[ / / / / / / / / / / / / / ] [ dir / animu / ausneets / dempart / doomer / s / strek / tingles / vg ][Options][ watchlist ]

/tech/ - Technology

You can now write text to your AI-generated image at https://aiproto.com It is currently free to use for Proto members.
Email
Comment *
File
Select/drop/paste files here
Password (Randomized for file and post deletion; you may also set your own.)
* = required field[▶ Show post options & limits]
Confused? See the FAQ.
Expand all images

File (hide): 5bb35e0dedb41f9⋯.png (21.82 KB, 1200x500, 12:5, 1*lKvH1FlonI5bqKYyiMNWqw.png) (h) (u)

[–]

 No.989111>>989123 >>989137 >>989197 >>989274 >>994316 >>994353 >>994450 >>1005886 >>1011433 >>1018020 [Watch Thread][Show All Posts]

It's the current year. Is there anything out there better than LaTex for typesetting technical documents? The syntax used in LaTex is god awful, but of course the only reason anyone puts up with it is the beautifully typeset document that results. Since I've been out of academia for a while, I'm wondering if there is any viable alternative, or are people still doing things the old ways? I do see that many popular editors have preview functions for latex formulas, that's nice, but it doesn't entirely ameliorate the pain.

 No.989114>>989429 >>1011287

There isn't. I use LaTeX through org, so my texts are mostly org with some latex code here and there. It works with beamer as well, and you can even put code with the result going to the document.


 No.989121>>989182

Unbelievable really. Now I'm starting to get flashbacks of the horrible \sum_{k=1}{n} z \left( \frac{\sqrt[3]{x}}{x-1} \right) garbage. Mathematica has WYSIWYG typesetting abilities, entry is quick but it's a closed system and the quality isn't even close to LaTex.

Everything is a bloody hack. TeX may have been great, but since the 70s everyone seems to have just piled up a tower of shit over it. The situation with error messages is probably even worse than the syntax!!! ARGGGGGGGGGGGGGHHHHHHHHH.


 No.989123>>989129 >>989137 >>1011433

>>989111 (OP)

The best thing you can do is write your texts in a sufficiently high-level markup language like Markdown, reStructuredText, AsciiDoc, OrgMode or TexInfo, then convert it to LaTeX using something like Pandoc.

TeX is a typesetting system which is programmable. LaTeX sits somewhere between markup and typesetting, it abstracts a lot of details away, but it still has you do manual typesetting when its facilities are not sufficient. I write a lot of math, so sadly none of the above markup languages are really good enough for me. The best I can do is write most of it in a higher-level markup, convert automatically and then manually make adjustments.


 No.989129

>>989123

It's a shame that OCR tools are still so primitive, it'd be far more efficient for me to perform entry by hand (despite being a very fast typer) and then make the fine adjustments.


 No.989132>>989272

Based Xah Lee hates LaTex too: http://xahlee.info/cmaci/notation/TeX_pestilence.html it really is quite a pain. The sad thing is that LaTex results are too impressive not to use, but it is surprising that package writers have just put up with all of the shortcomings and worked on more useful features which make it harder to resist using LaTex.


 No.989133>>989135 >>989714 >>1002660

groff -ms is for men of class, intelligence, culture, and wit.


 No.989134>>1011287 >>1011294 >>1011433

Speaking of LaTeX and TeX, LuaTeX has hit version 1.0 a while ago.

http://www.luatex.org/

LuaTeX is TeX with the ability to script it in Lua. You can use Lua in your documents to for example write loops, but the real power is that it exposes the TeX internals for use with a proper scripting language. This should make the job of package writers much easier.

Even if you don't want to use Lua, you should be using LuaTeX: you can natively use Unicode characters and it supports OpenType fonts. It's much more readable when I can write out Greek letters or math characters in my document instead of having to spell out their names.

BTW, has anyone tried out ConTeXt? Apparently it's idea is different from LaTeX, rather than hiding the typography its aim is to make it accessible. Sounds interesting.


 No.989135

>>989133

groff is a broken pile of garbage. Use heirloom-troff, it at least supports a more modern way of handling fonts and fucking Unicode: http://heirloom.sourceforge.net/doctools.html


 No.989137>>994353

File (hide): 0e684b2ba7cda91⋯.png (173.75 KB, 798x1060, 399:530, pdf.png) (h) (u)

File (hide): 400dae38e8c2058⋯.png (73.28 KB, 696x984, 29:41, snapshot16.png) (h) (u)

>>989111 (OP)

I think it depends a lot on what kind of "technical documents". If you deal with a lot of math, LaTeX seems pretty standard.

I recently started working producing technical documentation in the aviation field, and the standard format we use is an XML-based one called S1000D. Rather than lots of math, I deal with lots of configuration management, like alternative parts and configurations for different customers. And there's a big focus on single-sourcing, using the same source to produce a richer variety of outputs than just PDF, like so-called Interactive Electronic Technical Publications (IETPs). This requires more focus on semantic markup, rather than typesetting.

I'm not sure if it necessarily counts as an alternative, though, because like >>989123 points out, you can write documents in one of these semantic markup languages, and then output (La)TeX commands from it to produce a PDF with the TeX engine. I guess the actual alternative to LaTeX in the XML world is probably XSL-FO.

I haven't seen a detailed comparison between something like Apache FOP and TeX, though. First pic is an example of S1000D's standard page-oriented layout rendered with Apache FOP and some FOSS stylesheets, the second is the details of the layout from the specification itself.


 No.989161

Is Postscript not enough for you wretched fiends?


 No.989182>>1011078 >>1011433

File (hide): eddba0bcf0f6dc2⋯.png (51.18 KB, 1212x426, 202:71, notes.png) (h) (u)

>>989121

>Mathematica

If you are not being forced to use LaTex, I strongly recommend staying with Mathematica, assuming you have good familiarity with it. I use Mathematica to type up all of my math notes/assignments because it allows very quick entry owing to the keyboard shortcuts, coupled with a sane syntax that can be turned into "traditional form" in place. It can also handle multiline equation alignments, although I think the UI for this part needs a bit of work. I especially like being able to stay in one environment, without having to work on the barely comprehensible latex file, and run the build cycle to view the pretty output in a pdf, plus you get the full power of Mathematica at your disposal if you need to run a calculation or include a chart etc.

It's not quite Latex quality as you say, but it's certainly good enough for many tasks. Pic related shows a homework question I typed in it last year. The downside is that Mathematica isn't free, so if that is a sticking point, perhaps the Sage folks have some alternatives you can look into.


 No.989197>>989199

>>989111 (OP)

> The syntax used in LaTeX is god awful

It's perfect, if you want something more braindead you can use R Markdown and still use some of your LaTeX knowledge


 No.989199>>989300

>>989197

It is absolutely disgusting. I would almost go as far as to call it a write-only syntax, which makes sense as you're writing, but atrocious to read back at a later point in time.


 No.989200>>989205

mydocument.txt


 No.989205>>989254 >>989355

>>989200

Submission denied.


 No.989254

>>989205

What if I was to | fmt > mydocument.txt?


 No.989272>>989434

>>989132

He is right in that the representational notation is shit, but what is the alternative? Sooner or later you will have to put your math on display, and that's when you will have to deal with typography. Even if you were to make all the operators semantic, you will run into cases where even that is not enough.

Let's take an equation that needs to be broken up over several lines, something like a = b = c = d = e. Except of course each of these will be a large collection of formulas. Where do you make the break? What do you align on? There is no automatic solution, so eventually you will have to insert line breaks and alignment manually.


 No.989274>>989374

>>989111 (OP)

>the only reason anyone puts up with it is the beautifully typeset document that results

No shit. That's what typesetting means. That morons actually try to use an end product system for taking notes says more about how fucking dumb they are than anything about how bad TeX might be.


 No.989300>>989304

>>989199

Is groff any better?


 No.989304>>989314 >>989391

>>989300

Have you ever looked at the source of a manpage? Because that's what roff looks like.

.\" DO NOT MODIFY THIS FILE!  It was generated by help2man 1.47.3.
.TH LS "1" "January 2018" "GNU coreutils 8.28" "User Commands"
.SH NAME
ls \- list directory contents
.SH SYNOPSIS
.B ls
[\fI\,OPTION\/\fR]... [\fI\,FILE\/\fR]...
.SH DESCRIPTION
.\" Add any additional description here
.PP
List information about the FILEs (the current directory by default).
Sort entries alphabetically if none of \fB\-cftuvSUX\fR nor \fB\-\-sort\fR is specified.
.PP
Mandatory arguments to long options are mandatory for short options too.
.TP
\fB\-a\fR, \fB\-\-all\fR
do not ignore entries starting with .
.TP
\fB\-A\fR, \fB\-\-almost\-all\fR
do not list implied . and ..
.TP
\fB\-\-author\fR
with \fB\-l\fR, print the author of each file
.TP
\fB\-b\fR, \fB\-\-escape\fR
print C\-style escapes for nongraphic characters
.TP
\fB\-\-block\-size\fR=\fI\,SIZE\/\fR
scale sizes by SIZE before printing them; e.g.,
\&'\-\-block\-size=M' prints sizes in units of
1,048,576 bytes; see SIZE format below
.TP
\fB\-B\fR, \fB\-\-ignore\-backups\fR
do not list implied entries ending with ~
.TP
\fB\-c\fR
with \fB\-lt\fR: sort by, and show, ctime (time of last
modification of file status information);
with \fB\-l\fR: show ctime and sort by name;
otherwise: sort by ctime, newest first
.TP
\fB\-C\fR
list entries by columns
.TP
\fB\-\-color\fR[=\fI\,WHEN\/\fR]
colorize the output; WHEN can be 'always' (default
if omitted), 'auto', or 'never'; more info below
.TP
\fB\-d\fR, \fB\-\-directory\fR
list directories themselves, not their contents
.TP
\fB\-D\fR, \fB\-\-dired\fR
generate output designed for Emacs' dired mode
.TP
\fB\-f\fR
do not sort, enable \fB\-aU\fR, disable \fB\-ls\fR \fB\-\-color\fR
.TP
\fB\-F\fR, \fB\-\-classify\fR
append indicator (one of */=>@|) to entries
.TP
\fB\-\-file\-type\fR
likewise, except do not append '*'
.TP
\fB\-\-format\fR=\fI\,WORD\/\fR
across \fB\-x\fR, commas \fB\-m\fR, horizontal \fB\-x\fR, long \fB\-l\fR,
single\-column \fB\-1\fR, verbose \fB\-l\fR, vertical \fB\-C\fR
.TP
\fB\-\-full\-time\fR
like \fB\-l\fR \fB\-\-time\-style\fR=\fI\,full\-iso\/\fR
.TP
\fB\-g\fR
like \fB\-l\fR, but do not list owner
.TP
\fB\-\-group\-directories\-first\fR
group directories before files;
.IP
can be augmented with a \fB\-\-sort\fR option, but any
use of \fB\-\-sort\fR=\fI\,none\/\fR (\fB\-U\fR) disables grouping
.TP
\fB\-G\fR, \fB\-\-no\-group\fR
in a long listing, don't print group names
.TP
\fB\-h\fR, \fB\-\-human\-readable\fR
with \fB\-l\fR and/or \fB\-s\fR, print human readable sizes
(e.g., 1K 234M 2G)
.TP
\fB\-\-si\fR
likewise, but use powers of 1000 not 1024
.TP
\fB\-H\fR, \fB\-\-dereference\-command\-line\fR
follow symbolic links listed on the command line
.TP
\fB\-\-dereference\-command\-line\-symlink\-to\-dir\fR
follow each command line symbolic link
.IP
that points to a directory
.TP
\fB\-\-hide\fR=\fI\,PATTERN\/\fR
do not list implied entries matching shell PATTERN
(overridden by \fB\-a\fR or \fB\-A\fR)


 No.989314>>1005832 >>1005842

>>989304

>all those escapes

I guess you can't expect much better from both a GNU program and an autogenerator. The way troff is supposed to be used is by having macro directives starting in newlines.

Also m*doc is a pretty terrible choice for papers, use me or mom instead. That's what they're made for.


 No.989355

>>989205

>The way troff is supposed to be used is by having macro directives starting in newlines.

But that would be too easy to read.


 No.989374

>>989274

What a strange comment.


 No.989391>>989428 >>990020 >>1005815 >>1005842

File (hide): 190880563c4f665⋯.png (264.39 KB, 334x471, 334:471, troff-books.png) (h) (u)

>>989304

I have now, so thanks. I just remembered it's widely available for typesetting, yet I rarely see it being mentioned - for good reason, it appears.

I don't think latex is that bad for inputting math, but then again I can't think of a reasonable hypothetical alternative within latex. I (also) use latex because I don't like WYSIWYG editing, so that one won't do me any good.


 No.989402>>989435 >>994353

Total fagshit if you aren't printing it.

Justified text and all of those great styling abilities are all available in LibreOffice Writer by default, no extra addons/packages/whatever required.

SVG does vector graphics. No need for describing the vg specifically for your document language. It's called interoperability.

LibreOffice Math does all formulas (not just math, also chemistry, physics and so on) which you can also use in all other LibreOffice applications.

>inb4 but you can't style grouped text

Utter bullshit. It's called preset and it's in the top left corner. You can assign presets to paragraphs and if you change the preset all paragraphs change.

It also supports subpresets/groups. Just like CSS or LATEX but with a window.

I know you faggots hate being productive.

>inb4 MUH BLOAT.

It's bloatshit you don't have to clean up after. It does what it's supposed to.


 No.989428

>>989391

It mostly doesn't look like that.

https://9p.io/7thEdMan/

This has some actual troff source.


 No.989429

>>989114

>being a baguette on the internet and not hiding it


 No.989434>>989469

>>989272

> Where do you make the break?

The problem is that people cannot decide if LaTex is a typing setting tool, or a utility for generating nicely typeset documents. The solution to that long equation would ideally, be to let the layout engine choose a best fit, but allow the user to hint/force the line breaks. This is how LaTex handles hyphenation in paragraphs.

The selling point of LaTex to mathematicians, is that it can handle the minutiae of laying out your publications, allowing you to keep your focus on mathematics. Sadly, this isn't quite true, as you often spend time tweaking/fighting with the system to achieve your desired result. That seems to be where Xah has a problem, coupled with the nasty syntax.


 No.989435>>989804

>>989402

Point and click is a big pain, it might be fine if you're adding the odd equation in your notes, but I cannot imagine doing a homework assignment in it.


 No.989469>>989482 >>989728

>>989434

> The solution to that long equation would ideally, be to let the layout engine choose a best fit, but allow the user to hint/force the line breaks. This is how LaTex handles hyphenation in paragraphs.

I'm sure if it was possible TeX would have already implemented it.

> The problem is that people cannot decide if LaTex is a typing setting tool, or a utility for generating nicely typeset documents.

It's in-between. If you are satisfied with vanilla presets you can write an entirely semantic document, but you rarely are. Even something like typesetting an honorific requires manual fiddling:

% wrong
Hello Dr. Frankenstein
% correct
Hello Dr.~Frankenstein

Of course you could define an "honorific" macro, but that makes the code ugly again:

Hello \honorific{Dr.}{Frankenstein}.
% If you need to address many doctors you can also define a special doctor command on top of the honorific command
Hello \dr{Frankenstein}.

I guess the latter would be the most idiomatic solution.

>>989434

>The selling point of LaTex to mathematicians, is that it can handle the minutiae of laying out your publications, allowing you to keep your focus on mathematics. Sadly, this isn't quite true, as you often spend time tweaking/fighting with the system to achieve your desired result. That seems to be where Xah has a problem, coupled with the nasty syntax.

I know. I had been taking lecture notes in LaTeX and wrote my thesis in it. I had to write a lot of math, so I know exactly how shit LaTeX is, but it's still the least bad thing we have. I yet have to see an alternative.

The best thing I was able to come up with is some s-expression tree-format which could then be converted to LaTeX code.

'((frac 1
(power '(2 π)
(/ n 2)))
(integral #:from (power ℝ n) #:d x
'((apply f x)
(pow e
'(-i y x))))

This would then expand to

\frac{1}{(2 π)^{n / 2}} \int_{ℝ^n} f(x) e^{-i y x} \dop{x}

This wouldn't even be hard to implement, but I cannot see many people actually liking this.


 No.989482>>989499 >>989611

>>989469

Well I for one like it already. I don't want to continuously shill Mathematica, but that s-expression approach is employed in that language, and you can use the same code for perform equations as for displaying them.

Since your code looks rather Rackety, that reminds me, there was definitely some work in the racket community for dealing with programmable typesetting, I need to take a look at that again.


 No.989499>>989509 >>989611

>>989482

>there was definitely some work in the racket community for dealing with programmable typesetting

https://docs.racket-lang.org/pollen/


 No.989509

>>989499

Seems to be more for interactive html books, rather than math specifically. For that, it's just providing hooks for MathJax.


 No.989611

>>989499

Pollen still uses LaTeX syntax for math:

https://docs.racket-lang.org/pollen/mini-tutorial.html#%28part._.Math_typesetting_with_.Math.Jax%29

>>989482

Yes, Racket was what I was having in mind, although this particular idea would work with any Lisp. Once you get a taste of s-expressions you really don't want to use anything else.

I could see such a system working as a layer on top of TeX, but the most important thing would be to make it programmable. Part of the reason why LaTeX requires constant fiddling is because you simply cannot anticipate everything an author will need. A text markup system needs to be extensible to allow for custom domain-specific extensions. LaTeX gives you direct access to the underlying TeX, which is why LaTeX has been so successful for so long. As much as I love Pandoc, its static markup formats are simply not good enough for my needs.


 No.989714

>>989133

Actually the -mom package is unironically the GOAT. Groff is so much better than faggy TeX is unbelievable why people dont use it more.

You can literally just run a plain text file through groff, with absolutely no markup, and it will at least produce sane output. Unlike TeX where its impossible to even get output without 10k of boilerplate bullshit.


 No.989728>>989741

>>989469

For a math solver AND equation renderer S-expressions makes sense. But without actual solving functionality I dont get the point of S-expressions. Its easier to type LaTeX than that.


 No.989741

>>989728

The idea is to have 100% semantic markup. A tree-like expression is the only sensible way of doing it because a math expression is a tree structure. People are just not used to thinking of an expression like "2 + 3 / 4" as a tree, they think "take two, add three, divide by four, except you have to take into account order of operations, so first read the division, then take two and add the result of the division". With a tree-syntax on the other hand the expression

(+ 2
(/ 3 4))
reads as "the sum of two and the quotient of three and four". But since people are so used to infix notation they find it more intuitive and easier to read.


 No.989804

File (hide): 0a490467f0fdfd1⋯.png (16.76 KB, 550x646, 275:323, not point and click.png) (h) (u)

>>989435

>Point and click

The point is that it's not point and click.


 No.989841

How does TeXmacs compare with it's abstraction to TeX or LaTeX?


 No.990020>>990147 >>1005815

>>989391

PDF to those books?


 No.990147

>>990020

Not those books, but something like them:

http://chuzzlewit.co.uk/utp_book-1.1.pdf


 No.992219

There is groff + mom


 No.992744>>992812

Anyone have some good documentation on groff/groff_ms?


 No.992812>>992868

>>992744

Look at your post. Then angle your eyes a few cm upwards.


 No.992868>>992907

>>992812

mom is a macro package for groff not documentation


 No.992901

File (hide): 5a3215007b7c750⋯.png (67.9 KB, 1274x897, 98:69, latex linguistics.png) (h) (u)

Latex is a godsend for linguistics, it takes a bit to learn but as other anons have said the results are worth it.


 No.992907

>>992868

A few cm more...


 No.994316

>>989111 (OP)

i just wanna say im glad I came to this imageboard, 4chan is shit what thefuck it's only grug discussion everywhere. there's even a doctor who board and a containment board for ricing tards. this is too based


 No.994321

.do xflag 3

.so comment_macros

.TL

The best comment

.AU

Anon

.AI

8ch.net

.RT

.PP

\\comment{Heirloom-troff with some patches I've written is top comfy for writing stuff in humanities, idk how about more complex stuff, but surely more comfy then \\dobold{LaTeX}}.

.PP

You just have to write a Makefile with postprocessors and usually -ms macro set is just

.B

nice "."

.[

uriel smart book

.]


 No.994353

File (hide): db0a46aa2344f23⋯.png (2.29 KB, 35x38, 35:38, frame.png) (h) (u)

>>989111 (OP)

>>989137

I really wish Framemaker had caught on, especially during the early days of the web when its separation of content and presentation still could've been saved as a concept. Even today, it's by far the most powerful and elegant GUI for what so many TeXfags labor under either the tedium of handcoding or the broken mess that is every graphical frontend for LATeX.

>>989402

The software being discussed in this thread, document processors, isn't just "make pretty looking math", but the ability to automatically single-source large, complex documents using conditional logic. Word processors like LibreOffice are a different and strictly subsidiary type of software.


 No.994450>>994455

>>989111 (OP)

Groff


 No.994455>>994458

>>994450

You could have searched the thread to see that was already suggested multiple times before posting your worthless one word answer? Do you have anything else to contribute to the discussion?


 No.994458

>>994455

Do you?


 No.1002632

Bump


 No.1002660

>>989133

>not using eqn | troff


 No.1003287

File (hide): af1a3a004de93d9⋯.jpg (139.09 KB, 714x929, 714:929, a.jpg) (h) (u)

What kind of packages do I need to do something like this?

I really like the footer and header and all that.


 No.1005535>>1005591

Microsoft word is pretty good.


 No.1005591

>>1005535

this but meta-ironically


 No.1005815

>>989391

>>990020

genlib has "Document formatting and typesetting on the UNIX system [1st ed]" by Narain Gehani, doesn't have the second edition

"UNIX Text Processing System" by Kaara Christian isn't on genlib

troff Typesetting For UNIX Systems by Sandra L Emerson(?) and Karen Paulsell(?) isn't on there either

My guess is you should drop by IRC and see if anyone there is carrying a copy


 No.1005832>>1005864 >>1005886

>>989314

Weren't GNU pushing for info pages instead?


 No.1005841

Tried to get a Live Preview plugin for LaTeX bit it only updates when I manually save.

The chink who wrote it forgot to tell Vim to write to disk or something.


 No.1005842>>1005963

>>989314

>>989391

Looks like troff is completely unreadable a d that LaTeX improved it in almost every way.

More people should use both though.


 No.1005864

>>1005832

Info is for book-sized reference manuals, but manpages are still good for a short reference. Though GNU generally encourages to generate the manpage from the --help option. That is why the manpages look so ugly, they were not written by hand.

https://www.gnu.org/software/help2man/


 No.1005886>>1005926 >>1011433

>>989111 (OP)

how the fak did you learn LaTex? doesn't that take like a few months?

>>1005832

yes, because man wasn't muh free software.


 No.1005926>>1005944

>>1005886

>how the fak did you learn LaTex? doesn't that take like a few months?

Not him, but it doesn't. I learned it during a one-week workshop at university. You can also learn it by yourself in that time frame or even faster.

The key point is that learning LaTeX is very simple, but if you actually want to create your own packages you will have to learn TeX, and that shit is hard. But if you just want to use the standard classes you can pick up enough of the basics in one day. Usually though you will want to add packages for extra features, and those can get really complicated.

>yes, because man wasn't muh free software.

Manpages are really awful for long documents. It's called a manpage and not a manbook for a reason. Just look at the manpage of Rsync, it has almost 4000 lines of text. Info is much better for large manuals.


 No.1005944

>>1005926

> Info is much better for large manuals.

Agreed.


 No.1005963

>>1005842

There's a version of roff with a varient of the Knuth-Plass algorithm to add support for microtypography, although I find the syntax a bit unappealing: http://n-t-roff.github.io/heirloom/doctools.html


 No.1011078>>1011089

>>989182

Mathematica allows generating TeX files, doesn't it?


 No.1011089

>>1011078

It does, you'll probably want to copy equations at a time, rather than exporting an entire document.


 No.1011287>>1011294

is >>989134 better or synergistic with >>989114? I use only the latter, but I've been looking for an excuse to learn Lua.


 No.1011294>>1011652

>>1011287

>>989134

Holy shit LuaTex is cool, thanks anon!!!

Why do you even care about ConTeXt? LuaTex seems way cooler imo


 No.1011433>>1011652 >>1011657

>>989111 (OP)

I don't think syntax is that bad, but it's obviously geared towards being very powerful and letting you get it just right. This is the one strength of Latex imo, which no other software has eclipsed: If you can write it with pencil and paper, Latex can do it, while preserving semantics (as opposed to say scanning the the page). But if when you don't care much about how it looks the syntax is overkill. While supposedly you can just omit the directives and let it fall back to defaults, some Latex defaults are very stupid (fuckhueg margins on every page).

Latex pays off when you have very strict formatting requirements (academic paper), need to write math and don't want to end up stuck because it happens to not support a certain symbol or expression you want, when you want the output to look very nice (resume) and when it's a complex document that will be edited a lot (paper, resume). In other situations it's not really worth it unless you've been using it so much you have godlike syntax knowledge.

My approach to notes is usually like this

>notes.txt

>if needed, add #, *, - and save as notes.md

>if needs light to moderate math add $$

>if needs to look not shit on paper pandoc -o notes.tex notes.md and compile

>if needs to look NICEEE go to notes.tex preamble and also preemptively load tex.stackexchange.com

>>989123

This, markdown, pandoc, latex combo is a godsend. Especially with being able to export into docx and html so you can transparently interface with brainlets.

>>989134

So I use latex through texstudio with tex-live/miktex currently. How easy is it to migrate to luatex? Will any Latex snippets I find online just werk? Is it compatible with all of CTAN? These alternate texes always seemed like special snowflakes that are more trouble than they're worth to me, but you make it sound nice.

>>989182

Mathematica is nice but with the caveats you say. I accomplish a similar WYSIWYG thing by using a markdown editor that can automatically render latex math. Typora is my favorite, but there are many open source ones as well.

>>1005886

I literally just followed the wikibook on latex for a few chapters, and the basic syntax clicked in a few days. I also unwittingly knew the equation typesetting from sites that use markdown. Making a simple document in Latex is actually very easy and you need very little syntax. But getting the environment set up so you can compile at all may be confusing because of how archaic it is. Then you need some motivation to stick with it and google things to learn your way around slightly more advanced stuff like tables, images and macros. I still often copypaste snippets I find online and from my old latex files, learning all of latex syntax is a bit of a waste because you rarely use most of it.


 No.1011652>>1011703 >>1011707

>>1011294

> Why do you even care about ConTeXt? LuaTex seems way cooler imo

The design and goals of LuaTeX and ConTeXt are completely independent of each other. I'll try to explain it with a drawing:

LaTeX    ConTeXt        LuaLaTeX    ConTeXt (Mk IV)
^ ^ ^ ^
| | | |
| | | |
+----+----+ +------+----+
| |
| |
TeX -------------------> LuaTeX

LaTeX and ConTeXt Mk II are implemented on top to TeX, which is a typesetting systems. LuaTeX is a re-implementation of TeX with Lua scripting, and on top of it are LuaLaTeX and ConTeXt Mk IV. LuaTeX aims to expose the mechanisms of TeX in Lua, and ConTeXt aims to make typesetting more accessible. You can have botho of those independent of each other.

>>1011433

> So I use latex through texstudio with tex-live/miktex currently.

I don't know about any of the dedicated editors, I use Vim for writing and a (short) makefile for automating the compilation.

>>1011433

> How easy is it to migrate to luatex? Will any Latex snippets I find online just werk? Is it compatible with all of CTAN?

Probably yes to all of those. I use LuaLaTeX mostly for its unicode capabilities and everything worked just fine. Only doing custom fonts requires different packages, but I prefer to stick with the default fonts.

>>1011433

> These alternate texes always seemed like special snowflakes that are more trouble than they're worth to me, but you make it sound nice.

What alternate TeXes? There is really just pdfTeX (which you are most likely using), XeTeX (which is mostly superseded by LuaTeX) and of course LuaTeX. LuaTeX is like a successor to all previous TeX implementations. I think LuaTeX is the future, unless someone comes up with something better of course.


 No.1011657

>>1011433

> Then you need some motivation to stick with it and google things to learn your way around slightly more advanced stuff like tables, images and macros. I still often copypaste snippets I find online and from my old latex files, learning all of latex syntax is a bit of a waste because you rarely use most of it.

You should have a look at the unofficial LaTeX manual. There is a PDF and HTML version, but I personally prefer the info version. That way I can read and browse the info manual in Vim (or Emacs of course) right next to my source code and copy-paste parts without switching applications.

https://ctan.org/pkg/latex2e-help-texinfo?lang=en


 No.1011703>>1011707 >>1011877

>>1011652

Thanks for the drawing and explanation. So is it recommended to use LuaTeX directly (assuming a familiarity with programming), or are the front ends basically required? I think I'd rather use LuaTeX directly and get nasty with the internals.

Just asking because I do have to prepare a series of professional documents this week, and it'd be a great excuse to simultaneously learn Lua (which gives me much more zest for mundane paperwork).


 No.1011707>>1011877

>>1011652

Nevermind >>1011703, I get it (I think). I don't want to typeset in LuaTeX for the same I don't want to do it TeX, but I'm definitely moving to ConTeXt (Mk IV) for this upcoming set of professional documents I have to create.


 No.1011877>>1011987

>>1011707

>>1011703

I don't know what the requirements for professional papers look like, I have only been using LaTeX (LuaLaTeX to be precise) for my Master's thesis, keeping lecture notes and writing some letters where I didn't feel like fiddling with Libre Office.

You won't need to use Lua unless you really want to get involved with the gritty details of typesetting (i.e. not the content itself but the appearance). LaTeX and ConTeXt are basically frameworks built on top of TeX; you could build your own framework on top of TeX, but that requires you to be very familiar with typesetting. For my purposes LaTeX has always been good enough.

> but I'm definitely moving to ConTeXt (Mk IV) for this upcoming set of professional documents I have to create.

Cool, let us know how it works out for you. I have never used ConTeXt myself, but it looks like a really interesting project. I have wanted to learn it, but in the end time forces me to just stick with LaTeX instead.


 No.1011987>>1012010

>>1011877

Why are you “forced” to use Latex?


 No.1012010>>1012259

>>1011987

Do you mean why I am forced to use the TeX family in general, or why am I forced to used LaTeX specifically. For the former, I have to write lots of math, and once you have the process down there is nothing that's as capable as the TeX family.

As for why LaTeX specifically, it comes down to lack of time to learn something new, and the fact that there is usually a package that's good enough for everything. Usually when I write I have to produce output in a reasonable amount of time (like my thesis), I don't have the time to learn something new first. There is no deeper reason.


 No.1012259>>1012328

>>1012010

It was the latter, thanks for writing both (much more efficient). I’ve also learned LaTeX for the same reasons, but have no thesis to write (just some shitty business docs) and am ok with killing some time using ConTeXt. I was worried there might be some sort incompatibility in academics if I used something other than LaTeX (other than socially, i.e can’t really use ConTeXt in collaboration).


 No.1012285>>1012291

File (hide): 99df49d3108a9fc⋯.png (58.27 KB, 400x240, 5:3, sample-doc-hardcover.png) (h) (u)

I've been using groff a lot, and I like it for literature, although the documentation is sparse, especially because of the wide range of macrosets.

Rightnow I've been using mom, its really well supported, although ms is a vlose second(with html lexer support).

Please givegroff_mom ,or groff_ms a try. Its so pleasant to usr and you'll learn it in a fraction of the time it took to compile LaTeX. I'll post a few caps and sourcecode later today if you want.

Pic related was generated using groff_mom


 No.1012291>>1012416

>>1012285

>fraction of the time it took to compile LaTeX

I'm guessing you don't mean rendering the pdf in LaTeX, rather compiling LaTX itself. Because when the document is rendered every two minutes or takes too long, you didn't use the features right (e.g. dividing the document, so not everything gets rerendered, etc.).

I'd say LaTeX gets more "worth it" the longer you use it.


 No.1012328>>1012429

>>1012259

> I was worried there might be some sort incompatibility in academics if I used something other than LaTeX (other than socially, i.e can’t really use ConTeXt in collaboration).

Well, if you are given a LaTeX template you will have to use LaTeX obviously. The same if people expect you to hand in LaTeX source so they can copy-paste it into their own textbase. And if you are working in a team where no one knows ConTeXt you will also either have to train everyone or just accept LaTeX. And of course it works the other way around as well, if everyone is using ConTeXt and you come in with LaTeX you will be laughed out of the building. It's really not the fault of ConTeXt or LaTeX, it's simply how working in a group works.


 No.1012416

>>1012291

No I mean that groff is easier to learn than compiling your first LaTeX. The fact that it's faster is secondary.


 No.1012429

>>1012328

>working in groups

I imagine LaTeX is vastly more popular than ConTeXt though, no? And yeah at any rate, at least I will know them both if I learn ConTeXt. I think I’ll prefer it too.


 No.1018020>>1018024

>>989111 (OP)

unironically just write everything in plaintext. That's how it was before word processors replaced typewriters and we did just fine.


 No.1018024

File (hide): 8a266b964d38a10⋯.jpg (70.18 KB, 639x407, 639:407, selectric_changing.jpg) (h) (u)

>>1018020

Right, because typesetting mathematical documents was so much easier with a typewriter. Why don't you just suggest we go back to movable type letter pressing?


 No.1025297>>1025332 >>1025380 >>1025404

I decided to use LaTex for a bunch of reports I had to write. It was a fucking nightmare. Easily multiple times longer than using any office software.

>using images

enjoy manually resizing everything and still getting ass results

>tables

might as well use paint

>having nice word breaks

ser-

iously

The results don't even look any better than modern office software can produce. Total waste of time.


 No.1025332>>1025477

>>1025297

I use \includegraphics[width=0.6\textwidtg] which works nicely. Not easy to decide what % of text width you want the image to be.

Tables are a drawback of LaTeX. I use booktabs which looks nicer. You should use a tool like tablegenerator.com or the wizard in TexStudio.

What's wrong with the word breaks?

Generally, yes, some things are easier in MS Word. If those things are what you want to do, why bother with Latex? The point of using it is to do things that are hard in other prorams.


 No.1025380>>1025477

>>1025297

>resizing

nah you just set the width to the textwidth or whatever

>tables

I haeve never see one

>word breaks

fine by me.

LaTeX isn't useful for trying to make it pretty, it's easy to have it pretty by default since it has sane defaults.


 No.1025404>>1025409 >>1025477

>>1025297

> images

\includefigure{faggot}[width=\linewidth]

> tables

\usepackage{booktabs}

\toprule

...

\midrule

...

\bottomrule

> having nice word breaks

\usepackage{babel}[en]

Use a premade style like revtex41 and you will be fine fine. LaTeX has a steep learning curve yes, but eventually you will be 100x more efficient.


 No.1025409>>1025414

>>1025404

>Use a premade style

This is why Latex is so fucking cancerous, everyone just piles on styles and styles and styles that look nice while having no idea how they actually work. When you think about it, it's actually worse than the NPM garbage, because at least there's built in version checking. Latex niggers just find a random version of some package they like and never, ever update it again.


 No.1025414

>>1025409

To be honest, I only use "academic" formats like revtex and other conference templates, these are typically well-documented https://d22izw7byeupn1.cloudfront.net/files/revtex/auguide4-1.pdf. Comparing these against npm garbage would be a bit dishonest, but in general I agree with you. Some of these are just a mess. I think an other interesting parallel with the JS world is in the Stackoverflow answers. Some answers are just completely wrong.


 No.1025477>>1025510

>>1025404

>width=0.6\textwidth

thanks man I never even knew about that

>>1025380

works for you? Shit I guess it is good then.

>LaTeX isn't useful for trying to make it pretty, it's easy to have it pretty by default since it has sane defaults.

Yeah don't try to make it pretty but have it pretty by default. Now I get how it works.

>>1025332

>width=\linewidth

mind blowing stuff where did you come up with this.

>\usepackage{babel}[en]

holy shit latex is based. Now words run over the right hand margin just like I always wanted.

>you will be 100x more efficient

yeah right.


 No.1025510>>1025519

>>1025477

> Now words run over the right hand margin just like I always wanted.

Nigger, Babel does not do that https://ctan.org/pkg/babel?lang=en.

Enjoy Office pajeet.


 No.1025519

>>1025510

>no u

lol how about no u

Enjoy abandonware.




[Return][Go to top][Catalog][Screencap][Nerve Center][Cancer][Update] ( Scroll to new posts) ( Auto) 5
96 replies | 14 images | Page ?
[Post a Reply]
[ / / / / / / / / / / / / / ] [ dir / animu / ausneets / dempart / doomer / s / strek / tingles / vg ][ watchlist ]