Dear all,
Here are the meetup notes.
TL;DR: neosilator; feature/all-tex-fonts; backwards compatibility; \end{...}%
in latex;
typst; multi-page export; async LaTeX preview; ox-hugo, hugo; LaTeX formula
alignment;
MathML; org-table formula parser; how global id: links are searched; live LaTeX
previews;
indicating LaTeX errors in fragments; GNU licensing and money; Org
mode is not a GNU project; GNU & ethical payments; two meetups per month?
It was a rather long meetup. 4+ hours.
- As usual, we started from
https://sachachua.com/blog/2025/06/2025-06-09-emacs-news/
- Before the meetup I went through the news, capturing interesting
links. But not many people were present yet.
- We had a bit of (off-topic) discussion about russian word
"neosilator" (неосилятор), which I did not immediately recognize (despite
being
native-ish speaker). Pedro did though :)
- Apparently it is one of jargon words popular in the early days of
internet that refers to someone who has difficulties with learning
curve of a software (or not software)
- Zororg asked if I am russian
- I am not. I am Ukrainian. My russian is mostly spoken, often
mixed with Ukrainian (also English now). I am also not trained
in writing in russian.
- (that's where the proper meeting started)
- we also had issues with several people keeping their microphones turned on
- should not happen in the future. Sacha changed BBB settings to disable
microphones by default when people join.
- Pedro demoed an update on the WIP feature branch that implements
out-of-the-box export for non-latin Org documents
https://list.orgmode.org/CAO48Bk9oiqRy0Gj-MRVjNjkkuhyrK0nVxq7q2g+jnhG+vqpJ=a...@mail.gmail.com/T/#t
- He showcased Greek with LaTeX's polyglossia and babel (automatically
configured)
In lualatex. Materials:
https://list.orgmode.org/87ecvn4g94.fsf@localhost/T/#u
- Presentation as plain text is provided at the end
- *Pedro is interested in people familiar with various non-latin*
*scripts to test the branch and report any issues*
- We briefly discussed where to find examples of non-latin texts
- I recalled that I once found free version of Quran
(https://tanzil.net/download/)
- That won't help with checking if export right 100% right though
- We need people who know how things should be typeset
- Jacob S. Gordon asked about the new variable
~org-latex-fontspec-config~ introduced in the branch
- the variable sets (1) default fonts to be used for unicode scripts,
(2) backup fonts is the defaults are not available.
- the fonts are set independently for different rendering types:
main, sans, math, and mono
- that includes "math" that is used in equations (unicode-math
package that is used by default in the new branch)
- Pedro asked me about guidelines to handle backwards compatibility
- He has some issues maintaining backwards compatibility as the existing
code
does a lot of latex source fiddling with complex regexp replaces
that handle special "AUTO" language names (see "13.10.3 LaTeX
header and sectioning structure" section of Org manual
describing polyglossia/babel setup (that part is what is being
automated on the branch)
- We generally try to avoid breaking things as much as possible
See https://orgmode.org/worg/org-maintenance.html#guidelines
- For the purposes of developing the branch, Pedro may go ahead
and break things for the first iteration of the branch development.
First, let's make sure that the new feature is working
- We will tackle backwards compatibility later, closer to the time
when the review is completed and the branch is ready to be merged
- During his presentation, Pedro mentioned some issue with minipage
environment for LaTeX
- Apparently, it is necessary to end the minipage as
~\end{minipage}%~ - a common trick to suppress newline
and triggering new paragraph at the end of the environment
- without this trick, two adjacent minipages cannot be arranged
horizontally side-by-side
- I asked Pedro to consider fixing this in ~org-latex-special-block~
- We probably want to introduce a list of special latex environments
that are sensitive to the final newline
- we then moved to alternative typesetting system: typst
https://typst.app/
https://oylenshpeegul.gitlab.io/blog/posts/20250427/
https://github.com/jmpunkt/ox-typst
- The main selling point of typst is that it is /much faster/ than
LaTeX (because of incremental compilation + rust)
- A couple of years ago someone proposed to include typst export
support to Org mode:
https://list.orgmode.org/orgmode/87jzwjlovj.fsf@localhost/
- At that time, typst just had it early alpha release, so I rejected the idea
- Considering our backwards compatibility policies, it is not the best idea
to
add support to unstable/new software
- At this time, typst is getting fairly popular, but still has some
issues and is still fairly young (by Emacs's standards, heh)
- For example, I heard that it has problems typesetting large documents
- Although, Zororg shared his experience writing 50 pages thesis
purely in typst
- "... for me, it generates pdf within 1 second. For 500+ lines of 50
pages"
- Pedro A. Aranda shared his prior bad experience with typst
- They tried to replicate latex + overleaf workflow for typst
- However, getting collaborative editing part proved to be
tricky (that's what unstable software sometimes do)
- this might have nothing to do with compilation though
- Rudy asked about the status of the WIP multi-page HTML export
- Unfortunately, it is something that have been sitting on my backlog
since last August
- Since that time I have much, much less time and is still struggling to
keep up with the inflow of the new bug reports, patches, and feature
requests
(I can barely manage to, but anything extra is tricky)
- But that feature is important, so I need to do something to get things
going there
one way or another
- So, I now did HOLD -> NEXT status switch for that project. It gotta be
prioritized after waiting for so many months
- Since designing generic API for ox.el for multipage export remains tricky
(I will need to dedicate more than several days in a row), I can at least
attempt to get more
people involved into brainstorming and discussion
- In https://list.orgmode.org/orgmode/87tt4i1a2t.fsf@localhost/,
I attempted to get the ox-hugo maintainer involved. ox-hugo
does support multi-page export (by essentially implementing it
for MD export, similar to how Orm Finnendahl, the author of
multipage export idea, implemented it for HTML export)
- Rudy also asked about the status of async LaTeX preview by Karthink and
Timothy
- The project is taking quite a lot of time, partially because
Timothy is extremely loaded IRL and because Karthink is now
developing gptel (enough said)
- As Karthink pointed, the original plan was to merge the branch back in
2023
- the deadline kept shifting since then
- The branch has actually been sent for the review quite a while
ago, but I gave comments mostly on Timothy's part. That ain't
happening unfortunately due to RL©
- So, the plan is to get to do what we can and first try to review
the parts written by Karthink without putting too much load on
him (in other words, slowly, and prioritizing the most important
parts)
- see https://list.orgmode.org/orgmode/87bjqr4g2f.fsf@localhost/
- there are some backwards-incompatible changes for ob-latex
- but those are not as critical because ob-latex itself is buggy
and we have no goal of keeping all the bugs at all costs :)
- Unfortunately, we cannot merge the branch as is
1. It has breaking changes that should be carefully considered
2. It does not come with the manual, with many instructions scattered
across blog posts, readmes, etc. This needs to be sorted out
- Rudy commented that citations did not have a good manual at
the time of merging
- That's exactly the reason I am emphasizing to have the manual!!
We got enough headache with citations
3. I had some questions to specific design decisions and wanted to
make things maximally flexible in several areas
- Corwin Brust asked about the advantages of using ox-hugo vs. hugo
itself (hugo supports Org markup natively)
- The advantages are similar to advantages of ox-pandoc
- Hugo has no way to honor Elisp customizations/hooks/advices/custom
export config one can do for Org export:
https://orgmode.org/manual/Advanced-Export-Configuration.html
- Hugo also cannot handle babel and code evaluation
- Advanced Elisp macros also cannot be done with hugo (one needs Emacs for
those)
- Jacob S. Gordon asked about using MathML during export
- I am not very familiar with MathML, except that it is
something used by ox-odt
- I did not touch this area except a single security fix we did
for ~org-latex-to-mathml-convert-command~
- Corwin Brust shared his work on re-implementing Org table
formula parser and specifically regexps used there
https://cgit.git.savannah.gnu.org/cgit/dungeon.git/tree/src/ox-ox-test.el
https://cgit.git.savannah.gnu.org/cgit/dungeon.git/tree/src/ox-ox.el
- Contributing such improvements will be *very* welcome
- org-table is one of the legacy libraries I myself (despite being the
maintainer)
is not familiar with
- As a result, I have limited ability to fix things (let alone add new
features)
to spreadsheet functionality
- Even the fixes/improvements took a lot of time because org-table's global
state
often makes it difficult to keep things working after the most innocent
changes
- It would be really nice is someone worked on org-table code to make it more
modular and maintainable
- For example one can start by splitting ~org-table-eval-formula~ into
smaller pieces
and clearly documenting the process of formula reference expansion
- converting table formula regexps to rx notation, as Corwin did in
the above links would also be helpful
- somebody asked how global id: links in Org mode are searched (more
specifically, where they are searched)
- Org mode uses a combination of pre-defined list of files + all Org
buffers to search id: links in and dynamic cache that is
periodically saved to disk and contains links that have been
interactively created/references during Emacs session (links
created by manually typing :ID: <id> will not be auto-tracked
unless referenced via org-store-link later)
- by default, Org searches all agenda files, archives, and
~org-agenda-text-search-extra-files~, but this can be customized.
See ~org-id-extra-files~.
- The docstring of ~org-id-extra-files~ also details where Org
searches the id: links
- around this time or partially before Org table discussion we went on
discussing
LaTeX formula alignment in HTML pages and LaTeX previews
- most of the rendering systems (except Wikipedia's) do it wrong
- in many cases formula baseline is not well-aligned with the rest of the
text
- ... which gets worse when you zoom-in/out the HTML page
- the async latex preview branch does it much better than anything out there
(credit to Timothy for the implementation)
- Rudy tried to install the async preview branch live with screen
sharing and see how the LaTeX formulas are rendered live and in
HTML export
- Rudy has Mac machine, and we ran into a couple of svg library
problems (documented in FAQ, related to awkward Mac's
packaging), but eventually solved them
https://abode.karthinks.com/org-latex-preview/#orga2efcf6
- we had a strange issue with SVG previews being oversized
- We have managed to get an order of magnitude faster previews (~30
seconds for thousands of previews in the document) compared to
Org release branch (though Karthink says that it should be no
longer than 14 seconds, and that Rudy has some weird issues and
things are too slow on his machine)
- then we went on and looked into live latex preview with complex tikz
diagram being updated instantly as we were editing the latex sources
- when we made syntax errors, those were highlighted (and there
were errors in the example tikz diagram we checked, to Rudy's
surprise)
- that happened on mouse-over, but Karthink also shared a code
for bind error popup on C-c C-c
#+begin_src emacs-lisp
(defun my/org-latex-preview-show-error ()
(display-local-help t))
(add-hook 'org-ctrl-c-ctrl-c-final-hook
#'my/org-latex-preview-show-error -10)
#+end_src
- we also discussed ways to integrade the error indicator into
something like
flycheck
- I hope that we can eventually make org-lint library
integrate with flycheck (and be generally faster)
- then, it will be a natural fit for auto-indication of
problems like with latex previews
- Karthink also accidentally shared a similar hook, but for
looking through latex preview logs (will also work on main)
#+begin_src emacs-lisp
;; Custom DWIM functionality for `org-ctrl-c-ctrl-c'
(use-package org
:after (org ox-latex org-latex-preview)
:config
(defvar my/org-output-buffers
'("*Org Preview LaTeX Output*"
"*Org Preview Convert Output*"
"*Org PDF LaTeX Output*"
"*Org Preview Preamble Precompilation*"
"*Org LaTeX Precompilation*"))
(defun my/org-output-next-buffer ()
"Cycle through Org output buffers"
(let* ((org-buffer (current-buffer))
(bufnames
(cl-sort
(cl-remove-if-not
#'buffer-live-p
(cl-mapcar #'get-buffer my/org-output-buffers))
#'> :key #'buffer-modified-tick))
(next-output-buffer
(lambda () (interactive)
(setq bufnames
(nconc (cdr bufnames) (list (car bufnames))))
(switch-to-buffer (car bufnames))))
(back-to-org
(lambda () (interactive)
(while (memq (current-buffer) bufnames)
(quit-window))
(pop-to-buffer org-buffer))))
(when bufnames
(pop-to-buffer (car bufnames))
(set-transient-map
(define-keymap
"C-c C-c" next-output-buffer
"<remap> <keyboard-quit>" back-to-org)
(lambda () (eq (current-buffer) (car bufnames)))))))
(add-hook 'org-ctrl-c-ctrl-c-final-hook
#'my/org-output-next-buffer 99))
#+end_src
- (this is where some people in the chat got a bit too excited, up
to the point of posting NSFW memes; not posting those links here 🙂)
- I am also not posting the huge amount of flattery dumped onto Karthink
:)
- Karthink was still concerned about the degraded performance
though (by his standards!). So, we also tried to explore a bit
what might be the cause and tweaked a couple of internal-ish settings.
No luck live.
- ... so we moved on to trying the HTML export
- SVG-rendered latex formulas were indeed well-aligned on their
baseline with the normal text, even as we zoomed in/out the page
- After some tweaking of CSS stylesheet, we even managed the SVG formulas
to respect dark/light browser theme
- (it also worked for latex exported as PNG images)
- We then talked about GNU project, licensing, and money in free software
- nat asked about why GNU license can't somehow force companies using free
software to pay some money back
- my personal answer is the following
- first, ethically, forcing to pay for libre software is against
the idea of "libre". Although this ethical consideration
can be challenged and, indeed, it has been challenged by others
in the meetup. But there is other, more practical, consideration
- of course, GPL does not prevent monetization, but that monetization
usually concerns some extra services (support, cloud services, etc),
not
using the libre source code itself
- the current situation is that companies *do not want* to use
free software because of the copyleft that forces them to
open their own code
- the whole thing about "Open source" vs. "Libre software"
that FSF pushes is caused by companies actively trying to
sabotage copyleft and try to get open source code that does not
enforce them to open their code, let alone paying for that open
source software
- this sabotage is also the reason why assigning copyright to
FSF (or other similar organizations) is necessary
- by default (in most countries), if a company employee writes
GPL-licensed software, that GPL-licensed code belongs to the
company
- such ownership does not prevent using that GPL-licensed software
freely, but what it does is transferring the right to sue
any GPL violations to the companies
- if GPL violations do happen, companies never sue with
effective nullification of the whole idea of copyleft
- see
https://sfconservancy.org/blog/2021/jun/30/who-should-own-foss-copyrights/
- now imagine if we add payment to GPL license
- companies already try to avoid GPL like fire. If we add more to it,
the situation with GPL adoption will only get worse.
- of course, whether adding payment requirement will make things /worse/
is debatable. Maybe it will be about the same, but with great benefit
for GPL software. But ethical considerations will likely prevent
GPL changes like the proposed. And only FSF/similar companies can
practically
enforce such requirement in courts
- More generally, the problem with financing is indeed very real in
libre software world
- Bastien has been collecting some links on this topic in
https://github.com/bzg/opensource-challenges?tab=readme-ov-file#business-models
- Not everyone may know it, but Org mode is *not* a GNU project, even
though we practically try to follow all the ethical guidelines set by GNU
- In fact, for some time, I have been exploring the possibility for Org mode
to become an actual GNU project (just like TRAMP or Auctex)
- Note that Org mode being "a part of Emacs" does not mean that we
are subordinate of Emacs developers
- Similar to any other ELPA package, Emacs is free to distribute the source
code of any of the packages as a part of Emacs. Simply because
/all/ ELPA packages are GPL-compatible.
- But it does not put any authority of Emacs maintainers onto how
ELPA packages are developed and maintained. For example, GNU or Emacs devs
cannot prohibit using Github. They cannot prohibit advertising
proprietary software either.
- Despite not being a GNU package, Org is allowed to use savannah
for hosting, and we do use GNU mailing list hosting as our mailing
list server. What we do independently is website hosting (done by
Bastien), hosting an alternative mailing list UI (done by Kyle), and
hosting our CI infrastructure (done by me and Bastien on Sourcehut
via paid accounts)
- If Org does become a GNU project, we will be enforced to follow GNU
software guidelines
(https://www.gnu.org/prep/standards/standards.html), including
1. New maintainers will have to be approved by GNU
2. Obligation not to reference proprietary software (we currently don't)
3. Obligation to have our home page at https://gnu.org/software/org-mode
(easy; a simple redirect to https://orgmode.org is acceptable)
4. Obligation not to refer to websites that use non-libre JS
- There are at least two potential consequences of these obligations
1. WORG will not be able to reference proprietary software or even
JS-based websites like reddit/youtube/github
2. We won't be able to accept donations via Liberapay, because Liberapay
only provides Paypal and Stripe as payment options - both require
non-free JS
to do the payments
- The GNU's demand about payments is somewhat debatable though
- The ethically OK options available to do payments without running
non-free software are (1) crypto (well.. this one is just ok from GNU
perspective);
(2) GNU Taler (only working recently and only in Switzerland)
- GNU itself suggest "FSF working together fund", which is,
awkwardly offering Paypal for payments (with a warning that it
requires running non-free JS). Although, admittedly, they do
allow card payments without running JS at all
- notably, FSF fund charges 10% fee. It is fine as it is also working as
fiscal host and provides other services, but awkward when GNU
software is forced to use it
- the only alternative is https://www.spi-inc.org/ that charges 5%
but I have no clue about its reputation. (and they also offer
Paypal...)
- Here, someone asked why FSF is not offering financial support to GNU
maintainers
- The answer is simple - FSF is short on money by itself
- AFAIK, they had to lay off a number of people and the current FSF staff is
already overstretched. They have no ability to hire people they need
for necessary operations, let alone allocating money to support GNU
package maintainers
- ... and yes, it was related to scandals with RMS, which did hit FSF
pretty badly
- that said, people in the chat who worked with RMS shared their
experience
that RMS is not a bad guy, especially within the context of free
software
- the fact that his reputation in the areas he has less knowledge
affects FSF
is rather unfortunate side effect of how small communities work in
our society
(community reputation is often tied to leader reputation,
even outside the community scope)
- At the very end, we discussed a possibility to have this meetup twice a month
- I personally doubt that I can make it, given the amount of time I usually
spend to create meeting notes
- Crowin offered to help with the extra meetup time, possibly also to fit
better
to non-european time zones.
- But we will need to figure if meeting notes can work
- (we are discussing this option further now via email)
The Org source of Pedro's presentation during the meetup:
#+STARTUP: beamer
#+LATEX_COMPILER: lualatex
#+TITLE: feature/all-tex-fonts
#+SUBTITLE: Progress
#+AUTHOR: Pedro A. Aranda Gutiérrez \linebreak mailto:paag...@gmail.com
#+LANGUAGE: en
#+MACRO: pause @@latex: \pause@@
#+LATEX_CLASS_OPTIONS: [presentation,11pt,t,aspectratio=169]
#+LATEX_HEADER: \renewcommand*\familydefault{\sfdefault}
# +LATEX_HEADER: \input{packages}
#+BEAMER_THEME: Marburg
#+OPTIONS: H:2 toc:t num:t
* What's new?
** What's new?
- This is WIP: *things WILL break*
- Relying on local variables and document options
- Maximum flexibility for the user. {{{pause}}}
- By providing a minimalist default you can complement with ~#+LATEX_HEADER:~
{{{pause}}}
- Splitting the different configuration tasks into functions
- PDFLATEX -> WIP
- LUALATEX/XELATEX:
- Plain FONTSPEC: same as last time
- BABEL: initial support
- POLYGLOSSIA: initial support
** What's new?
- Instead of guessing font management /a posteriori/, we proactively generate a
full configuration from document options for lua/xelatex {{{pause}}}
- Use ~#+LANGUAGE:~ for the language list
- ~#+LATEX_MULTI_LANG:~ to derive the backend
1. ~babel~
1. ~polyglossia~
3. plain ~fontspec~ (if ~nil~) {{{pause}}}
- We use short language names and Org translates them appropriately to long
names.
- Suitable for the multi-language package selected
** Configuration for ~babel~
# +BEAMER: \framesubtitle{\textcolor{red}{\textbf{CAVEAT:}} Work in
\textbf{early} progress}
# - Early stage, still some coding needed (!)
- We start by defining ~#+LANGUAGE:~ and ~#+LATEX_MULTI_LANG:~
#+BEGIN_SRC org :results node :exports code
,#+LANGUAGE: de,el
,#+LATEX_COMPILER: lualatex
,#+LATEX_MULTI_LANG: babel
#+END_SRC
- draw font definitions from ~org-latex-babel-font-config~ only.
** Configuration for ~polyglossia~
- We start by defining ~#+LANGUAGE:~ and ~#+LATEX_MULTI_LANG:~
#+BEGIN_SRC org :results node :exports code
,#+LANGUAGE: de,el-polyton
,#+LATEX_COMPILER: lualatex
,#+LATEX_MULTI_LANG: polyglossia
#+END_SRC
- First language in the resulting language list is chosen as the main language
for the LaTeX export.
- Font definitions drawn from ~org-latex-polyglossia-font-config~ and
~org-latex-fontspec-config~
* Demo
** For the ~polyglossia~ demo
- I'm using the following header options:
#+BEGIN_SRC org :results none :exports code
,#+LANGUAGE: de,el-polyton
,#+LATEX_CLASS: article
,#+LATEX_CLASS_OPTIONS: [11pt,oneside,a4paper]
,#+LATEX_COMPILER: lualatex
,#+LATEX_MULTI_LANG: polyglossia
#+END_SRC
- and the following font configuration in ~.dir-locals.el~
#+BEGIN_SRC elisp :results none :exports code
((org-mode
. ((org-latex-fontspec-config
. (("main" :font "CMU Sans Serif")))
(org-latex-polyglossia-font-config
. (("greek" :font "Noto Serif" :props "Script=Greek"))))))
#+END_SRC
* Next steps, discussion
** This is a living development
- My objectives:
- Cover all features people need:
- My documents are basically monolingual and can live with a ~fontspec~ only
configuration. {{{pause}}}
- Reduce the complexity of the code
- Making it more maintainable in the future {{{pause}}}
- Test it if you have free cycles.
- I need /your/ feedback, your font setups will be better than mine.
- It's only a =git checkout= away if you are using =main=.{{{pause}}}
- I'm on LuaLaTeX
- /Please/ send me MWEs with your (PDF)LaTeX font configurations.
** Keeping package detection from ~#+LATEX_HEADER:~
#+BEAMER: \framesubtitle{Or not...}
- The feature branch provides the same functionality
- on lualatex/xelatex {{{pause}}}
- You can continue to use ~#+LATEX_HEADER:~ for
- your font settings
- extra options using ~\PassOptionsToPackage~ {{{pause}}}
- We don't need the stacked functions with the string replacement stuff
- Hard to read and understand
* Extras from the discussion
** Controlling the =math=
#+BEGIN_SRC elisp
((org-mode .
((org-latex-fontspec-config .
(("main". (:font "FreeSerif"
:fallback (("emoji" .
"Noto Color Emoji:mode=harf")
("han" .
"Noto Serif CJK JP:")
("kana" .
"Noto Serif CJK JP:"))))
("sans" . (:font "FreeSans"
:fallback (("emoji" .
"Noto Color Emoji:mode=harf")
("han" .
"Noto Sans CJK JP:")
("kana" .
"Noto Sans CJK JP:"))))
("math" . (:font "TeX Gyre Termes
Math"))
("mono" . (:font "Noto Sans Mono"
:features
("Scale=MatchLowercase"))))))))
#+END_SRC
- From last month's meetup (see ZIP in mailing list)
# local variables:
# eval: (auto-fill-mode t)
# spell-check-lang: "en_GB"
# org-latex-fontspec-config: nil
# End:
:chat:
[17:34] yantar92 is now the presenter
[17:39] [yantar92 : MODERATOR]: The most recent Emacs News:
https://sachachua.com/blog/2025/06/2025-06-09-emacs-news/
[17:44] [Obsidian : VIEWER]: Nice typing skills
[17:46] [yantar92 : MODERATOR]: The official start is in 15 minutes
[17:46] [yantar92 : MODERATOR]: we should have more people joining around that
time
[17:46] [Obsidian : VIEWER]: np. I just take obsidian nick
[17:50] [Obsidian : VIEWER]: Iam switch to obsidian for neosilator reason. Want
to see how actually people working in org.
[17:52] [yantar92 : MODERATOR]: what is "neosilator"?
[17:53] [Obsidian : VIEWER]: On russian neosilator means man who not mastered
in something)
[17:56] [Obsidian : VIEWER]: Google it like неосилятор
[17:58] [Zororg : VIEWER]: wait, yantar not russian?
[17:58] [Zororg : VIEWER]: Oh I thought name was like russia. Ihor Ra....
[17:59] [Fade : VIEWER]: 👋
[17:59] [Lumi : VIEWER]: Heya
[17:59] [Zororg : VIEWER]: hi fade, welcome 😊
[18:03] [Zororg : VIEWER]: I was recently trying typst. Imo it was so easy and
I got hyped up for it. I hope ox-typst expands more.
[18:04] Pedro A. Aranda is now the presenter, set by yantar92
[18:10] [Matthew Page-Lieberman : VIEWER]: Can the people who are not
presenting turn off their microphones?
[18:10] [yantar92 : MODERATOR]: Done 😊
[18:15] [Rudy : VIEWER]: This is all Greek to me.
[18:16] [yantar92 : MODERATOR]: if someone wants a different language we can
request a change
[18:16] [yantar92 : MODERATOR]: maybe we can break something 😁
[18:18] [oylenshpeegul : VIEWER]: I might suggest a right-to-left language, but
it looks justified anyway, so maybe it wouldn't make much difference.
[18:19] [yantar92 : MODERATOR]: yeah
[18:19] [yantar92 : MODERATOR]: can try https://tanzil.net/download/
[18:19] [Jake : VIEWER]: I think a couple people are activating their
microphone without realizing it.
[18:19] [yantar92 : MODERATOR]: yeah
[18:21] [Fade : VIEWER]: this is very cool.
[18:22] [Jacob S. Gordon : VIEWER]: Nice work! Slight tangent:
I like the org-latex-fontspec-config, is the unicode-math font done in a
similar way?
[18:24] [Jacob S. Gordon : VIEWER]: Fallback is great too, thanks!
[18:27] [yantar92 : MODERATOR]: About backwards compatibility:
https://orgmode.org/worg/org-maintenance.html#guidelines
[18:34] yantar92 is now the presenter, set by yantar92
[18:35] Pedro A. Aranda is now the presenter, set by yantar92
[18:36] yantar92 is now the presenter, set by yantar92
[18:38] [oylenshpeegul : VIEWER]: I messed with typst a bit
https://oylenshpeegul.gitlab.io/blog/posts/20250427/
[18:39] [Zororg : VIEWER]: I wrote my thesis of about 50 pages
[18:40] [yantar92 : MODERATOR]: https://github.com/jmpunkt/ox-typst
[18:42] [Zororg : VIEWER]: Yes, I used ox-typst and wrote
[18:42] [Zororg : VIEWER]: I had to embed lot of typst code tho, since ox-typst
was not complete
[18:44] [Rudy : VIEWER]: I would like to ask about the status of (1) the
revamped LaTeX subsystem by Karthink and (2) the WIP multi-page HTML export.
[18:47] [Zororg : VIEWER]: Maybe your issue can be reproduced.
But for me, it generates pdf within 1 second. For 500+ lines of 50 pages
[18:49] [Zororg : VIEWER]: yea, what i believe more is that typst syntax is
similar to org
Looks more like yaml
#image(("file.png", width: 80%), caption: [title], ..)
[18:50] [Zororg : VIEWER]: ^
[18:59] [Rudy : VIEWER]: Last year, it was estimated that this work would be
shipped at around Christmas. Do we have a new estimate now?
[19:00] [Rudy : VIEWER]: Sounds like the project is stuck, if Timothy is needed
but moved on. Oops!
[19:00] [Rudy : VIEWER]: 2023? Ha-ha.
[19:01] [Rudy : VIEWER]: Cannot we merge it into main and take it from there?
[19:02] [Rudy : VIEWER]: I asked because citations also did not have a perfect
manual.
[19:02] [Rudy : VIEWER]: [They still do not.]
[19:02] [Rudy : VIEWER]: So, about 2030. Optimistically. 😊
[19:06] [Zororg : VIEWER]: karthink nice to hear kannada lol.
I'm from bangalore cheers 😉
[19:23] [Corwin Brust : VIEWER]: I think hugo doesn't require any js what so
ever. there are many themes.
[19:23] [Corwin Brust : VIEWER]: am i wrong? i think it is a ruby program
[19:25] [Corwin Brust : VIEWER]: Whoops. No. It's Go.
[19:26] [Corwin Brust : VIEWER]: I have been using hugo directly to read
org-mode files, without using ox-hugo at all.
Can you talk about advantages of using ox-hugo?
[19:27] [Rudy : VIEWER]: Per the docs, Hugo supports MathJax and KaTeX, both JS.
[19:27] [Corwin Brust : VIEWER]: ohhhh!
[19:27] [Corwin Brust : VIEWER]: thank you.
[19:28] [Corwin Brust : VIEWER]: Nice! Makes sense! Right now II don't have
that but it sounds great.
[19:30] [Jacob S. Gordon : VIEWER]: Has anyone tried / had success with using
MathML?
[19:33] [Jacob S. Gordon : VIEWER]: Not necessarily! In HTML
[19:33] [Jacob S. Gordon : VIEWER]: Sorry no mic, will try to have one next time
[19:35] [Jacob S. Gordon : VIEWER]: Thanks! I'll take a closer look
[19:42] [Corwin Brust : VIEWER]:
https://cgit.git.savannah.gnu.org/cgit/dungeon.git/tree/src/ox-ox-test.el
[19:46] [Corwin Brust : VIEWER]:
https://cgit.git.savannah.gnu.org/cgit/dungeon.git/tree/src/ox-ox.el
[19:48] Rudy is now the presenter, set by yantar92
[19:49] yantar92 is now the presenter
[19:49] [Sacha Chua : VIEWER]: I'm here
[19:50] [Sacha Chua : VIEWER]: oh yeah, I should change that setting
[19:50] [Sacha Chua : VIEWER]: let me go do that
[19:50] Rudy is now the presenter, set by yantar92
[19:50] [Sacha Chua : VIEWER]: updated
[19:50] [yantar92 : MODERATOR]: thanks!
[19:51] [Sacha Chua : VIEWER]: ha =)
[19:57] [Corwin Brust : VIEWER]: I also am usually on the release branch, fwiw
[19:59] [Corwin Brust : VIEWER]: There's a NSFW meme I really want to use here.
Wow!
[20:00] [Jacob S. Gordon : VIEWER]: this is awesome
[20:05] [Corwin Brust : VIEWER]: <redacted link to NSFW meme>
[20:05] [Corwin Brust : VIEWER]: NSFW ^
[20:13] [Corwin Brust : VIEWER]: ha!
[20:28] [Corwin Brust : VIEWER]: 👏👏👏
[20:50] [Sacha Chua : VIEWER]: gotta head off, see y'all!
[20:50] [Jacob S. Gordon : VIEWER]: I have to run, thanks for the discussion!
[20:53] [karthink : VIEWER]: This message was deleted by karthink
[21:10] [karthink : VIEWER]:
https://paste.karthinks.com/eece0f9c-show-latex-errors.html
[21:12] [karthink : VIEWER]:
https://paste.karthinks.com/c9773599-show-latex-errors-2.html
[21:12] [Corwin Brust : VIEWER]: 🎵 love it!
[21:42] [nat : VIEWER]: I have a question:
If the GNU license is free as in libre, not money, why it does not include a
clause to make companies making money out of using free software pay back even
a very very very small percentage. When the profit is higher the percentage is
higher. For any personal or educational use, it will be totally free.
[21:45] [Corwin Brust : VIEWER]: He isn't a bad guy
[21:45] [Corwin Brust : VIEWER]: he says stuff
[21:45] [Corwin Brust : VIEWER]: and and turns people off
[21:46] [Corwin Brust : VIEWER]: preach
[21:46] [j8 : VIEWER]: i second this. I worked for rms a little bit. he isn't
a bad guy
[21:47] [yantar92 : MODERATOR]: auctex is a gnu project
[21:47] [yantar92 : MODERATOR]: https://www.gnu.org/software/auctex/
[21:48] [Corwin Brust : VIEWER]: Gotcha. That's one of the most sensitive
pieces of information that is on any shared VM we have.
[21:49] [nat : VIEWER]: What is the difference of org being a gnu project ? is
it some persones signature needed ?
[21:49] [nat : VIEWER]: that is all ? or is there something else ?
[21:50] [Rudy : VIEWER]: EMMS is also a GNU project /and/ an Emacs package.
[21:51] [Corwin Brust : VIEWER]: GNU is a project to create a free software
only os
[21:51] [Corwin Brust : VIEWER]: packages are anything that is installable
within that
[21:51] [nat : VIEWER]: no
[21:51] [Corwin Brust : VIEWER]: It's tricky because the frame of reference for
most users has changed since the 80s when this paradime was created
[21:51] [nat : VIEWER]: it is free as in libre
[21:52] [nat : VIEWER]: not free money
[21:52] [nat : VIEWER]: I did not get it
[21:52] [Corwin Brust : VIEWER]: When RMS created GNU linux didn't exist. We
now think about GNU as coming with a "distro" which wasn't a thing when GNU
started out.
[21:52] [nat : VIEWER]: can you please repeat ?
[21:53] yantar92 is now the presenter
[21:53] [nat : VIEWER]: If the GNU license is free as in libre, not money, why
it does not include a clause to make companies making money out of using free
software pay back even a very very very small percentage. When the profit is
higher the percentage is higher. For any personal or educational use, it will
be totally free.
[21:54] [nat : VIEWER]: no no
[21:54] [nat : VIEWER]: exactly
[21:54] [nat : VIEWER]: thanks Ihor
[21:54] [nat : VIEWER]: it is etheical
[21:54] [nat : VIEWER]: no no no
[21:54] [nat : VIEWER]: I disagree
[21:55] [Corwin Brust : VIEWER]: 👏
[21:55] [nat : VIEWER]: it is totally ethical
[21:55] [nat : VIEWER]: no no no
[21:55] [Corwin Brust : VIEWER]: 👏👏👏
[21:55] [nat : VIEWER]: everybody agrees
[21:55] [nat : VIEWER]: that it is ethical
[21:55] [Corwin Brust : VIEWER]: It's a great debate! I agree with Yantar (so
far!)
[21:55] [nat : VIEWER]: GPL does not prevent you from making money
[21:55] [nat : VIEWER]: out of it
[21:56] [nat : VIEWER]: I totally agree with you
[21:56] [nat : VIEWER]: but making money is totally ethical
[21:56] [nat : VIEWER]: from a GPL POV
[21:57] [karthink : VIEWER]: Ihor is saying that companies will not use free
software if they have to pay a percent of their profit in return
[21:57] [karthink : VIEWER]: So it doesn't matter if GNU thinks it's ethical
[21:58] [nat : VIEWER]: @kathink even if it is a 0.00001%
[21:58] [yantar92 : MODERATOR]: yup
[21:58] [nat : VIEWER]: it does not impact them
[21:58] [karthink : VIEWER]: This is generally not how they think
[21:58] [nat : VIEWER]: but it is a big help for free software
[21:58] [nat : VIEWER]: Ihor it is just0.000001%
[21:59] [karthink : VIEWER]: Companies avoid GPLv3 code like the plague
[21:59] [nat : VIEWER]: but they will have to pay much more for other softwares
[21:59] [nat : VIEWER]: it is just 0.00000001%
[21:59] [nat : VIEWER]: basically nothing
[22:00] [nat : VIEWER]: and if the profit are higher
[22:00] [karthink : VIEWER]: The part of GPLv3 they disagree with is having to
share what they build using it
[22:00] [nat : VIEWER]: the % became higher
[22:00] [karthink : VIEWER]: So the money is irrelevant
[22:00] [Corwin Brust : VIEWER]: +1
[22:00] [karthink : VIEWER]: What they build with it is their moat against
their competitors
[22:01] [karthink : VIEWER]: So they don't want to release it
[22:01] [Corwin Brust : VIEWER]: If someone's point is to have a technology
nobody else does there is not a good way to compromise that with the principles
of free software.
[22:02] [Corwin Brust : VIEWER]: Even removing monel it self likely would not
fix it. it is a motive conflict.
[22:02] [j8 : VIEWER]: Like if I use emacs to create some software do I have to
release the created software under GPLv3?
[22:02] [nat : VIEWER]: companies dont have to share anything
[22:02] [j8 : VIEWER]: I'm good. Thanks for your time 😊
[22:03] [nat : VIEWER]: I just hope if you can take tmie to think abou the idea
[22:03] [nat : VIEWER]: not now
[22:03] [nat : VIEWER]: but later
[22:03] [Corwin Brust : VIEWER]: i will!
[22:03] [nat : VIEWER]: Thank you all
[22:03] [Corwin Brust : VIEWER]: 💗
[22:03] [karthink : VIEWER]: Thanks for the meetup Ihor
[22:04] [nat : VIEWER]: because everyone needs to make a liveing
[22:04] [nat : VIEWER]: Thanks Ihor for your time
Ihor
22:04
Reply to message 164
one more thing
is it possible to make the meetup twice a month ?
22:04
2 shorter better than one longer ?
Reply to message 167
everyone is leaving early
:end:
--
Ihor Radchenko // yantar92,
Org mode maintainer,
Learn more about Org mode at <https://orgmode.org/>.
Support Org development at <https://liberapay.com/org-mode>,
or support my work at <https://liberapay.com/yantar92>
