« first day (2006 days earlier)      last day (2934 days later) » 

Anonymous
12:41 AM
@cfr What about the images? tex.stackexchange.com/questions/306754/…
 
Anonymous
+ I'll add another quote who claims it was made in LaTeX.
 
Paulo Cereda
Paulo Cereda
@VincentVerheyen What's the purpose of this question? I really don't follow (besides, I don't get the point of including the paper's text...)
 
Anonymous
@PauloCereda The point is to try to discover which (kind of) person(s) created the document. If that person knows how to work with LaTeX (or a certain package of), that can be a lead. The paper's text can be included for people who know to analyze a document in a different way.
 
Paulo Cereda
Paulo Cereda
@VincentVerheyen I still cannot follow. If the bloke's identity is a mystery, IMHO, 1. there's no way of telling the PDF was originated from him, 2. there's no way of telling metadata was modified, 3. you can tweak a lot of LaTeX helper macros and produce a Word-like document, 4. the statements are too strong from a mere document and no info from the person behind it, so nothing can be inferred.
But that's me, a duck. :) Just my two Bitcoin cents. /ba dum tss
There are some English folks in the chatroom. There might be a Sherlock-esque vibe at some moment. :)
 
Anonymous
12:57 AM
@PauloCereda Somebody should be able to figure out how he made those images (analysis of pixels, transparency et cetera); and/or re-produce the formulas exactly in the same way.
 
Anonymous
@PauloCereda And people who tweak metadata to fake origin of documents can make mistakes, and they might be spotted. That doesn't mean we can be sure that the document was completely made in LaTeX, but it can demonstrate that somebody tweaked with the meta-date in attempt to fool or laugh with the public.
 
Paulo Cereda
Paulo Cereda
@Vincent: well, in that case, IMHO the question is off-topic for TeX.sx. I'd say it would be a better fit for a typography forensics group.
 
Anonymous
@PauloCereda But the question, to stay on topic, is specifically whether or not some parts of the document seem likely to have been made in Tex. If it was VERY DIFFICULT (but still very likely that it needed to be created in Tex) to create the document as such in Tex, then it might require certain packages, which might reveal a person's compiling fingerprint, or hint at a certain group of persons who specialize in using advanced packages.
 
Anonymous
1:13 AM
@PauloCereda A funny (and perhaps enlightening) answer would be when somebody re-creates a .tex code needed to exactly compile the paper.
 
Paulo Cereda
Paulo Cereda
1:26 AM
@cfr: Cwac!
 
cfr
cfr
@VincentVerheyen What about them? I agree with @Paulo (cwac! & I am going to reply to your email ... - thanks): I can't see the point of this. It seems an exercise in futility. Of course it can be reproduced in TeX if somebody really wants to. Of course it can be created in another programme if somebody really wants to. I think it is unlikely to have been made with LaTeX unless the author was from the beginning attempting to fool people and thought the effort worth it.
@VincentVerheyen I 'tweak meta-data' almost every time I compile a document. This is routine and requires no special expertise at all. I also routinely wipe meta-data in post-processing when I can't be bothered or don't remember to preserve it. This stuff is not intended to certify anything or be of any use for forensic analysis. It is designed to be easy to modify so that people can use it effectively for communication.
 
cfr
cfr
1:42 AM
@VincentVerheyen ^^ Produced by pdfLaTeX.
@PauloCereda Cwac!!
 
Anonymous
2:03 AM
@cfr Dear cfr, that's very interesting. Perhaps you can post this image as a comment. But, are you sure there wouldn't be a difference in encoding, when you would produce it this way? Perhaps some other things might be copied along, which would then need to be deleted/altered manually by the fooler? It might just be interesting to note all such things and little details.
 
Anonymous
@cfr I think it can be good to focus on the images as well. In what application do you think they are made? Perhaps there is an easy package in LaTeX to create (most) of them?
 
Anonymous
@cfr Under Creation of the paper (LaTeX or not?), I have added 2 sources which claim that the doc was prepared in LaTeX. It might be nice to discuss their arguments in an answer.
 
Anonymous
If you agree with the arguments in the second source

>> "The layout (I.e. margins, font, etc) are the default stylesheet for LaTeX, [...]" <<,

that means that, if you think the document was NOT prepared in LaTex, that the author must at least have known LaTeX, or that he (when he would not know about it) otherwise must at least have tried to mimic papers made with LaTeX?
 
cfr
cfr
@VincentVerheyen I agree with @PauloCereda The question is off-topic. You can debate the what-ifs and what-if-nots, but it is all pure speculation. I have no idea what made the images. Probably not LaTeX but I don't know. The 'encoding' stuff you're talking about is just the language of PDF as far as I can see. Any PDF has to look roughly similar.
 
Anonymous
@cfr The question is on topic, as it specifically asks if the document (or parts, such as the images) might be created with LaTex, and if so; if traces should remain visible of such workflow.
 
Anonymous
2:15 AM
@cfr Why are the images probably not made in LaTeX?
 
cfr
cfr
@VincentVerheyen Or at least seen a LaTeX document ... or a document based on a LaTeX document ... or something ... And this will help identify the person?! All you can do is speculate.
 
Anonymous
@cfr The questions might or might not be actually identify the person(s). I agree with you that such identification is off-topic, but nevertheless it might be of interest. I am not asking for the exact identification in the question though. --- There might be this nifty detail which could amount to extra information about the creator. It can be a puzzle piece, not a complete solution of the puzzle.
 
Anonymous
@cfr Fore example, I discovered a detail (and LaTeX experts might know if this is inherent to LaTeX or not ... just a guess). If you look at the final image (in the original PDF), and you go to the boxes around the words "Trusted Third Party" & "Counterparty" respectively; then the top-left pixel of each of those boxes is coloured differently. For the left box, it is RGB (246), and for the right box, it is RGB (235).
 
Anonymous
@cfr There might also be a maximum-zoomability in some PDF readers, depending on the way in which a document was created? I am just guessing wildly here.
 
yo'
yo'
5:59 AM
@VincentVerheyen On second thought, I agree with @cfr, this question is not really suitable for this site. We are not a forensic analysis place. I asked the folks at Skeptics whether such a question is suitable for them, I'll keep you posted when they reply.
 
 
1 hour later…
Anonymous
7:08 AM
@yo' There is also a bitcoins (beta) StackExchange, but unfortunately, those people might miss the expertise of LaTeX needed to answer the questions.
 
Anonymous
Mico for example, writes an interesting comment, which however doesn't elaborate:

**I agree with @cfr that the document doesn't look like it was produced by LaTeX. A tip-off is the complete absence of discretionary hyphens and, coupled with that, the readily apparent uneven amounts of interword spacing. The presence of Arial also hints at a non-LaTeX origin.**

If he would elaborate on why it is very difficult or so to have uneven amounts of interword spacing in LaTeX, then he could write up a real answer? I think people on Tex's SE have enough knowledge to write up a decent answer, but it…
 
Anonymous
@yo' The question is about whether there is anything possible in LaTeX, which might leave such a fingerprint which is impossible to create in OpenOffice (or in any other software perhaps). The question is thus about the specificity of LaTex, which may be best answered on this forum.
 
Anonymous
I am also surprised that there seems to be no-one who knows how to create those images in LaTex?
 
yo'
yo'
@VincentVerheyen You can do anything in LaTeX and then post-process by some PDF processing tool to remove the traces. But the point is: why in the world would you do that?!
 
egreg
egreg
@VincentVerheyen If you want my opinion, I can't see the slightest hint that LaTeX has any role. What the people at Aston think is just speculation. The argument about double spacing at end of sentences is simply stupid. As you know, two spaces in a LaTeX typescript are exactly the same as one; non French spacing is obtained in a really different way than those “expert” seem to be claiming.
 
Torbjørn T.
Torbjørn T.
7:18 AM
@VincentVerheyen There are plenty who could, look through questions tagged with tikz-pgf, pstricks or metapost for example.
 
egreg
egreg
@VincentVerheyen By the way: hinting that something has been written with TeX and saying that without the TeX source it's impossible to say whether it really is must make one think: are those people expert in something?
@VincentVerheyen It's like “the pyramids were built by aliens and they would surely confirm it”.
 
phell
phell
Hello (La)TeX lovers!
 
egreg
egreg
@phell Good morning!
 
phell
phell
IMHO Mico is making a good point in the comment: "the readily apparent uneven amounts of interword spacing". To me, as a novice user, the inter word spacing is one of the signatures of (La)TeX.
 
Joseph Wright
Joseph Wright
7:51 AM
@VincentVerheyen Remember that in the end a typesetting system simply puts glyphs onto the page: you can make any decent system put glyphs anywhere. As already noted, the question has to be why would one bother: the easiest path to creating the document you've linked to would be to do it in exactly the way it claims it was made, using OpenOffice.
 
David Carlisle
David Carlisle
8:17 AM
@VincentVerheyen the question is clearly off topic (I voted to close) but also I can't see why you think tex is involved at all, the document says it is produced by openoffice and is using fonts that are rather uncommonly used with tex and much more easily used by open office so unless there is something else you haven't mentioned, assuming it is produced by open office seems a reasonable approach.
 
Ulrike Fischer
Ulrike Fischer
@DavidCarlisle @PauloCereda Paulo would have used this reasoning en.wikipedia.org/wiki/Duck_test ;-)
2
 
David Carlisle
David Carlisle
@UlrikeFischer that page is of course @PauloCereda's thesis. He must have finished it after all and never told us:-)
 
Anonymous
8:35 AM
If there is no reason to believe that LaTex was involved, than that could be argued as an answer (if elaborated upon). LaTeX is put forward as the compiler by some sources cited. Yo' said:

You can do anything in LaTeX and then post-process by some PDF processing tool to remove the traces.

So, if there are traces in regular Tex documents, one could write up an answer as to what those traces are, demonstrate whether such traces are present in the PDF cited, and/or demonstrate that they can be easily removed, if they can.
 
Anonymous
I think all of you guys have a lot more knowledge to elaborately say why there is no possibility of figuring out why (this) and/or any document / picture / formula is made in LaTex.

An another interesting thing is to say, that it can not be made in LaTeX, unless one is aware how to use advanced package A, B, C or D, ... Perhaps those packages might or might not only be used by a handful of people.
 
Anonymous
@TorbjørnT. Thanks, added those tags.
 
Ulrike Fischer
Ulrike Fischer
@VincentVerheyen You can find for every claim - how absurd it may be - a source. I won't certainly not spent my time to refute every nonsense people did put somewhere on the net.
 
David Carlisle
David Carlisle
@VincentVerheyen sorry it would be off topic even if there was any suggestion of tex involvement but as there is no indication at all of tex involvement and every indication that it was not made by tex then unless you think that discussing any pdf file found on the internet is on topic for this site there is no reason to re-open the question.
 
Anonymous
@UlrikeFischer It's a good thing to refute nonsense! People can learn from it. For example, somebody who knows a lot about images might know whether an image from the document COULD (easily or not easily), or COULD definitely not be made in LaTeX.
 
Anonymous
8:44 AM
@DavidCarlisle How is there every indication that it was not made by Tex? It could be interesting to write up those indications and elaborately say that it could not be made by Tex, if somebody wishes to claim so.
 
yo'
yo'
in Hub of Reason, 1 hour ago, by Oddthinking
(To be clear: If you send it over here, it will be ruthlessly cut down to the core claim - I hope that satisfies the author.)
@VincentVerheyen (so it can be placed in Skeptics, but it has to be slightly modified.)
 
Anonymous
@yo' How about it can just stay put on hold on the Tex forum? I fear people at Skeptics don't have the technical knowledge to discuss the question, which is Tex-related.
 
yo'
yo'
@VincentVerheyen well, it can't receive an answer now, and it'll be closed in 48 hours if it's not reopened or migrated. It shouldn't get deleted since it has got an upvote, but still..
 
David Carlisle
David Carlisle
@VincentVerheyen no you picked up a pdf from the internet why on earth do you think tex is involved? the document says it is not made by tex it is using fonts essentially never used by tex and it doesn't look like tex output.
 
Anonymous
Look. I can also post another question:

**Can you know, from an image created in LaTeX, if it was created in LaTeX, with high certainty (e.g. if it would require hours of Photoshopping otherwise)?**

The current question is merely an application of that question (although it also asks about the whole document).

So the question about the whole document is just an application of:

**If a document is created in LaTeX, is there any way to tell whether or not it was created in LaTeX, or can all traces be removed easily (and if difficult, what difficulties need to be faced?**
 
Anonymous
8:51 AM
@DavidCarlisle I just quote some computer scientists on fora, as well as a whole line of forensics students and their mentor. Further more @DavidCarlisle, I think you have enough knowledge to write an answer, considering the things you have just said. Why does it not look like Tex output? That means you would know what Tex output looks (has to look) like.
 
David Carlisle
David Carlisle
@VincentVerheyen I wouldn't write an answer as it's off topic.
 
Anonymous
@DavidCarlisle How about my 2 questions presented above? Do you think they are off-topic? My OP is just a specific version of those 2 questions.
 
Ulrike Fischer
Ulrike Fischer
@VincentVerheyen Then -- if needed -- I will vote to close your question as nonsense. Beside this you are a very good example why I normally not bother with this type of discussion: You are obviously completly deaf to arguments. You are trying to make your question unfalsifiable and from there on discussion no longer work.
 
yo'
yo'
@VincentVerheyen The problem is that nobody can put a definitive answer if you want a definitive answer. The answer is: "with a lot of effort, you could have done it in LaTeX probably."
 
Anonymous
@UlrikeFischer Look yo' just created an answer. A very interesting answer too. It could be elaborated upon. It is also core to people with Tex knowledge, thus on topic. And Ulrike, how can my question be unfalsifiable? People might claim There is no way to know, for this and this reason. That is an answer, right? It can even be overthrown by evidence to show that there might be substantial reasons to demonstrate that it is (very likely) / has to to be made in Tex or not.
 
Anonymous
8:56 AM
Just an example of a possiblity: Perhaps Tex somehow puts a trace into each document, about which only 2 persons know. Then those 2 persons could explain such traces, and perhaps show that they are present in the document. That means that, if those 2 persons really were the only ones knowing about the production of such traces, that the paper was either typeset in LaTeX, or that the author of the paper must be (1 of) those 2 persons.
 
phell
phell
@VincentVerheyen You can't tell how many people know about such details. I would focus on trying to estimate the probability it was made with LaTeX/OpenOffice/Word/...
 
Ulrike Fischer
Ulrike Fischer
@VincentVerheyen Why is a proof of any sort if someone answers? Sorry but I'm certain that I ask in some suitable place if Abraham had a Tattoo on his left shoulder someone would answer. This wouldn't make the question interesting. It also wouldn't make it on-topic here -- even if I added the question if the tattoo could be reproduced by a latex and if not why not.
 
Torbjørn T.
Torbjørn T.
@VincentVerheyen Yeah, not really appropriate. Your question isn't about any of those packages specifically. The more generic diagrams is better, so I edited.
 
yo'
yo'
9:18 AM
@VincentVerheyen but I wouldn't post it as an answer. This is not the type of questions I'm seeing value in. Yes, I have answered questions before that are considered opinion-based by some people, but it was in places where I could share some expertise. Here, anything is either pure speculation or meaningless rubbish.
 
Paulo Cereda
Paulo Cereda
9:39 AM
@UlrikeFischer ooh the best philosophical approach of all time. :)
@DavidCarlisle :)
 
yo'
yo'
@PauloCereda in one Czech movie, one person says: "He looks like Crow but he's Rabbit." (meaning Mr. Crow and Mr. Rabbit)
 
Paulo Cereda
Paulo Cereda
@yo' oooh :)
@phell Quack!
 
yo'
yo'
(actually, the plot is funny, the guy, Mr. Crow, is earning dirty money and his family doesn't know. But his neighbor spots him in an expensive car. So Mr. Crow says: "I'm Mr. Rabbit, what do you want?" But the neighbor drags him home to Mrs. Crow who says: "Oh goodness you two must have been drinking or what?!")
(to this the neighbor replies the above sentence)
 
Paulo Cereda
Paulo Cereda
9:56 AM
@yo' :D
 
Anonymous
10:45 AM
I think a nice answer could be (e.g.): it can not be known [Note: the sentence doesn't end here] based on these and these factors; or it can be known based on these and these factors. This can be different, as an answer, for parts of the PDF. Further more, it seems like the comments are getting a lot of votes (most of which deal specifically with Tex) ...
 
Anonymous
@yo' Do you think the 2 questions provided in chat.stackexchange.com/transcript/message/29309203#29309203 can only generate opinion-based answers? Do you think nobody can learn from the answers of those questions? I repeat that I think my OP is merely a specific application of these more general questions.
 
Anonymous
@UlrikeFischer I think it's very interesting to know if Abraham had a tattoo on his left shoulder. I am curious already! ;) One can argue some evidence (that is: the state of current evidence) as to why it might be very difficult to find out. And somebody with a lot of expertise might give some reasons as to why this or that hypothesis might be more or less plausible. The answer to the question about whether the tattoo could be reproduced by LaTeX depends on what you mean with "reproduced by".
 
yo'
yo'
@VincentVerheyen if you are as paranoid as you are, yes.
 
phell
phell
@PauloCereda Quack quack! Hee hee hee.
Would a question about debug experience that was extra tricky fall into meta forum?
 
Anonymous
11:05 AM
@yo' I'll take that as a compliment, as I think to doubt is important for intellectual development. So, disregarding my situation, which might or might not be characterized by paranoia; I assume that you agree with my curiosity that those are interesting questions, considering that you escape the answer by putting forward a joke?
 
David Carlisle
David Carlisle
@phell ? meta is just for asking about the site itself, if you want to ask about debugging tex documents that would be for the main site.
 
yo'
yo'
@VincentVerheyen well, you should know that I'm cynical and I put forwards jokes on anything and everything. I think your question is ridiculous, see the Duck Test above.
 
Paulo Cereda
Paulo Cereda
MY DUCK SENSE IS TINGLING
 
phell
phell
@DavidCarlisle I started thinking of formulating a question after tracking a problem to the new LaTeX kernel not liking an old class file re-definition of date and time. I am curious to hear others experiences with "tricky" debugging but I am not sure if it is within scope for the main site and if it is too much of a "list" question.
 
David Carlisle
David Carlisle
@phell possibly. aside from whether it's on topic though what did we change in the format about date and time? (I can't think of anything since 2e came out in 93)
 
phell
phell
11:13 AM
I have a feeling those answers could teach others how to think in new ways while debugging.
 
yo'
yo'
@PauloCereda 7th sense
 
Paulo Cereda
Paulo Cereda
@yo' :D
 
David Carlisle
David Carlisle
@phell possibly although the site mechanisms try to avoid tutorials and force specific question and answer format. so probably it's the wrong place:-)
 
phell
phell
@DavidCarlisle Well, it is possible I didn't find the root cause to the incompatibility.
Maybe I should pose a question about that specific instance of debugging insetad.
 
David Carlisle
David Carlisle
@phell as I say I'm interested (as if a recent format broke something it was probably me that did it) so you could tell me here even if you don't post:-)
 
phell
phell
11:20 AM
@DavidCarlisle I will look through my ScreenShots folder. The problem surfaced when one of my computers had updated a file related to l3kernel.
My temporary solution was to stop using the SIunitx package.
 
Anonymous
@yo' That's fair enough, I still assume that you think the questions are interesting, although you also find them ridiculous (that's still possible, right?) Further more, you refer to a source about abductive reasoning, which I think is not too much different from induction; and I think to understand that, in simple terms, it incorporates quite a large room for uncertainties. The Duck Test seems like it's a humorous aspect of our ability to criticize uncertainties...
 
David Carlisle
David Carlisle
@phell ah the l3 kernel, I thought you meant the latex format kernel. That won't be my fault; we can blame @JosephWright for that:-)
@VincentVerheyen The only part of this that I find vaguely interesting is why you are asking about it here in a tex forum rather than a forum for indesign or troff or something? You have a pdf that claims to be and so almost certainly is, an openoffice document, so why (without giving any hint of a reason that leads you to think that it wasn't made by openoffice) ask here rather than than on a site for any other software?
 
Anonymous
@DavidCarlisle The first things that comes to my mind (a hint of a reason that leads you to think that it wasn't made by openoffice) is that those diagrams can not be made in OpenOffice (I have to specify: as far as I know that software).
 
David Carlisle
David Carlisle
@VincentVerheyen really? (I wouldn't know, but just a few boxes and text, it can't do that?) but anyway there are thousands of drawing programs, just picking tex at random and asking here doesn't make it on topic. But anyway unless you edit the question enough to persuade 5 people that the question is on topic and should be re-opened it doesn't matter anyway as answers can't be posted.
 
Torbjørn T.
Torbjørn T.
@VincentVerheyen I'm fairly certain those can be made in OpenOffice/LibreOffice. Look at the Draw program.
 
Anonymous
11:35 AM
@TorbjørnT. Interesting, I will take a look.
 
Anonymous
@DavidCarlisle To follow your criticism, I would say: it could also be interesting to post the same question on other forums, with regards to other coding languages and software: why not?
 
David Carlisle
David Carlisle
@VincentVerheyen because asking about random pdf files unrelated to tex and saying "was this made by tex" is just a pointless open door exercise in encouraging spam. The site is designed to close down questions that are not related to the subject.
 
phell
phell
@DavidCarlisle Hee hee hee. I guess the root cause of my need to debug is me being lazy and use the universities unofficial, an outdated, class file instead of starting from scratch.
 
Paulo Cereda
Paulo Cereda
@VincentVerheyen Crossposting, IMHO, just evidences the lack of scope and focus of the question itself. Besides, if inside the StackExchange universe, crossposting is disallowed.
 
Anonymous
@DavidCarlisle But, if there was, for example, another OP which would ask "How to (unambiguously) identify pictures made in LaTeX"?, then the answers from that question could be perhaps very much used in my OP, right? Perhaps I asked too much questions at once again, from which arose a lot of criticism. Or perhaps the answer requires too much (rather than being off-topic, as some people say) knowledge of LaTeX?
 
Joseph Wright
Joseph Wright
11:45 AM
@VincentVerheyen There's really no logic to such a question: images are just a description of where to place stuff on a canvas, and ultimately you cannot link back from an image per se to a definitive creator program
@VincentVerheyen TeX isn't even an image-production software in that sense: you can arrange to ask the low-level PDF mechanisms to draw things via interfaces that make it look more TeX-like, but that's really not directly TeX-related
 
Anonymous
@JosephWright But if you would know a lot about (I'm just guessing out loud, because I lack the expertise) about pixel output in relation to a specific package which seems very likely to be able to generate the desired output, then one might argue that its quite possible that the image was created using a LaTeX package, if all outputs match.
 
Paulo Cereda
Paulo Cereda
One image was produced with LaTeX, the other is not. Spot which one.
 
Anonymous
@PauloCereda I can see a big difference in the pixel output, when you zoom completely into the text. I wouldn't be able to tell why exactly, but an expert might know.
 
Paulo Cereda
Paulo Cereda
@VincentVerheyen /sigh I won't even bother anymore. Sorry for the noise.
 
Anonymous
@PauloCereda ;D No, it's interesting; because it's something concrete which we can discuss. There might also be a difference in the amount of shades of gray generated.
 
phell
phell
11:57 AM
@JosephWright Do you know if there was any changes involving date and time in l3kernel?
 
Joseph Wright
Joseph Wright
@phell We added an interface for the \date and \time primitives relatively recently
 
phell
phell
@JosephWright I think I narrowed it down to happening November or December last year.
I was using SIunitx and then an update of l3kernel made some incompatibility with re-defined date and time in a class file.
 
Joseph Wright
Joseph Wright
@VincentVerheyen Doesn't prove anything, and as has been pointed out several times the paper you linked to says it was created with OO and that is almost certainly the easiest way to make it, based on the various issues with the typography. But in the end you cannot say 'PDF X must have been created by application Y'
@phell siunitx is one of the test packages that gets run before any release of l3kernel: it's fine with the latest release (oddly enough I don't break my own packages)
 
Paulo Cereda
Paulo Cereda
@VincentVerheyen Outputs are not unique, they have no fingerprint. You could use the very same font metrics in TeX and Word-like applications and the result would be the same. Sorry, it is not anything near to concrete, it's just pure speculation. Saying pixel level, output analysis, etc., to every question posed here without considering experts' voices (e.g, David and egreg) sounds counterproductive...
 
phell
phell
@JosephWright Okay, thank you. :)
 
Joseph Wright
Joseph Wright
12:05 PM
@phell There was a report of some daft class redefining (I think) \time which caused issues
 
phell
phell
@JosephWright One of them are
\renewcommand{\month}[1]{\renewcommand{\MDHmonth}{#1}}
I am not saying you broke anything. :)
 
Joseph Wright
Joseph Wright
@phell Ah, that would be an issue, yes
 
phell
phell
@JosephWright I guess I should start making my own style file instead.
My LaTeX teacher is pretty old school. I am starting to understand the advantages with using less packages to avoid compatibility issues in the future. But I am still in the "oh cool, another package that looks nice" phase.
 
David Carlisle
David Carlisle
@VincentVerheyen not really, no.
 
 
2 hours later…
phell
phell
1:46 PM
What is your view on forcing an image to a certain place by putting it in a mathemathical environment?
 
David Carlisle
David Carlisle
@phell why put it in a math environment?? (images only move if you specify they can move, so not sure what you mean by "force")
 
phell
phell
@DavidCarlisle I mean instead of putting barriers or setting the image to stay in a section. In my experience images moves around quite a bit from where I expect them to be. This is because of the rules for punishing "bad layout".
For example images likes to be at the top or bottom of a page.
I hope I didn't annoy you too much with the question. :)
 
David Carlisle
David Carlisle
@phell no, images and text are both positioned in the same way, in either case they will appear in the position in the source file unless you wrap them in an environment who's only purpose is to allow it to move. Compare hello and \begin{figure}hello\end{figure} the purpose of the figure environment is to allow the content to be moved (whether it is hello or \includegraphics it makes no difference. You could put \includegraphics in math but you can just put it in the main text.
 
phell
phell
@DavidCarlisle Oh I see. I never understood that the figure environment was not needed.
 
David Carlisle
David Carlisle
@phell but of course latex moves figures for a reason, if you routinely put in large blocks like tables and figures without allowing them to float the page breaking will be horrible with lots of white space.
 
phell
phell
1:58 PM
@DavidCarlisle You are right. I just think it looks a bit funny when a page has a text block, then an empty block, and finally the image at the bottom. I guess it is the least worst solution.
 
David Carlisle
David Carlisle
@phell bottom is latex's last choice so it will only place it there if you have not allowed it to be mid page (with h) and not allowed it to be top (somehow)
 
phell
phell
@DavidCarlisle I will read up on the includegraphics package.
 
David Carlisle
David Carlisle
@phell as I say float positioning is completely unrelated to graphics inclusion (or any package) it is handled by the latex format.
 
phell
phell
@DavidCarlisle Oh, the HTP is for any block.
 
David Carlisle
David Carlisle
@phell not sure what you mean by block it is for floats (figure, table and any similar environment) but the positioning of a figure only depends on the vertical size, the algorithm doesn't care if the content is includegraphics or hello or vspace{3in} see:
603
A: How to influence the position of float environments like figure and table in LaTeX?

Frank MittelbachTo answer this question one has to first understand the basic rules that govern LaTeX's standard placement of floats. Once these are understood, adjustments can be made, for example, by modifying float parameters, or by adding certain packages that modify or extend the basic functionality. LaTe...

 
phell
phell
2:08 PM
@DavidCarlisle I will read the answer. Yes, I meant float when I said block.
 
phell
phell
2:41 PM
Aha, I have to give all options I allow and not just try to forbid with !b.
 
David Carlisle
David Carlisle
@phell :-) I did wonder. The most common mistake people make is use [h] and wonder why the float goes to the end of the document, [h] means not t not b not p so if it is too low on the page to fit here it's not allowed anywhere... (in latex2.09 it really went to end of document, 2e complains and changes it to [ht] to give itself a chance.)
 
phell
phell
@DavidCarlisle Hee hee hee. I guess we novices always come up with strange things. :)
 
David Carlisle
David Carlisle
@phell ! in !b does not mean not it means ignore the constraints and try to put in in b even if it would not otherwise be allowed.
 
phell
phell
@DavidCarlisle Ah right. List the options you want and remove the ones you don't want.
I am impressed by your patience. You must have told people about this a million times.
 
 
2 hours later…
yo'
yo'
4:38 PM
@DavidCarlisle I would have advocated for writing b! in the manuals, like an imperative sentence...
btw, from the right pane, I have a feeling that the chat is full of cookie-eating ducks (cookie ducksters?) :) @Paulo
 
Paulo Cereda
Paulo Cereda
@yo' Quack! :)
 
phell
phell
4:58 PM
/me hands yo' a cookie.
 
yo'
yo'
> \newcommand{\fl}{\hspace*{-\mathindent}}
I mean, the moment when you suddenly feel 20 years younger :)
 
Joseph Wright
Joseph Wright
@yo' Ooops
 
yo'
yo'
5:23 PM
@JosephWright iopart.cls
 
yo'
yo'
5:34 PM
\begin{center}$
\begin{array}{cc}
lovely
 
David Carlisle
David Carlisle
5:57 PM
@JosephWright last seen 2011, not a good sign:( tex.stackexchange.com/users/1237/hideo-umeki
 
Christian Hupfer
Christian Hupfer
 
Paulo Cereda
Paulo Cereda
@ChristianHupfer Das cookie! :)
 
Christian Hupfer
Christian Hupfer
@PauloCereda My god, it's full of stars cookies (Tis tyme wis korekt speling ;-))
 
Johannes_B
Johannes_B
6:12 PM
 
yo'
yo'
@Johannes_B sorry, I'm unable to follow any discussion on any wiki
 
Christian Hupfer
Christian Hupfer
7:08 PM
\setsansfont{Comic Sans MS}
0
Q: Problem with AMS packages

DavidI have a problem that occurs every time I use one of the AMS packages (regardless of the document class or template I use), for instance amsmath or amsthm. In the next line after the \usepackage command LaTeX identifies an error File ended while scanning use of \@newenv. It also shows several war...

amsmath
 
David Carlisle
David Carlisle
@ChristianHupfer we Davids have excellent choice in fonts
 
Aaron Hall
Aaron Hall
:)
 
Christian Hupfer
Christian Hupfer
@DavidCarlisle Your Alter Ego, I assume? ;-)
 
Aaron Hall
Aaron Hall
it's going on my second year using orgmode to beamer, and I still have not figured out how to skin my beamer slides with different background colors.
 
David Carlisle
David Carlisle
@ChristianHupfer as you can tell by my grasp of Cyrillic
 
Aaron Hall
Aaron Hall
7:18 PM
I've been hoping someone's going to tell me the magic insertion point, but no-one does...
 
Christian Hupfer
Christian Hupfer
@AaronHall SamCarter is one of our beamer experts, in my point of view -- but he's/she's not in chat very often
 
David Carlisle
David Carlisle
@AaronHall there's never a site mod around when you need one.
 
Christian Hupfer
Christian Hupfer
@DavidCarlisle: What have mods ever done for us @PauloCereda ;-)
 
David Carlisle
David Carlisle
@ChristianHupfer @JosephWright I should flag the above to defend your honour
 
Christian Hupfer
Christian Hupfer
@DavidCarlisle Go ahead :D
@DavidCarlisle @AaronHall: I've left a comment addressed to @samcarter, perhaps she will appear
 
David Carlisle
David Carlisle
7:23 PM
@ChristianHupfer I would but flags are network wide and not all people pinged would understand the strange norms in this room
 
Christian Hupfer
Christian Hupfer
@DavidCarlisle There are norms here? /Shocked
 
yo'
yo'
Oh the badges at Music: Practice & Theory are the coolest badges in the world (by design I mean)!
 
Christian Hupfer
Christian Hupfer
@yo' mozart badge? [tag:AC/DC] badge?
 
yo'
yo'
@ChristianHupfer Mozartkugeln
 
Christian Hupfer
Christian Hupfer
@yo' :D
 
samcarter
samcarter
7:42 PM
@ChristianHupfer @ChristianHupfer Hi!
@AaronHall @AaronHall Unfortunately I don't know anything about orgmode, but if you want background colours, you can find enough to last a very long time at tex.stackexchange.com/a/171482/36296
 
Christian Hupfer
Christian Hupfer
@samcarter Hello ;-)
 
Aaron Hall
Aaron Hall
7:59 PM
Woop!
Apparently I'm going to have to figure out how to read the implementation of OrgMode to Latex in Emacs, (I've spent hours tweaking headers to no avail) and/or figure out how to actually make templates for Beamer (wah).
 
egreg
egreg
8:11 PM
Those who can, do. Those who Kant, lipsum! Indeed, indispensible. — Steven B. Segletes 2 mins ago
 
samcarter
samcarter
@AaronHall Is there any way to insert something like \setbeamercolor{background canvas}{bg=yellow!20} into OrgMode?
@AaronHall Actually google suggest #+LATEX_HEADER: \setbeamercolor{background canvas}{bg=yellow!20} but I have no idea how to test this.
 
yo'
yo'
8:24 PM
@egreg lol
 
Aaron Hall
Aaron Hall
ok...
I should really just learn how to make templates...
and where to keep them (github?)
 
Canageek
Canageek
8:41 PM
Help? I've got U+301 somewhere inside my bib file, so it is spewing errors (well, one error), but searching for [^\x00-\x7F] finds nothing
Anyone have ideas?
 
egreg
egreg
@Canageek Look for an acute accent like in Montréal
 
Canageek
Canageek
Yeah, I figured that, any way to do it non-manually
 
David Carlisle
David Carlisle
@Canageek hexdump?
 
Anonymous
8:58 PM
@PauloCereda Regarding Was anything in Satoshi Nakamoto's original Bitcoin paper compiled in LaTeX?, I would like to answer a question; based on the expertise in this chat ... But, unfortunately, the question is on hold.
 
David Carlisle
David Carlisle
@Canageek sed -e 's/\xcc\x81/X/g' turns UTF-8 U+0301 into X
 
Canageek
Canageek
@DavidCarlisle Installing Cygwini and running sed is the next step after testing if emacs can find it
 
David Carlisle
David Carlisle
@Canageek Mx-hexlify-buffer then look for cc81
 
Canageek
Canageek
@DavidCarlisle Failed i-search
 
cfr
cfr
@VincentVerheyen Wanting to answer it doesn't make it on topic. That why nobody else has answered it.
 
Anonymous
9:02 PM
@DavidCarlisle Do you think a pdftohtml -xml (as proposed in bitcoin.stackexchange.com/a/42466) could add any info on the source of the document?
 
David Carlisle
David Carlisle
@VincentVerheyen see the standard banner added when the question was closed, you need to edit the question to bring it on topic to make 5 people vote to open it (although that seems unlikely as it os so clearly not on topic)
@VincentVerheyen no none at all.
 
Canageek
Canageek
! Package inputenc Error: Unicode char ́ (U+301)
(inputenc) not set up for use with LaTeX.
 
David Carlisle
David Carlisle
@Canageek well that's easy to fix!! just use \DeclareUnicodeCharacter{0301}{HERE} and run it again
 
Canageek
Canageek
@DavidCarlisle ?
OH ok
IU get you
Wait
The offending character is Garc{\'\i}a
Which I didn't look at as it is escaped.
Is it inserting a character that it can't print?
 
David Carlisle
David Carlisle
@VincentVerheyen why do you have any thought at all that it isn't made by openoffice? the whole thread is just completely uninteresting:-)
@Canageek it's a combining character so added as i U+0301 but tex can't really handle unicode suffix combining characters,
 
Canageek
Canageek
9:08 PM
@DavidCarlisle So is there a unicode character there, or is the {\'\i} making it?
 
Anonymous
@DavidCarlisle I already told you that I just quote some higher education institution and some on-line computer scientists. --- I don't have enough knowledge to figure out whether it was created in OpenOffice or not. --- But, another question: is this (verbeia.com/tex/fontsupport/arial/doc/arialstydoc.pdf) the most basic way to output ArialMT from a Tex document?
 
David Carlisle
David Carlisle
@Canageek \'\i wouldn't make any unicode by default, is this biblatex?
 
Canageek
Canageek
@DavidCarlisle yes, biblatex+biber
 
David Carlisle
David Carlisle
@Canageek complain to Joseph:-) It could make the precomposed U+00ED
 
yo'
yo'
@VincentVerheyen You can use XeLaTeX or LuaLaTeX and then any TTF or OTF font
 
Canageek
Canageek
9:11 PM
@JosephWright Can you explain why Garc{\'\i}a is screwing up pdflatex?
Ok, replaced with \'{i}
 
Anonymous
:29324874 Thanks, so using:

\usepackage{fontspec}
\setmainfont{Arial}

hé. Ok.
 
yo'
yo'
@VincentVerheyen Also, I saw your message on top of the post. Next time, you better flag the post for moderator attention and put your message in the popup form. I did that for you.
@VincentVerheyen you can't do what you want to do in 1 message in the chat ;-)
 
Canageek
Canageek
@DavidCarlisle That seemed to work. I think it should look the same, right?
 
David Carlisle
David Carlisle
@VincentVerheyen I don't know, perhaps they just don't know about the document properties menu
@Canageek unless redefined \'\i and \'i make identical output in classic latex definitions
 
Anonymous
@DavidCarlisle But, as people have pointed out, contents of the document properties menu, is the most easy changeable feature, and should thus not be naively trusted.
 
Anonymous
9:14 PM
@yo' Thanks, so I'll delete that message on top of post. I will look into more things anyway, beforehand.
 
egreg
egreg
@Canageek There's a question on the site about that problem; using Garc{\'i}a is the solution (or directly García, of course).
 
Canageek
Canageek
@egreg Arg. I just spent like half an hour finding that, but dismissed it as it was properly escaped and such things should always be safe.
 
David Carlisle
David Carlisle
@VincentVerheyen sure if you have any reason to doubt it, but when I drive into a town and it says "welcome to Smaztaroo" on a roadsign, I don't spend weeks worrying if someone has illegally repositioned the signpost, I assume unless there is any evidence to the contrary that I am in fact in Smaztaroo
 
egreg
egreg
@DavidCarlisle When I go to Slovenia I see many signposts with even funnier names, mostly without any vowel.
 
Anonymous
@DavidCarlisle We have reasons to doubt, as the person(s) who wrote such paper has deliberately chosen to hide his/her/their identity(ies). Therefor, everything might have been done to try and remove fingerprints from both spelling, word use, and grammar; as well as from document design.
 
Canageek
Canageek
9:16 PM
@egreg Is there an option to have biber tell me about unsafe characters, so that I get an actual line number?
 
David Carlisle
David Carlisle
@VincentVerheyen sorry the whole thread is just cyclic and off topic for this site.
 
egreg
egreg
@Canageek See also
6
Q: Input encoding error after upgrading from Biber 1.9 to Biber 2.1

cthUnder biber 2.1 my biblatex produces input encoding errors. The same procedures worked smoothly under biblatex 2.9a/biber 1.9, now that I upgraded to biblatex 3.0/biber 2.1 I get the following message when pdflatex runs after biber: Package inputenc Error: Unicode char \u8:́ not set up for u...

biblatex biber input-encodings
 
Canageek
Canageek
@egreg I have unkind things to say about the person that decided being technically correct was more important that producing output that the only LaTeX engine that matters can use.
 
Anonymous
If I use

\documentclass{article}
\usepackage{fontspec}
\setmainfont{ArialMT}
\begin{document}
A purely peer-to-peer version of electronic cash would allow online payments to be sent directly from one party to another without going through a financial institution. Digital signatures provide part of the solution, ...
\end{document}

Then, the Adobe Acrobat document properties panel shows: ArialMT, Embedded subset (Type: TrueType (CID), Encoding: Identity-H).

But in the Bitcoins PDF, it is shown: ArialMT, Embedded subset (Type: TrueType, Encoding: Built-in).
 
Canageek
Canageek
@egreg I left a comment on that one
Well, just so whomever made that call knows I just wasted about an hour combing through my .bib file for non-unicode characters, converting it into hex and looking for cc81 until someone finally showed me \DeclareUnicodeCharacter{0301}{HERE}. So thanks for making biber generate unsafe output for pdflatex by default. Wouldn't that be better as the non-default option?
 
David Carlisle
David Carlisle
9:30 PM
@VincentVerheyen the main body text in the document isn't sans serif
 
egreg
egreg
@Canageek I'm not sure why the converter spits out the wrong combination.
 
Canageek
Canageek
@egreg That is detailed in the second link you sent me.
@egreg Being technically correct is more imporant that working.
 
egreg
egreg
@Canageek Well, I don't really understand the real reason.
 
Anonymous
@DavidCarlisle Hmm, yes indeed. It seems that ArialMT is supposed to be sans-serif. So, perhaps another font is used in the original Bitcoins doc (at that place) then. --- David, I understand that the post is off-topic quite a bit, as a more complete answer might ask for information which is not wholly Tex-specific (in some way). I will perhaps transfer the question and answer to e.g. my own website, provided I find some time for that. My apologies and thank you for all of your help.
 
Canageek
Canageek
6
A: Input encoding error after upgrading from Biber 1.9 to Biber 2.1

PLKThis is a change in biber 2.1 with \i in particular. Now biber properly encodes this as a dotless i (ı - 0x0131) with a combining accent. Even though biber always converts to precomposed (NFC) form on output and is therefore generally as friendly as possible to inputenc, there is no precomposed f...

 
Paulo Cereda
Paulo Cereda
9:39 PM
Smaztaroo! :D
 
David Carlisle
David Carlisle
@PauloCereda duck free zone
 
Paulo Cereda
Paulo Cereda
@DavidCarlisle oh :(
 
David Carlisle
David Carlisle
@PauloCereda of course it could be that ducks are allowed, but someone has faked the sign, you never can trust these things
 
Paulo Cereda
Paulo Cereda
@DavidCarlisle aaaah good point. :)
 
David Carlisle
David Carlisle
10:33 PM
If there is any justice in the world I'll steal @egreg's \includegraphics tick.
 
egreg
egreg
10:45 PM
@DavidCarlisle The solution with just one \includegraphics is already online.
 
David Carlisle
David Carlisle
@egreg unsurprising:-)
 

« first day (2006 days earlier)      last day (2934 days later) »