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

Ask HN: Did TeXmacs' doubly misleading name hold it back for 20 years?

news.ycombinator.com

Post reply on HN
Print view View original

21–30 of 61 posts

Re: Ask HN: Did TeXmacs' doubly misleading name hold it back for 20 years?

#21

Post by wosined »

davrosthedalek wrote: post #9

One could argue that it is held back by not being based on TeX (depends on how good the export is) and emacs (probably).

The exporter is very good. There is also an importer, which can import latex. It can import almost all standard latex macros one to one.

You can convert org-mode files into texmacs files by exporting to latex and then importing the latex into texmacs.

Emacs cannot render latex with such quality even with image previews. Trust me, I'm using Emacs.

The typesetting algorithm is based on TeX, but it is a custom rewrite with tweaks afaik.

There are interfaces to python, maxima, octave, matlab, asymptote, graphviz, sympy etc. that let you type the code into the buffer and see the graphical output of those programs, i.e. figures. Similar to a notebook. Texmacs can typeset the symbolic equations that you get from maxima, so you don't have to see the source code with code syntax but the result a rendered equation. Does emacs have a repl that returns rendered equations?

It is also possible to use the python, maxima, etc. plugins to evaluate expressions in the body of your text in a texmacs file.

You can create presentations with animations with interactive code sessions and typeset all the math in the presentation with the algorithm derived from TeX.

The main implementation uses gnu guile as an extension language, thus if you like Emacs lisp you will get along with texmacs too.

You can link texmacs files together and create a wiki structure (think org-mode links etc.) and you can export this structure into HTML. This allows you to write a website in texmacs and then export it into HTML with your custom css.

You can typesetting algorithms in the same manner as with the algorithm* packages of latex, but with realtime rendering without having to recompile or guess the layout.

Texmacs also offers Emacs like keybindings so that you can open files with C-x C-f, move the cursor with C-f, C-b, C-n, C-p and other Emacs like bindings.

Re: Ask HN: Did TeXmacs' doubly misleading name hold it back for 20 years?

#22

Post by mgubi »

II2II wrote: post #17

It's really hard to say. Producing documents for publication is a big part of the reason for using software like TeX, both because of its capabilities and requirements of the publisher. Given the audience of TeXmacs appears similar to that of TeX, in terms of capabilities but not the requirements of publishers, it was not terribly appealing to a wide audience and a significant factor holding it back for its specializ…

TeX is a typesetting language. Like Postscript. There is no particular reasons humans need to write it. That is the reason LaTeX was invented (and Context). We need formats, like PDF or HTML, to exchange and reliably use informations and describe documents. TeXmacs is a document preparation system from the UI to all the other things needed, in one place, in a uniform architecture. It has its own internal format and it exports to TeX/LaTeX as it can export to HTML or PDF. Publishers do not need TeX to store documents, indeed TeX is no a particularly good way to store data. It is what it is: a typesetting language, not a document description format. So I do not see any conflicts between using TeXmacs and then sending a TeX file to the editor. Anyway in most of the cases them (or nowadays most likely you) need to reformat it to the standards of the particular journal. I (and many others, including the main author of TeXmacs) work in academia and I have no problem using TeXmacs everyday to do my work and then send to the publisher a nice LaTeX file, which I do not have to edit by hand. I'm (mostly) TeX/LaTeX free since 2006 and very happy about this.

Re: Ask HN: Did TeXmacs' doubly misleading name hold it back for 20 years?

#23

Post by GiovanniP »

amichail wrote: post #19
II2II wrote: post #17

It's really hard to say. Producing documents for publication is a big part of the reason for using software like TeX, both because of its capabilities and requirements of the publisher. Given the audience of TeXmacs appears similar to that of TeX, in terms of capabilities but not the requirements of publishers, it was not terribly appealing to a wide audience and a significant factor holding it back for its specializ…

What about for producing documents not for publication such as writing up assignments, lecture notes, personal/research notes, etc.?

And slide presentations.

Re: Ask HN: Did TeXmacs' doubly misleading name hold it back for 20 years?

#26

Post by mgubi »

zitterbewegung wrote: post #18

When I used to use TeXmacs I would have annoying glitches to the point where I just called back to TeX with regular Emacs. I think its general clunkyness really held it back. The problem now is that people still use regular Emacs and share the TeX file locally or use a service like overleaf that allows for collaboration online. TeXmacs in my opinion tried to solve a problem for a small subset of users that it didn't…

I'm not sure what are you referring to. TeXmacs is used everyday by many people, at lest 3/4 of my students wrote their PhD thesis with it, some of them enterely, other the initial draft. All my papers are written with it, lecture notes, presentations, letters. My experience is the same of many other people. Even comparning TeXmacs with AucTeX+Emacs is misleading. Totally different user experience. TeXmacs does not have yet a fully stable collaborative functionality, but the functionality is there and it experimentally work. Maybe one day will be completed and you will not need to rely on a central server to use it, just fire up a TeXmacs instance to be the server and connect the others: https://mobile.twitter.com/gnu_texmacs/status/13910657725338... So, slowly, but we will be there too. :)

Re: Ask HN: Did TeXmacs' doubly misleading name hold it back for 20 years?

#27

Post by wosined »

hprotagonist wrote: post #11
Freak_NL wrote: post #10

The name doesn't help, but while it may produce output out of the box that looks as good as LaTeX, the website also doesn't directly answer another question which matters to people who need to do typesetting things with it that have been solved by others: the huge package ecosystem of LaTeX (and the batteries-included distributions like TeX Live).

Bingo. I usually need to lean relatively hard on CTAN, and if you’ve got a niche tool with better UX but no widespread adoption in a very entrenched user base, you’re not going to take over. when i want “texmacs”, i use the other GNU tools: https://www.gnu.org/software/auctex/ , emacs, and pdf-tools.

Auctex, emacs and pdf-tools are good tools. I still use them. But you either compile LaTeX often, which gets very tedious once the compile time gets longer for larger documents, or you can get a confusing error message from LaTeX that takes you half an hour to deciper. That is my experience. Texmacs does the job of all of those tools better, except for Emacs, because Texmacs is not yet as big as Emacs, but it has more potential.

Re: Ask HN: Did TeXmacs' doubly misleading name hold it back for 20 years?

#28

Post by kzrdude »

Things that are great usually come through anyway. Is that too optimistic? If you think Texmacs is great, maybe it also has come over a threshold to greatness - it wasn't there for all of the past 20 years(?) Things such as your outreach in this post is what's needed for a great program to become more well known.

My spontaneous feeling is that it's very upsetting that it has TeX in the name but does not use Tex or Latex.

Re: Ask HN: Did TeXmacs' doubly misleading name hold it back for 20 years?

#29

Post by mgubi »

Freak_NL wrote: post #10

The name doesn't help, but while it may produce output out of the box that looks as good as LaTeX, the website also doesn't directly answer another question which matters to people who need to do typesetting things with it that have been solved by others: the huge package ecosystem of LaTeX (and the batteries-included distributions like TeX Live).

This is not our experience as TeXmacs users. 90% of the time you are always using the same features, which are well implemented in TeXmacs. Maybe you can show me some of your documents which require TeX, so that I can understand the uses cases you think about. If you look at documents on arXiv there is no reason not to use it. Of course TeXmacs will never take over but it allows us to focus on the content and not on writing TeX. Time gained for more productive tasks with no backside. I guess people praising LaTeX should give explicit examples of what they mean by "huge package ecosystem", in the sense of a measure of how many of those packages are really used by 80% of the scientific community. And btw, TeXmacs can just run TikZ fine, if you really need it. https://twitter.com/gnu_texmacs/status/1509200571537973255

Re: Ask HN: Did TeXmacs' doubly misleading name hold it back for 20 years?

#30

Post by wosined »

kzrdude wrote: post #28

Things that are great usually come through anyway. Is that too optimistic? If you think Texmacs is great, maybe it also has come over a threshold to greatness - it wasn't there for all of the past 20 years(?) Things such as your outreach in this post is what's needed for a great program to become more well known. My spontaneous feeling is that it's very upsetting that it has TeX in the name but does not use Tex or La…

Installing it sucks. Once I got through that hurdle I liked it. However, I do think most people won't bother if the executable downloaded from the main website just doesn't run.
Post reply on HN