What's new in TeX, part 1

Posted Sep 18, 2015 5:42 UTC (Fri) by rsidd (subscriber, #2582)
Parent article: What's new in TeX, part 1

I still use (pdf)LaTeX but partly that's because I've been using it since the 1990s. The newer stuff in luatex and xetex are still uncharted territory for me. Out of curiosity, how many people here have become latex converts since the mid-2000s, and how many would use it for complex formatting and graphics rather than just go with scribus, Adobe Pagemaker and suchlike? Also, the unicode equations are fascinating, but what's a good editor on linux that makes entering unicode math symbols easy? There's no question the source is more readable this way, but entering familiar LaTeX code still seems faster unless there's a really good unicode editor around.

What's new in TeX, part 1

Posted Sep 18, 2015 7:34 UTC (Fri) by jezuch (subscriber, #52988) [Link]

> Out of curiosity, how many people here have become latex converts since the mid-2000s, and how many would use it for complex formatting and graphics rather than just go with scribus, Adobe Pagemaker and suchlike?

Well, I'm a lazy bastard so I went with LyX :)

What's new in TeX, part 1

Posted Sep 18, 2015 14:10 UTC (Fri) by jsanders (subscriber, #69784) [Link] (1 responses)

I'm still on pdflatex, too. I wonder whether publishers know how to deal with lualatex, etc? One of the main journals in my field still requires figures in postscript format, so even pdflatex is too modern.

What's new in TeX, part 1

Posted Sep 18, 2015 16:33 UTC (Fri) by rsidd (subscriber, #2582) [Link]

Ah, don't get me started on the publishers. Using their style files is an exercise in masochism. Many (I'm thinking of certain OUP journals for example) developed their style files on particular Windows-specific proprietary releases of TeX, and never bothered to test on any other system. Using their styles/templates is such a nightmare.

What's new in TeX, part 1

Posted Sep 18, 2015 16:15 UTC (Fri) by leephillips (subscriber, #100450) [Link] (3 responses)

I use Vim for everything - it has a "digraph" system that provides mnemonic abbreviations for quick entry of some Unicode characters. And, of course, you can define your own shortcuts or macros for anything.

What's new in TeX, part 1

Posted Sep 18, 2015 16:35 UTC (Fri) by rsidd (subscriber, #2582) [Link] (2 responses)

I use vi(m) for quick and dirty things, emacs for latex and programming... but given that I still have residual vim finger-memory, maybe I should try switching back... emacs too does seem to have unicode support, but not particularly faster to type than latex math code.

What's new in TeX, part 1

Posted Sep 25, 2015 17:06 UTC (Fri) by Wol (subscriber, #4433) [Link] (1 responses)

Seeing as emacs is the kitchen sink, surely it has a vi emulation mode? :-)

gd&r

Cheers,
Wol

What's new in TeX, part 1

Posted Sep 26, 2015 5:54 UTC (Sat) by jem (subscriber, #24231) [Link]

Well, Emacs has not only one vi emulation mode, it has three.

As an Emacs fan it bothers me that Emacs is accused of being bloated. "Kitchen sink", and all that. Unlike some other editors, Emacs is modular. Extensions are loaded on demand. If you are not interested in vi emulation, or have no need for Python mode, just pretend they are not there.

What's new in TeX, part 1

Posted Sep 18, 2015 16:58 UTC (Fri) by lsl (subscriber, #86508) [Link] (2 responses)

> Also, the unicode equations are fascinating, but what's a good editor on linux that makes entering unicode math symbols easy?

A well-written, mnemonic XCompose file goes a long way for that and the result is usable in all X programs. For example:

<Compose> NN: ℕ (double-struck capital n, works for other letters too)
<Compose> mo: ∈ (element/member of)
<Compose> pd: ∂ (partial differential)
<Compose> fa: ∀ (for all)
<Compose> te: ∃ (you can guess it I think)

Then there's prefixes for reaching other alphabets, like * for greek, so that *p is π and *P is Π, with Σ being *S. The latter being distinct from the summation operator ∑ (su). Add a bunch of logical operators, set operators and more obvious things like ≤, ± or ≠.

The initial problem is coming up with one that works for you. I started with Plan 9's /lib/keyboard, as I knew it already and it has nice mnemonics for math symbols (all things mentioned above come from there). Also I think the format is easier to edit than XCompose. Plan 9 from User Space (the Unix port of the Plan 9 tools) has a program (mklatinkbd) to convert from Plan 9 keyboard files to XCompose ones.

What's new in TeX, part 1

Posted Sep 18, 2015 17:44 UTC (Fri) by rsidd (subscriber, #2582) [Link] (1 responses)

Thanks for that! I had no idea.

What's new in TeX, part 1

Posted Sep 18, 2015 18:12 UTC (Fri) by lsl (subscriber, #86508) [Link]

If you want to try the Plan 9 mappings, the keyboard(7) manpage of plan9port describes some of the rules underlying the chosen sequences (like the * prefix for greek), how to convert the keyboard file to the XCompose format and how to get rebellious GTK+ programs to abide by it (in short: set GTK_IM_MODULE=xim in your environment).

https://swtch.com/plan9port/man/man7/keyboard.html

The manpage could use an update: it still talks of 16 bit runes/codepoints. The actual code was updated to use 21 bit runes, though.

For plan9port itself see https://swtch.com/plan9port. Note that it recently migrated to Github, so you should ignore all references to Mercurial and/or Google Code and instead use https://github.com/9fans/plan9port (or maybe your distro-packaged version, if it exists).

What's new in TeX, part 1

Posted Dec 30, 2020 23:16 UTC (Wed) by ceplm (subscriber, #41334) [Link]

My wife switched to LaTeX sometime in the mid-2000s. Yes, it is mostly about specific requirements: she is a linguist and their linguistic trees are just such pain to do in Word, that LaTeX was the only option for her. Also, even for my sociology thesis, LaTeX was the best option. There was not that much math, but working with large bibliographies and citations, was just such pain in Word/LibreOffice (then OpenOffice). Perhaps Zotero now makes life easier even for the Word users, but I have never seen the reason to switch back (although, now I write mostly in rST).

What's new in TeX, part 1

Posted Dec 31, 2020 15:00 UTC (Thu) by jem (subscriber, #24231) [Link]

>Also, the unicode equations are fascinating, but what's a good editor on linux that makes entering unicode math symbols easy?

An online alternative is mathcha.io. It is a simple word processing application, with a wysiwyg math editor that is easy to learn and productive once you have got the hang of it. I wouldn't recommend writing a large document using only Mathcha, but you can export the selection to LaTeX, with the following disclaimer:

"Note: Latex below is just for reference, it does not guarantee to be full compatible (or compiled) in Latex Document."

A Mathcha document can also be saved online (if you log in with a Google, Facebook, Twitter, or Github account), or saved locally as a zipped HTML or (proprietary) Mathcha file. Links to online documents can easily be shared. I have mainly used Mathcha to help a friend do his math and physics homework by sharing short documents.

Entering Unicode math symbols easily

Posted Dec 31, 2020 17:18 UTC (Thu) by peniblec (subscriber, #111147) [Link] (2 responses)

I know that this comment is 5 years old now, but since it has been resurrected by jem, and I figure a fair share of LWN readers might be Emacs users:

Also, the unicode equations are fascinating, but what's a good editor on linux that makes entering unicode math symbols easy? There's no question the source is more readable this way, but entering familiar LaTeX code still seems faster unless there's a really good unicode editor around.

Emacs has many ways to enter arbitrary Unicode symbols:

C-x 8 RET prompts for a character name (or its hex codepoint): I frequently use it for emojis; e.g.
- C-x 8 RET face TAB
- C-x 8 RET thumbs TAB
C-x 8 C-h enumerates a bunch of shorcuts for specific characters; e.g.
- C-x 8 < yields «
- C-x 8 a > yields →
Emacs 28's new "iso-transl" input method (C-\ iso-transl RET) allows the user to type those characters without the C-x 8 prefix;
it's fairly easy to define your own input methods for symbols you frequently use;
TeX symbols already have their own input method (C-\ tex RET); I frequently use it for greek characters; e.g.
- \alpha yields α
- \oint yields ∮
- \nabla yields ∇

Entering Unicode math symbols easily

Posted Dec 31, 2020 19:19 UTC (Thu) by jem (subscriber, #24231) [Link] (1 responses)

>I know that this comment is 5 years old now, but since it has been resurrected by jem

I realized the article was old only after posting my reply. I wouldn't have replied if I had known this. But it wasn't me who dug up the ancient article, it showed up in my unread comments because of user cpelm commented on it yesterday.

Entering Unicode math symbols easily

Posted Dec 31, 2020 23:24 UTC (Thu) by peniblec (subscriber, #111147) [Link]

Apologies! I must have glossed over ceplm's comment yesterday; I got nerd-snipped today by the text you quoted :)