Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

I hate LaTeX, I love LaTeX

commutative.xyz

Post reply on HN
Print view View original

151–160 of 245 posts

Re: I hate LaTeX, I love LaTeX

#151

Post by dhosek »

moffkalast wrote: post #143
amelius wrote: post #106

Earlier quoted context omitted.

My main issue with LaTex is that you can't take a perfectly correct piece of markup, insert it into a container like a table, and expect it to work. LaTeX makes it difficult to compose stuff. Even HTML did this better.

Not to mention that when it fails to compile, at least in Overleaf you get an error log that is hopelessly useless and the only thing you can do is just remove your changes until it works again. LaTeX is a cool idea, but the implementation? Absolutely horrid.

There are two big challenges: One is that TeX, the underlying executable behind LaTeX, is pushed to its limits¹ to run. The macro expansion paradigm behind LaTeX made sense in 1982 but there are a lot of decisions underlying the executable and TeX-the-language that were dictated by the computing technology of the 1970s (forget Unicode, standard ASCII wasn’t universally available when Knuth wrote TeX).

The other problem was the questionable decision by the LaTeX core team to, rather than release LaTeX3, they instead decided to evolve LaTeX2e which means that they are encumbered by backwards compatibility issues.

The noisy output from LaTeX is also not so great. One of the things I’m doing with finl is (a) assuming that there won’t be an attempt to fix errors mid-run (one of those 1982-era decisions that no longer makes sense) so a comprehensive list of errors can be presented to the user (or revealed programmatically) and (2) Adding indications in the output (which can be disabled) of errors in the input. I’m debating between errors being PDF annotations so users would click on an error or warning icon in the PDF to see the message and indication of where the issue was found, or just making it a big blot of overprinting (so many times on tex.stackexchange there are so many issues where a user—usually on Overleaf—ignored the error messages being output.

⸻

1. Actually beyond its limits. LaTeX will no longer run on a Knuth TeX executable but requires eTeX extensions to run.

Re: I hate LaTeX, I love LaTeX

#152

Post by ska »

Tepix wrote: post #89

In the last three decades of having the opportunity to learn and use TeX, i have managed to avoid it. In the same time i've helped a friend with word processing for her dissertation. I'm still not sure how we managed to get it to work with all the horror stories about data loss and Microsoft Word at the time. But it did its job and I'm still in the WYSIWYG camp. Maybe in another life...

People write dissertations in Word all the time. How painful it is depends a bit on how complicated the dissertation is as a document.

This is of course only an anecdote, but I know many people who started in word, gave up on it and moved to Tex/Latex, but nobody who went the other way (that is to say, I know people who tried to give up on Tex, but eventually went back - not anyone who actually published with it). This might have quite a bit to do with subject matter - Word has always been an awkward fit for mathematics.

Re: I hate LaTeX, I love LaTeX

#153

Post by elmolino89 »

marcosdumay wrote: post #104
wesleywt wrote: post #78

Having written my thesis in LaTex, I can say it is wonderful and a wonderful waste of time.

Having written a report in Word that was about 1/4 of the size of my dissertation, but had some of the same content, and then the dissertation in LaTeX, I can say that as much as LaTeX wastes your time, it's by far the least wasteful option.

I did typesetting in Lyx/LaTeX of two PhD theses: my own (eng) and my GF's (esp) a decade+ later. In both cases the adventure started with admitted a vanilla Word doc without all the bells and whistles (= no templates, no version control).

Once you enter the area of strange formatting, chars typed by who knows either a cat walking on the keyboard or sleep deprived student typing madly, fixing these in a binary format is a real PITA. Lyx/LaTex text formats are trivial to grep, diff, etc.

Today I would probably start and try to stay as long as possible in Markdown, then reformat it at the very end.

Re: I hate LaTeX, I love LaTeX

#154

Post by ska »

mhh__ wrote: post #3

LaTeX needs replacing with something faster and more program-y, and that isn't just a python script. LateX's successor should be opinionated, safe, easy to build in parallel, and also good at integrating with other tooling: Let a CAS generate the next equation for you, for example. Make sure that figure is never out of date (e.g. a build system rather than relying on a web of files and button clicks) Etc. This will l…

> Make sure that figure is never out of date

For anyone with a software background, this is pretty straightforward - once you start thinking of the latex file(s) as source code rather than the document, it's a short jump to setting up a build environment...

Re: I hate LaTeX, I love LaTeX

#155

Post by dhosek »

cruegge wrote: post #72
camel-cdr wrote: post #42

I'm always amazed how people could bootstrap something as feature rich and versatile as LaTeX from plain TeX. I'd really like to get to know plain TeX a bit better, but I've got no idea where to start.

If found “TeX by Topic” a great introduction, it's less overwhelming than the TeX book. Also really helpful in wrapping your head around macro expansion is the `texdef` cli tool, which allows you to quickly evaluate a line of macro code. Finally, https://www.tug.org/utilities/plain/cseq.html is nice for reference.

There are three books that are worth looking at to get plain TeX. I’d start with Viktor Eijkhout’s TeX by Topic (which I think is free in PDF on CTAN). Then the TeXbook to really get everything (the PDFs that float around the net are illegal. Please buy your copy. DEK donated his royalties to TUG so your purchase will help support the continuation of things like CTAN and the LaTeX project). And finally, if you really want to dig deep, Stephan v. Bechtolsheim’s four-volume magnum opus, TeX in Practice will tell you more than you ever wanted to know.

Re: I hate LaTeX, I love LaTeX

#156

Post by elmolino89 »

plopilop wrote: post #139
kaba0 wrote: post #131

Earlier quoted context omitted.

Strong agree. My fiance and I wrote our thesis simultaneously, she in Word, while I used LaTeX. The amount of stupid figure- and page- rearrangements she suffered because she removed half a sentence from the middle, while LaTeX generated sane output with basically any text seriously made me much more productive.

Word has had automatic reference numbering for a while now (see for instance [0]). Not denying that it's not very known (nor pleasant to use), and I'm not even talking about bibliography. I made my partner switch to Latex thanks to the bibliography. Writing tables almost made her switch back to Word. [0]: https://www.sfu.ca/~ljilja/cnl/info/UseCrossReference/index....

Among fellow students countless hours were lost playing with spaces, line breaks, font sizes just to get tables/figures stay or move to the next page. This did look like an endless task: adding few chars in a chapter could have messed up positioning of bunch of figs.

Re: I hate LaTeX, I love LaTeX

#157

Post by ska »

ProfXponent wrote: post #11

What is the intended use for LaTex these days? I’m not trying to flame, I just have no idea why anyone would choose to use it over html/css or just a word processor.

> What is the intended use for LaTex these days?

Writing anything with complicated mathematical typesetting, for one; their have been improvements in word processors and thinks like MathML/MathJax, but they are still behind.

Works pretty well for autogenerated docs with more going on than is easy in Markdown, also - but that's often a bit of a wash.

Re: I hate LaTeX, I love LaTeX

#158

Post by dhosek »

clircle wrote: post #4
mhh__ wrote: post #3

LaTeX needs replacing with something faster and more program-y, and that isn't just a python script. LateX's successor should be opinionated, safe, easy to build in parallel, and also good at integrating with other tooling: Let a CAS generate the next equation for you, for example. Make sure that figure is never out of date (e.g. a build system rather than relying on a web of files and button clicks) Etc. This will l…

What is LuaLaTeX?

LuaLaTeX is a lot slower than pdfLaTeX or XeLaTeX and default settings give different output than the other two, e.g., around line breaks at – or —. I only recommend it if you need it. Most people don't.

Re: I hate LaTeX, I love LaTeX

#159

Post by samatman »

netizen-936824 wrote: post #135
samatman wrote: post #107

Earlier quoted context omitted.

The poster didn't say "hands", they said "your hands". Your human hands, which co-evolved with throwing things, we have abundant paleontological evidence of this and you may read as many papers on the subject as you would care to.

Got links to any of those papers?

Are you vegan? Just curious.

Re: I hate LaTeX, I love LaTeX

#160

Post by krinchan »

ProfXponent wrote: post #11

What is the intended use for LaTex these days? I’m not trying to flame, I just have no idea why anyone would choose to use it over html/css or just a word processor.

LaTeX is particularly suited to print (and by extension PDF). When you get into writing your own custom LaTeX macros/commands, yes it's difficult. However, the vast majority of users are simply consuming these sorts of custom commands in the article, not producing them. This leads into environments where people writing papers or articles just use what's handed to them.

LaTeX works out amazingly for two very specific use cases: 1) Large, collaborative documents and 2) producing the exact same, highly structured print document over and over. I have pretty extensive experience with the second use case.

When I was Secretary for a non-profit organization, I had an entire little personal library of macros that sat on top of the minutes[1] package. I could open an empty text file, type in \attendee{Name} as people arrived. Type in \maketitle and do a quick preview and the document would instantly build out a front page that included if we had quorum or not, etc. I typed the document exactly as the meeting happened while marking up things like tasks (and who they're assigned to), decisions made (with voting tallies and who motioned what), and so on.

All that data would get duplicated to the end of the document in nice lists for quick reference. It had consistent formatting. The build system (a make file and latexmk iirc) would wrap my new file in the basic boilerplate and spit out the PDF for me to email. I consistently had the meeting minutes published an hour after every meeting whereas my predecessors often took 24 hours or more to publish meeting minutes due to Word being Word. Those documents also lacked a table of contents or the various appendices with summary data.

That said, there were things that I tended to avoid doing, like complex tables or something. There's something to be said for staying inside the box of what comes built into LaTeX. I recall giving up with floating an image to the left of a table on our title page and just plopping the image inline in a horizontally centered paragraph and then putting the rest of the title content beneath it.

Also, many documentation frameworks (sphinx comes to mind) use LaTeX under the covers to produce their PDF output. So you might think you're just using markdown or RST, but somewhere someone is converting that to LaTex or TeX in order to leverage a very good typesetting engine.

[1]: https://www.ctan.org/pkg/minutes

Post reply on HN