathematics and figures. I cannot say the same because here people do not care if a document looks good or not
@manooooh I've seen documents that cannot be combined into one as the images were simply too big. That's problematic if you need to make a TOC: I think we ended up doing it all by hand.
@manooooh Point there is that if the source is TeX, it can be used 'locally' then re-used by a publsiher. If it's Word or whatever, much less easy. So again, it depends on your target audience.
@DavidCarlisle if you say that the people who should worry about the format of the document should be editors, I totally agree. But you can not steal time from a mathematician to learn rules about formatting that are conventions and that he does not care
@manooooh It depends, though: it's not a hard line between formatting and meaning. For example, as a chemist I care a lot about the difference between C^2 and C_2!
@manooooh For example think about tables. A well laid-out table makes it easy to comapre values. That's partly appearance, but it impacts on the conclusions you might draw. So it's not just 'some layout thing'.
@manooooh -- "A mathematician should not learn format rules." That is why a serious publisher will create a document class with necessary and proper structure so the author needs only to use the provided environments and create the content. Everybody wins.
@JosephWright I have 20 and in my high school the documents were 10 pages average. In my university there are no documents with 200 pages, and the books that I have to read are not written in LaTeX
@manooooh sorry I do not understand, nor do I mind much either way. If you want to use Word that is fine, it is there and lots of other people use it. I think if you use it to produce digital documents that you want to be archival and used in many years time you will be disappointed, but it's your choice.
@PauloCereda -- Yes, at least mostly. Unfortunately I was not able to save out much of my accumulated address list before the plug was pulled. (I think i lost your address in the process; I think I will be able to recover much of it later, but at the moment, things are pretty much bare bones.)
@manooooh Oh come on, a thesis is typically at least 100 pages in most subject areas: talk to geologists, they go to around 400. Books are typset normally in something dedicated, be it LaTeX, InDesign, Quark, ...
@JosephWright yes, that could causes errors. However, I do not have the enough experience to use OneNote. For that reason I am asking you to create an example (realistic) in which these errors can exist, and I will try to convince myself. I do not know anything about OneNote
@JosephWright it is actually easy; use Google Docs
@manooooh I'm really no sure what you are trying to argue. For the majority of people in the world, a word process is going to be fine. That's not in question. But when one wants high-quality typeset output, some other tool is needed. LaTeX is one such tool; it's certainly not the only one.
@manooooh No. A publisher will still have to extract the content to their tool, for example 3B2.
@manooooh I never mentioned OneNote: I said EndNote
@manooooh google docs has no reasonable support for mathematics, and a primitive editing interface, it's fine for a one page letter the idea of using it for a thesis is scary.
@JosephWright that is not format. You can type C_2 or C^2 on Word and you can edit that as many times as you want. I mean "the document is nice or not", and I do not mean errors of that style
@manooooh Have you sat an read many papers? That's part of my job. I much prefer the typeset versions in PDF to the HTML or Word 'pre-prints'. They are just easier to read, more consistently formatted, etc. The content is the same (more or less), it's about being easy to follow.
@DavidCarlisle Yes
@manooooh Right. So please believe people who do have some experience. This is part of my day job. I've had students using Word, and students using LaTeX. Word really does struggle a lot with big documents. But it was never designed to produce them, at least in any real sense.
@JosephWright is it really hard for you to understand a document in which the first line is blank and in another follows? (Like the example you wrote it)
@manooooh It's a little thing, but it makes a difference. I notice more bad word spacing in word processed output: that does break the flow and make reading slower.
@DavidCarlisle what kind of math are you pretending to write? Of course math has to be written in LaTeX! But do not pretend that a mathematician knows the difference between 1. (a) and 1. \\ (a)
@manooooh my objection to using Word is nothing to do with appearance, I do not want ato end up with a proprietary binary blob of data that may well be unusable after the next version update of software over which i have no control 9and have not bought).
@manooooh there are no theorems or proofs in a typical software manual.
@manooooh I'm not sure what your aim is here. LaTeX is good at some stuff, and if you care about that stuff, it's worth the effort. If you don't, and you don't have to use it, then don't.
@manooooh My day job is largely Word-based: almost all of my academic work has been done in Word, it's pervasive in my subject area, most of my colleagues don't know LaTeX.
@JosephWright it is true. You dedicate yourself to one thing and I to another. You can not expect a mathematician to separate the word exactly as the AMS requires. You can not put a bad note to a student because he did not use LaTeX or because he forgot to correctly separate the word at the end of the line; you must concentrate on the content of your document, in its essence. That is what I would do, that I should not tell you how your work is haha :P
@manooooh In my own case, I've made it clear to my students that I'll help them with LaTeX or Word, and that the choice is ultimately their own. They have to be comfortable with the tools they are using and the outcomes they get.
@DavidCarlisle ER, to some extent. Probably maths is rather different from chemistry in this regard: I'd be chosing a journal based on reputation/likelihood of being accepted/impact
@JosephWright yes I noticed that, not sure what happened to that version, title (not to mention the author) is wrong, oh well.... still I wouldn't have wanted to make that in Word:-)
@DavidCarlisle I don't know how many maths publishers there are. In chemistry, depending on exact area of interest, I can think of five worth any consideration. For my own work, I'd cut that to three (OK plus Nature and Science if one is really lucky)
@JosephWright of course. I love LaTeX. A mathematician who wants to write a proof that he considers important should not care if it is written in Word or LaTeX, because most of the problems are questions of style (an enumeration starts below, the theorems have a cursive format, etc.). In fact here in Argentina. 80% of my university's exams are done with Word. you are referring to an editor's work, and I think it is nice that you correct those things
@DavidCarlisle I guess. But that's different for maths to other subjects. For me, the old print or now PDF version-of-record is absolutely fine. We never expect to re-use text 'as is' in the way a mathematician might.
@manooooh No. I would like not to have to use Word templates with multiple columns. And I'd like not to have to do admin work in Word ...
@DavidCarlisle of course not, since my PC slows down when I exceed 10 pages of a document based on 60% mathematical expressions. In that case is preferable to make it using LaTeX
@JosephWright that document has text that was written 40 years ago, routines get added or withdrawn but some stay and their documentation lives on, that text has been in tssd in dynatext, in SGML in xml+tex maths and (now) in xml with mathml maths, and it has survived the introduction of pdf and other web and other changes. The one thing it has never been in is a word processor, not Word or WordPerfect or framemaker or any other thing that makes the text hard to access (@manooooh)
@manooooh In a two column layout, getting images in the right place in Word can only be done once all the text is finalised. So one has to fiddle about no end. It's pretty painful.
@JosephWright doing that is a bad idea, because the image takes place, and you can lose the space you wanted to use to add more text or a table does not separate in two
@DavidCarlisle you are right, it is useless. You found mistakes and you preferred to use LaTeX, it is fine. A mathematician does not find those mistakes; he finds the errors in a demonstration
@PauloCereda actually for the word file itself (rather than any pdf or html produced) the accessibility is probably pretty good, not because it meets any standards but because the screen readers spent a large part of their development effort making that work as it's a rather large use case.
@JosephWright yes, it is a guide of how the document will look, but you should know that the image occupies space. Is not that what happens in LaTeX? Why does not it happen in LaTeX?
@DavidCarlisle yes. It is not new. You have to know how to take it, for that I use CTRL + LEFT_ARROW or CTRL + RIGHT_ARROW in addition to CTRL + N. Problem solved
@manooooh the correct answer is that people should care about those with disabilities. It's not always easy to care, with deadlines and other constraints but they should at least know that they should be caring.
@manooooh The entire way that figure works in LaTeX is based on the idea that it appears 'near' the text but is not fixed in place. That allows LaTeX to deal with e.g. an image which doesn't fit exactly where it is in the source. Basically, Word doesn't do that.
@JosephWright in Word, when I paste a big image into a page it is it is automatically reduced to space, and with manual adjustments (as there are in LaTeX), I can reach the desired size. I do not know what's wrong with that. Maybe I'm misunderstanding
@DavidCarlisle you want people to care about things that do not change the essence of the document at all; and this is wrong. For the format, leave it to an editor, who has his own rules
@manooooh Imagine you have a picture that is one column (8.5 cm) wide. You don't want to resize it. You've got some text in column one, the image at the top of column two, then more text. When you add a sentence to column one, everything moves in Word.
@manooooh So you have to move text a word at a time around the image so it's back at the top of the column. And that happens every time you edit the file
@PauloCereda do you think they care? I mean that if you believe that a document made by them was intended for people with disabilities. If the document supports it naturally, good; but otherwise, there is no problem, there will be people who modify the document
@manooooh @DavidCarlisle's point is that at the very least one should care that things are achievable. For example, if you deliberately paste text in to a file as an image, you are breaking any accessiblity. That's a choice.
@PauloCereda the same argument can be applied to a document that is made in English. Do you think the mathematician cares that it is translated into Esperanto or Spanish? He gives absoultely the same
@JosephWright why should not he move? In Word, when you put an image with the option "In line with the text" it does exactly what it should do; in LaTeX I suppose the same thing happens, that as you write the image moves, which is the natural
@JosephWright why do you say that in LaTeX that does not happen?
@manooooh No. Increasingly, working in any public body means that all documents have to be considered for accessiblity. Unlike translation, making documents accessible is part of the whole requirement to make reasonable adjustments.
@JosephWright no. I do not share David's vision. The mathematician writes in his language (or the one he likes). You can not translate the text to all possible languages; it would be just crazy. The same for people who have vision problems; you can not create a document for the person who is blind, for whom he is colorblind, for whom he has only one eye ...
@manooooh Having images move with the text is not the outcome one usually wants. The moment they hit a page/column boundary, loads of space is left. That's ugly. And it's wasteful of space. Most communication journals give a page limit: you have to use the space efficiently. So you have to reformat.
@JosephWright we are not talking about the public. In fact, here in Argentina I do not know any official document that is accessible to all; it is impossible to satisfy all the people of the country
@manooooh Well you have a very different situation to us. Most of western Europe, the US, etc. is getting on with this idea. It's not really acceptable to say to people who have sight issues 'tough, you have to request special documents'. And that's before there are ideas like data mining.
@manooooh It's a long-standing part of book design for good reason. Unless you take extreme steps (see books by Tufte), text will not just have 'spaces' for images that don't lead to bad blocks of whitespace.
@manooooh No. I'm arguing that one can't simply ignore the fact that accessiblity is increasingly important. Not just for big publishers, but for everyone. Having text reusable is part of that. As @DavidCarlisle says, there are solutions for e.g. Word. So it's not TeX-specific. It's about the 'ecosystem'.
@manooooh No, I never said I change that. I said one has to move text around.
@JosephWright "you" are taking time away from mathematicians who want to go deeper in mathematics and "you" are "forcing" them to modify all their documents so that more people are able to understand something they may not understand. I do not think that is a good idea
@JosephWright I am not sure if I understand you, sorry
@manooooh No. I'm only sayign that one cannot ignore these issues entirely. At present, they are not all easily solvable. But there is real legal pressure to make accessiblity the default position.
@JosephWright what I understood was the following. If you want to leave an image motionless so that it does not move every time the text is written, you are breaking the line spacing. It's like the justified text, which in LaTeX is by default. Do you find it difficult to read a document in Word but do not find it difficult to read a document in which some of its lines have more spaces than another (by the justified text)?
@manooooh If I have 'texta textb [image] textc', and I want to change to 'texta textd textb [image] textc', I'll need to move something to give 'texta textd [image] textb textc'
@JosephWright I agree that we should be aware of all people, but we can not pretend to write a type of document for each one. That's impossible at least for a non-editor
@JosephWright I take charge of my words, and when I make a mistake I accept it. I tell you that it is impossible to satisfy everyone, but we must not forget them
@manooooh -- More and more governments are requiring documents to include facilities by which a blind person can be provided with the content of documents in an understandable form. (Provided the blind person has the requisite knowledge to understand the content.) Especially important for textbooks. This is a sincere effort to "level the playing field".
