@manooooh I can only speak for physics (even though I have a lot of math books written by excellent authors which have upright d's). The main reason I am so vocal is that others have tried to discredit upright d's and I disagree with them. And I have a top physicist who got the highest distinction in mathematics who supports my views. Try to find a renowned mathematician who got a Nobel prize in physics and writes non-upright differential d's....
@marmot I love reading you. @egreg is a mathematician specializing in Algebra, and he writes "d" without italics. It is indeed a matter of taste, no matter how much recognition there is through
What is C here?If these are the complex numbers, I'd use \mathbbm (\usepackage{bbm}, unfortunately only pixels) or \mathbb. I personally use \mathrm{e} for abbreviating exponentials, to distinguish them from electrons and because all functions are typeset upright.
@egreg I know that AMS says that. It is totally inconsistent! Because the mathematical object refers to something that is not separated by addition or subtraction DOES NOT WANT TO SAY that the space must be erased. It is total madness
@egreg well, I did not test it... see the examples of the image please
If we have the space in \max\theta, why can not we have the same space for \max\langle\theta,p\rangle??
Silly question: does the AMS style guide work for people who want to write a math formula somewhere on the Internet? And for those who wish to write a math article?
@manooooh A guide is a guide. (Just for the records: I agree with most of what @barbarabeeton is saying). A guide provides suggestions. Think of it as a recommended speed. Very often it is very reasonable to respect it, but you won't get a ticket if you don't.
@manooooh I guess this is an attempt to weed out bad practices when writing articles. I agree and support the idea, and agree with far over 90% of what @barbarabeeton is proposing. (I referee a lot and can see how people can get frustrated.)
@marmot do not take it so seriously. At the end of it all, "normal" people do not pay attention, so for them there is no difference between a well-done document and a bad one
@manooooh Yes and no. I do know cases where the notation has lead to misinterpretations and thus to errors. You do not want to sit in a plane where a notational error in the design plan is at work. (Therefore I am very picky on making different objects distiguishable, such as a distance $d$ and a differential $\mathrm{d}$.)
@marmot what happens is that we live in different environments. You have cases where, because of typographical errors, disasters have occurred, and I do not have (nor will I ever have). The people we write for are different, and therefore "normal" people are a different definition for each one of us
@manooooh Again yes and no. I do think that we all benefit if the notation is clear. If you look at the physics site, you will find tons of questions that arose just because the author of some book was very sloppy. And @egreg is right when saying that several physicists are sloppy. And then there are those who know what they are doing, and those who copy without knowing what they are doing, in which case an unclear notation is even more fatal.
Do you think that it is beautiful, or on the contrary, should it be positioned differently when there is an underbrace command? How can a correct alignment be achieved?
@manooooh I am not sure I would add a bulky word like hypothesis to an equation like this. But if you really want to, you may do \documentclass{article} \usepackage{mathtools} \begin{document} \begin{align*} A&=2A-A\\ &\underset{\mathclap{\text{Hypothesis}}}{=}2(A-B)-(A-B)\\ &=\cdots \end{align*} \end{document}
@manooooh ^^^^ this is much clearer and can be produced with \documentclass{article} \usepackage{mathtools} \usepackage{tikzducks} \begin{document} \begin{align*} A&=2A-A\\ &\overset{\mathclap{\tikz[scale=0.2]{\duck}}}{=}2(A-B)-(A-B)\\ &=\cdots \end{align*} \end{document}
@manooooh No, there is no need. Everybody knows that ducks cannot be trusted, and therefore any equality marked by a duck is questionable, i.e. (at best) a hypothesis.
@manooooh You seem to be really biased against ducks. (I guess you need to play with the various ....lap commands that mathtools offers and/or ask on the main site.)
@samcarter -- here's the scoop. tug owns the drawings, so yes, the tug 2019 drawing can be used. but please check to see if it can be changed mid-year -- it will look "stale" after the meeting is over. here's the suggestion for the replacement at that time: tug.org/art/welcome-bw.pdf .
Now I make everyone's eyes here bleed.... \documentclass{article} \usepackage{mathtools} \begin{document} \begin{align*} A&=2A-A\\ &\underset{\mathclap{\underbrace{}_{\text{Hypothesis}}}}{=}2(A-B)-(A-B)\\ &=\cdots \end{align*} \end{document}
@marmot -- this is a fair question. ed witten is a physicist, not a mathematician. and he is the only physicist, to the best of my knowledge, who was ever awarded a fields medal. but the basis for the award was the immense strength of the mathematical ideas he used in developing his physics theories, ideas that had very powerful and lasting effects on several purely mathematical fields -- probably fields he wasn't intentionally addressing. (cont'd.)
(cont'd.) so, as a physicist, it isn't surprising that he uses an upright d to represent the differential.
@marmot -- delightful! you might look at it another way -- with the marmot standing still and the sphere rotating around the marmot.
@manooooh -- if a colon (or semicolon) appears in a theorem (in text, not to end a heading), at ams it would be set with \textup. mixed in with math, where all punctuation is upright, the slanted form looks goshawful and out of place. the same treatment is accorded to parentheses -- they are made upright.
@marmot -- try to find any mathematician who has ever won a nobel prize. the combination is essentially nonexistent. and it's simply the case that the conventions of math (pure or applied) are different from those of physics. i'm not trying to say that either is wrong; they're just different.
@manooooh -- no difference between a well-done document and a bad one? if you're writing a textbook, it's better to conform to accepted practices so that students will learn more easily. there are people who say it doesn't matter, but it does. randomly formatted math expressions can easily be misunderstood, or be difficult to interpret. that's why there are copyeditors. (@marmot -- you have my sincere sympathies if you have to referee such stuff.)
I have just seen an edit suggest in which an user (rep 1) edit much another user's post (also rep 1) (change the code, log files, etc.). They shares the same identicon. I wonder if they are one person.
@marmot -- but why should you give up that idea? just because it's possible to use an existing drawing of a lion that is not a tikzlion? there should be plenty of room for everybody, i think.
@manooooh -- traditionally, with cases the left-hand member on a line is flush left; relations (= signs and the like) are not aligned. but i've seen some authors do that when there is no right-hand member, and it hasn't been edited away. (however, i am not a mathematician, and would have to ask someone who is for a ruling.)
@manooooh -- there should almost always be a comma after the left-hand member of a line in a cases statement, and often also after the right-hand member. the expression is punctuated as if it is a sentence. (hence a period at the end, if it isn't followed by continuation text.) in one of the examples, there appear \cdots after the = sign, and that is the correct choice there.
@barbarabeeton so do you mean that x^2+y^2=2,\\x=y-2. is correct?
@barbarabeeton if there is a piecewise function, AMS suggest either of the two options. One of them is not to put punctuation marks after the expression on the left (e.g. f(x)=\begin{cases}1&\mathrm{if}\;x\geq0,\\0&\mathrm{otherwise},\end{cases})
@JouleV -- some of the identicons for different persons are exceedingly similar. there are a few that are so similar to mine that i have to look very carefully to spot the difference. so my conclusion is that without seeing them for myself, i wouldn't be willing to make a judgment.
@barbarabeeton of course there are differences between a well-done and a badly done document, but for most people it does not notice the differences. Regarding the use guide, I agree that students should use one, but unfortunately it does not happen
@manooooh -- in the raw edited documents i've looked at recently, a comma is almost always added after the left-hand element. and by the way, the recommended coding for the right-hand member would be `\text{if $x\geq 0$,}\` -- the style of the word should follow the style of the text in the surrounding paragraph.
@manooooh -- a semicolon should, in an environment where math is mixed in, be upright. since it's so difficult to distinguish an italic period or comma from a roman one, nobody bothers. but ideally, a "theorem font" should have italic letters and roman punctuation. it's just too complicated to build and use one properly.
@manooooh -- this one's a tossup. the result should look the same (although with some fonts there are subtle differences, but not usually large enough to be noticed). what's important is, as said earlier, consistency.
@barbarabeeton ok, thank you! What about egreg's comment? I have never seen something like "Theorem 2;" but "Theorem 2:" or "Theorem 2.". I like the last one. This dot should be bold, like "Theorem 2"?
@manooooh -- "Theorem 2;" wouldn't make sense unless it was at the end of a clause, certainly not as a heading. the ams document classes by definition make the period in "Theorem 2." bold if the rest of the heading is bold.
@manooooh -- by "heading", i mean the "Theorem 2." as in your subsequent example. and "Proof." is likewise a heading. the particular font style used for such headings depends entirely on the publisher's style specs; bold is often used, but caps and small caps or italic is often used instead. a publisher's specs ("specifications") are what make the look of a journal or book series consistent and distinctive.
@manooooh -- this is indeed the specified style for ams journals. the specs are different for some book series.
I am submitting a manuscript through Editorial Manager. I am getting this error. The manuscript successfully compiles on my local computer. As I am new to Latex, I don't know how to fix this. This is compiled on the server side, so I cannot modify the installation folder. The problem is that I do...
@barbarabeeton I agree. Yet I personally would not use a brace at all, nor a bulky word. I almost completely stopped using underbraces. In papers the are not very common anyway, and in notes there is, according to my experience, always a more appealing way to go without them.
@marmot -- well at least it's not so much of a head-scratcher as if there were disagreement on this particular issue. (or at least disagreement without any reason being given.)
@manooooh -- "upright lightface roman" is just the normal text font.
okay folks, time to tuck in. i have to be at the office on time tomorrow. nighty night.
@barbarabeeton I guess we all agree in principle. Yes, there may be small disagreements on details, but this is what chats and discussions are for. It would IMHO be sad if one could not have such discussions.
@manooooh if you are happy with that, that's of course great. (My personal view has shifted, though. What is written on the blackboard and what is written in notes is IMHO rather different. Of course, on a blackboard I would write something along the lines you are proposing, but not in notes, let alone in a research paper.)
@manooooh I would just present the equation without the brace and the word "Hypothesis", and then continue after the equation: "As the alert reader certainly noted, in the step from the first to the second line there is an error."
It always depends for whom you write the stuff. And one does not have to introduce mistakes. Sometimes equalities mean really equal. It really depends on the context. (I do use things like \[a\overset{!}{=}b\quad\text{or}\quad a\overset{?}{=}b\]
@barbarabeeton Thanks a lot for checking the image rights! I asked a question on main meta if it will be possible to remove/change the add after the conference.
This question is similar to previous ones considering chapterthumbs -- answering this question will render them obsolete.
How to get chapterthumb work in KOMA-Script?
How to get chapterthumbs match their chapter titles in KOMA-Script?
Bleed correction for chapterthumbs in KOMA
How to fix ...
@Max actually it is an achievable constraint, if you set \lineskiplimit=-2cm (or anything large and negative tex will honour the constraint that you set and set the large text to a small baseline even if it causes over-printing.
`\begin{algorithm}[H] \caption{REINFORCE Algorithm} Initialise $\theta$ arbitrarily\\ \For{each episode $\{s_1, a_1, r_2, ..., s_{T-1}, a_{T-1}, r_T$\}\sim \pi_\theta}{ \For{$t=1$ to $T-1$}{$\theta \leftarrow \theta + \alpha \nabla_\theta \log \pi_\theta (s_t, a_t )v_t}$} \Return $\theta$ \label{alg:reinforce} \end{algorithm} ` How do I get the two ends of the for loops to appear in the right places?
@mtbrands sorry, I have no access to documentation right now, but I am almost sure closing macros are needed, but it also depends on the package you are using to typeset algorithms...
@mtbrands you need to specify the explicitly. It might be an idea if you post a full question on the site, remember to make a full minimal example, don't just post a sniplet like this. Also isn't algorithm a float env, thus you need an algorithm env as well.
@CarLaTeX -- obviously that journal either has no copyeditors, or has copyeditors that don't care. (it's too easy to blame the author. pity an author with great knowledge but also with dyslexia.)
@HaraldHanche-Olsen -- ah! that's genuine, but (like its opposite) vanishingly small. good work.
@egreg -- the main problem with that is that pdf is no longer linear; as a former colleague used to say, "it's hamburger".
@samcarter -- thanks a lot for posing the question. i'll try to keep watch for an answer, and hope it's positive.
@barbarabeeton I got an answer on main meta that one can edit/remove the ads at any time. I put a reminder in my calender in case I'll forget about the ad :)
@samcarter -- good news! thanks! two other, more generic drawings have been suggested as replacements when the time comes. i'll provide the details as soon as i'm sure i've identified the correct locations to find them.
@Skillmon Why are you asking me? (I personally would say these are equal time.) If the reason why you are asking me is the academic honesty discussion, then I do not see any academic dishonesty in being 4 secs lates.
@barbarabeeton I'm curious what the drawings will be! Putting the details in a comment below the ad will probably the best way to find this information again after TUG19 is over?
@Skillmon There are several users who are at least as cautious as I. I think there are two questions. 1. Did one steal from another? No. Does it make sense to keep two almost identical answers? Please ask someone else. ;-)
@samcarter -- one is a group of differently-dressed lions (including rather preppy ones like those in the latex manual) holding up or sitting on the letters "TUG"; the other is, i think, the "tex friendly zone" drawing. unfortunately, the way my mailer works, the url is obfuscated, and i have to ask for help to get accurate information that i can give to someone else. (i'll let you know as soon as i find out.)
@JosephWright @StefanKottwitz Hi, and excuse me for this comment. I have always same problem. I have flagged now. Thank you very much. My best regards.
Welcome to the new entry into chat. It is correct my English?
@Sebastiano Hi! If images or code are copied from somewhere else, a link to the source should be added. That's a correct comment or flag. That this user targets specifically you, is an annoying thing, I understand that. Just fix the issues (missing links to the source) and flag us again if you see something really wrong.
@Sebastiano I think the user in question is very keen on 'full attribution; as @StefanKottwitz says, images or code from elsewhere do need to be linked, although the user could just suggest an edit
@Sebastiano Please flag offensive or abusive stuff. Just don't flag comments that are theoretically correct just because they pile up. Is it ok if I remove the flags? Be sure, we will react on abusive behavior towards you and towards anyone else.
@Sebastiano You noticed that we do act once we can justify it. Last time I had to suspend an abusive user, a suspension lasting for 1 year (!) because a first serious warning did not help. So we can do something, once it's clear.
@StefanKottwitz Good afternoon everybody. Just a moment because I must use Deepl translator.
@StefanKottwitz @JosephWright @MartinScharrer One second I'm answering to all of you. Excuse me for being so slow first of all.
@StefanKottwitz These are questions from over a year ago that I haven't checked since. Surely and with all the sincerity of the world I copied an image from the net because I didn't know where to look for a good quality one. I can change it there are no problems. But if I flagged it's just because it's annoying to be contacted after the user has been reprimanded. I add that after almost two years as you can see I add where I got the source. This indication dates back to early 2018.
Translated with www.DeepL.com/Translator
@JosephWright These are old answers and I should check them out. I didn't have and haven't installed the programs I refer to. That's why I took pictures from the network.
@StefanKottwitz Remove the flags correctly. Thank you and thank you all. Even if I have been a member for more than two years, I am learning slowly and I hope that many things have changed since the date of my registration.
@CarLaTeX Honestly, when I returned to school after the Christmas holidays on January 7, my face was paralyzed by the cold. We didn't get used to it here. Yesterday it rained but the cold is coming back. I now have my hands frozen. And then I had the polls on January 22nd. A disaster. I work in a mountain village. Day 22 -1 degrees. Maybe now we are at 8 or 10 degrees.
@StefanKottwitz The suspension of the last user was 10 years and I saw. I blindly trust you all. Thank you very much to all modeators.
@Sebastiano One could, besides ignoring annoying repeated stuff, theoretically: add the missing link, flag the comment as obsolete, the mods would remove the comment. Don't search your old post for missing links if that stalker does the work for you. Still theoretically, one could lean back and wait until something abusive happens that justifies suspending from the network. Either nothing serious happens, or it does.
@StefanKottwitz I hope I understand. I remove the image that is the simplest thing at least for me and the fastest. A big hug to all you moderators and to all the users who love me. Hello and good work.
Just a quick question: In TikZ, is it possible to do something like asin(veclen(\x1,\y1)/\y1)? It seems that the division of two dimensions is not dimension-less but given in points…
@TeXnician When TikZ parses asin(veclen(\x1,\y1)/\y1) for, say, \x1=3pt,\y1=4pt, it will first get asin(5pt/4pt), then (and this is, well, "surprising") asin(1.25pt) and then it will compute asin of the dimensionless quantity 1.25. So in this case the overall result is fine but not the intermediate one.
@TeXnician If you do `\draw let \p1=(u1), \n1={asin(veclen(\x1,\y1)/(\y1*10))} in (0,0) -- (\n1:4cm) node[right] {\n1};` instead (now the argument is smaller than 1) you do not get an error.
@TeXnician If you do `\draw let \p1=(u1), \n1={atan2(\y1,\x1)} in (0,0) -- (\n1:4cm) node[right] {\pgfmathsetmacro{\pffft}{\n1}\pgfmathprintnumber{\pffft}};` you get an angle (no pt attached).
@PauloCereda Yes. But some terminologies are misleading. For instance, it is not true that those working at an accelerator are particularly fast, even if it is a particle accelerator.
@PhelypeOleinik, English is not my native language. Can I use typeseted instead of was typeset? Is it regular or irregular? I have no idea. I would like a short text, without document was...
@Sigur I'm pretty sure it's typeset. typeseted sounds weird. And as I said, that's an idea, you'd have to adapt. Are you writing in Portuguese or English?
@Sigur That's Google Translate's suggestion for "typeset". I like neither. I think in Portuguese we don't need a word for "typeset" because we use ABNT.
@manooooh -- well, the person who will do much of what i have been doing has been at ams for a year, and i've been trying to teach him all i know that should continue to be important to his job. (procedures are changing, so some things will become obsolete.) whether he will become active on tex.sx is up to him.
@manooooh -- yes, i will continue here, although i have to get a computer at home that behaves better than the laptop i'm using now (which doesn't have tex installed, so i can't answer questions that require testing, which is most of them). thank you for the kind words.
@manooooh -- oh, i hope to get more than a laptop. the screen i use at the office can accommodate two full-size letter-size pages side by side at 100%, plus a little extra real estate, and i will hate losing that. also, a laptop keyboard is too greatly compressed -- no difference between delete and backspace, for instance, and no independent arrow or "home" keys. so getting a new system set up will be a challenge. but i'll survive and hope for the best.
@manooooh -- that looks a lot like the macbook i have for a portable. but after apple managed to make me lose access to two gmail accounts, i got disillusioned, and haven't kept it updated. (the local apple store has no idea of what tex is either, and their "assistance" made installation impossible.) so we start over from scratch.
@barbarabeeton two questions. 1) What do we want to define should always be in italics ?, 2) If I want to use a name for the definition but without listing, 2.1) should it go without italics?, 2.2) should it also be mentioned in the consequent text? For example, is the following correct: "Definition (Math object). A math object is..."
@manooooh -- the style guide doesn't say how this is done. however, it is based on the \theoremstyle{definition} from amsthm. read the user guide -- texdoc amsthdoc for guidance.
@barbarabeeton it is the first time I read that Apple behaved badly with someone, but it is something that often happens. The brand of my laptop is much lower than Apple's, but there are specialized laptops for other types of needs
@manooooh -- ams style has deinitions set with a theorem-style header and upright text. but remember, this is the style of a specific publisher. other publishers may recommend a different style. the ams position is that definitions should have a clear beginning (the theorem-style heading) and a clear end (extra space to separate it from what follows), and be distinct from theorems/corollaries/lemmas/etc.
@manooooh -- yes. that is what i'd recommend (if you're adhering to ams style). the * will cause the number to be omitted, as you have already found out.
@barbarabeeton nobody tells me how I have to format the text, so I am guided by what the AMS guide says (which I find very good). Do you think that this example: "Definition (Math object). A math object is..." is correct based on what AMS say?
@yo' -- good information. what operating system are you using? (totally unrelated topic. in my sorting through papers, i just found something i had lost track of and have really been looking for. it's a copy of a business circular from my grandfather's bakery entitled "Christmas Stollen", and ending up with the note "This product is guaranteed to contain the best and richest ingredients obtainable, also New England Rum and Pure Brandy [Prohibition Permit Md. H 3118]" a real treasure!)
@yo' -- great! thanks! my new system will be linux, and i expect to be asking for help from norbert. (i really hope that gordon will have enough patience to be my resident sysadmin, but he's still got to get up to speed on linux; he's comfortable with windows, but the system he knows best is still the ibm 360 or 370. waaaaay different! somehow we'll manage.)
@manooooh -- what you show looks like what ams prefers. i'm glad you think the `amsthdoc' guide is clear. (i admit to writing much of it. part of my soon-to-be-former job.)
@manooooh that would be quite the AMS way. I personally prefer math object to be slanted rather than italic, but I have the same preference for theorem statements actually, and many other people dislike this.
@yo' -- actually, i personally like the slanted type in contrast to italic (it makes it really easy to spot an isolated variable "a" in the statement of a theorem), but it's necessary to toe the company line at times. in either case, though, the distinction between upright and non-upright type has a very long history in math publishing, and since "math has a long shelf life", recognizing that distinction and adhering to it is a valuable component of communicating.
This question is similar to previous ones considering chapterthumbs -- answering this question will render them obsolete.
How to get chapterthumb work in KOMA-Script?
How to get chapterthumbs match their chapter titles in KOMA-Script?
Bleed correction for chapterthumbs in KOMA
How to fix ...
