<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:dc="http://purl.org/dc/elements/1.1/">
  <channel>
    <title>DEV Community: letx app</title>
    <description>The latest articles on DEV Community by letx app (@letx_app_76f3cdf1c781dfd7).</description>
    <link>https://dev.to/letx_app_76f3cdf1c781dfd7</link>
    <image>
      <url>https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F4034760%2Fc7a88f76-c1ec-4bc7-ba48-067e163e9ed7.png</url>
      <title>DEV Community: letx app</title>
      <link>https://dev.to/letx_app_76f3cdf1c781dfd7</link>
    </image>
    <atom:link rel="self" type="application/rss+xml" href="https://dev.to/feed/letx_app_76f3cdf1c781dfd7"/>
    <language>en</language>
    <item>
      <title>How Does Your Lab Handle Co-Authored Papers?</title>
      <dc:creator>letx app</dc:creator>
      <pubDate>Sun, 08 Nov 2026 14:00:00 +0000</pubDate>
      <link>https://dev.to/letx_app_76f3cdf1c781dfd7/how-does-your-lab-handle-co-authored-papers-415p</link>
      <guid>https://dev.to/letx_app_76f3cdf1c781dfd7/how-does-your-lab-handle-co-authored-papers-415p</guid>
      <description>&lt;p&gt;This week I wrote about comment macros, Git, track changes, response letters, shared bibliographies and merge conflicts. All of it came from reading and testing, not from watching your labs. So here is a question instead of a guide.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;How does your lab handle a paper with three or more authors?&lt;/strong&gt;&lt;/p&gt;

&lt;h2&gt;
  
  
  The setups I keep seeing described
&lt;/h2&gt;

&lt;p&gt;Pick the closest one, or describe your own.&lt;/p&gt;

&lt;ol&gt;
&lt;li&gt;
&lt;strong&gt;Email and attachments.&lt;/strong&gt; One person holds the master file. Others send edits back.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;A shared drive.&lt;/strong&gt; One file in a folder, with a naming convention like &lt;code&gt;paper_v7_final&lt;/code&gt;.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Git or GitHub.&lt;/strong&gt; The paper lives in a repository, with branches or pull requests.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;A live online editor.&lt;/strong&gt; Several people edit the same project at once.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;A word processor first, LaTeX at the end.&lt;/strong&gt; Draft in a shared document, then someone converts.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Something else.&lt;/strong&gt; A wiki, notebooks, a mix of tools.&lt;/li&gt;
&lt;/ol&gt;

&lt;h2&gt;
  
  
  What I would like to learn
&lt;/h2&gt;

&lt;ul&gt;
&lt;li&gt;Which setup does your lab use, and &lt;strong&gt;how many authors&lt;/strong&gt; does a typical paper have?&lt;/li&gt;
&lt;li&gt;What is the &lt;strong&gt;single worst moment&lt;/strong&gt; in the process? An overwritten edit, a lost version, the last-week scramble, a co-author who will not learn the tool?&lt;/li&gt;
&lt;li&gt;What is &lt;strong&gt;one rule&lt;/strong&gt; you adopted that made things better?&lt;/li&gt;
&lt;li&gt;If you switched setups, &lt;strong&gt;what made you switch&lt;/strong&gt;?&lt;/li&gt;
&lt;/ul&gt;

&lt;h2&gt;
  
  
  Why I am asking
&lt;/h2&gt;

&lt;p&gt;I work on &lt;a href="https://letx.app/?utm_source=devto&amp;amp;utm_medium=post&amp;amp;utm_campaign=latex-collab-series&amp;amp;utm_content=day14" rel="noopener noreferrer"&gt;LetX&lt;/a&gt;, a collaborative LaTeX editor, so I have a stake in how teams write together. The posts in this series are my best attempt at general advice, and they are only as good as the problems they match. Your answers decide what I write over the next weeks, so please be specific, and critical if you disagree with something I wrote.&lt;/p&gt;

&lt;p&gt;If you want to catch up, the series starts with &lt;a href="https://dev.to/letx_app_76f3cdf1c781dfd7/how-to-choose-a-collaborative-research-writing-tool-a-9-point-checklist-4o2m"&gt;a 9-point checklist for choosing a collaborative research writing tool&lt;/a&gt;.&lt;/p&gt;

&lt;p&gt;Next week the topic is choosing between tools: Overleaf alternatives, free plan limits, LaTeX versus Typst, and how to migrate a project.&lt;/p&gt;

&lt;p&gt;Tell me in the comments: which setup is yours, and what is its worst moment?&lt;/p&gt;

</description>
      <category>latex</category>
      <category>discuss</category>
      <category>research</category>
      <category>academia</category>
    </item>
    <item>
      <title>Replaying a Merge Conflict in a .tex File, Step by Step</title>
      <dc:creator>letx app</dc:creator>
      <pubDate>Sat, 07 Nov 2026 14:00:00 +0000</pubDate>
      <link>https://dev.to/letx_app_76f3cdf1c781dfd7/replaying-a-merge-conflict-in-a-tex-file-step-by-step-48c8</link>
      <guid>https://dev.to/letx_app_76f3cdf1c781dfd7/replaying-a-merge-conflict-in-a-tex-file-step-by-step-48c8</guid>
      <description>&lt;p&gt;Two co-authors edit one paragraph, and Git refuses to merge. If you have never seen this, it looks alarming. It is actually mechanical, and once you have walked through one, the next takes a minute.&lt;/p&gt;

&lt;p&gt;Here is a conflict I created in a real repository, with the real output, and then the fix. I ran every command shown.&lt;/p&gt;

&lt;h2&gt;
  
  
  The setup
&lt;/h2&gt;

&lt;p&gt;A paper has one file, &lt;code&gt;intro.tex&lt;/code&gt;, with a single paragraph written on one line:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;We evaluate on three benchmarks. The model trains for ten epochs. Results are averaged over five seeds.
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;Two authors each work on their own branch:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;Sam&lt;/strong&gt; changes &lt;code&gt;three benchmarks&lt;/code&gt; to &lt;code&gt;four benchmarks&lt;/code&gt;.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Priya&lt;/strong&gt; changes &lt;code&gt;five seeds&lt;/code&gt; to &lt;code&gt;ten seeds&lt;/code&gt;.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;These are different sentences, so a human would see no conflict at all.&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight shell"&gt;&lt;code&gt;git init &lt;span class="nt"&gt;-b&lt;/span&gt; main
git add intro.tex &lt;span class="o"&gt;&amp;amp;&amp;amp;&lt;/span&gt; git commit &lt;span class="nt"&gt;-m&lt;/span&gt; &lt;span class="s2"&gt;"base"&lt;/span&gt;

git checkout &lt;span class="nt"&gt;-b&lt;/span&gt; sam
&lt;span class="nb"&gt;sed&lt;/span&gt; &lt;span class="nt"&gt;-i&lt;/span&gt; &lt;span class="s1"&gt;'s/three benchmarks/four benchmarks/'&lt;/span&gt; intro.tex
git commit &lt;span class="nt"&gt;-am&lt;/span&gt; &lt;span class="s2"&gt;"Sam: four benchmarks"&lt;/span&gt;

git checkout main
git checkout &lt;span class="nt"&gt;-b&lt;/span&gt; priya
&lt;span class="nb"&gt;sed&lt;/span&gt; &lt;span class="nt"&gt;-i&lt;/span&gt; &lt;span class="s1"&gt;'s/five seeds/ten seeds/'&lt;/span&gt; intro.tex
git commit &lt;span class="nt"&gt;-am&lt;/span&gt; &lt;span class="s2"&gt;"Priya: ten seeds"&lt;/span&gt;
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;h2&gt;
  
  
  The merge
&lt;/h2&gt;

&lt;p&gt;Merge Sam's branch into &lt;code&gt;main&lt;/code&gt;. That works, because nothing else has changed on &lt;code&gt;main&lt;/code&gt;. Then merge Priya's:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight shell"&gt;&lt;code&gt;git checkout main
git merge sam
git merge priya
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;Git answers:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;Auto-merging intro.tex
CONFLICT (content): Merge conflict in intro.tex
Automatic merge failed; fix conflicts and then commit the result.
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;h2&gt;
  
  
  Reading the conflict
&lt;/h2&gt;

&lt;p&gt;Open &lt;code&gt;intro.tex&lt;/code&gt;. Git has written both versions into the file:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;&amp;lt;&amp;lt;&amp;lt;&amp;lt;&amp;lt;&amp;lt;&amp;lt; HEAD
We evaluate on four benchmarks. The model trains for ten epochs. Results are averaged over five seeds.
=======
We evaluate on three benchmarks. The model trains for ten epochs. Results are averaged over ten seeds.
&amp;gt;&amp;gt;&amp;gt;&amp;gt;&amp;gt;&amp;gt;&amp;gt; priya
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;The markers mean:&lt;/p&gt;

&lt;div class="table-wrapper-paragraph"&gt;&lt;table&gt;
&lt;thead&gt;
&lt;tr&gt;
&lt;th&gt;Marker&lt;/th&gt;
&lt;th&gt;Meaning&lt;/th&gt;
&lt;/tr&gt;
&lt;/thead&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td&gt;&lt;code&gt;&amp;lt;&amp;lt;&amp;lt;&amp;lt;&amp;lt;&amp;lt;&amp;lt; HEAD&lt;/code&gt;&lt;/td&gt;
&lt;td&gt;Start of the version on your current branch (&lt;code&gt;main&lt;/code&gt;, which now contains Sam's change)&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;&lt;code&gt;=======&lt;/code&gt;&lt;/td&gt;
&lt;td&gt;Divider between the two versions&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;&lt;code&gt;&amp;gt;&amp;gt;&amp;gt;&amp;gt;&amp;gt;&amp;gt;&amp;gt; priya&lt;/code&gt;&lt;/td&gt;
&lt;td&gt;End of the version coming from the &lt;code&gt;priya&lt;/code&gt; branch&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;&lt;/div&gt;

&lt;p&gt;Git compares by line, and the whole paragraph is one line, so any two edits to it collide, even if they touch different sentences.&lt;/p&gt;

&lt;h2&gt;
  
  
  Resolving it
&lt;/h2&gt;

&lt;p&gt;Decide what the paragraph should say, delete the markers, and keep both changes. Here that is a single line:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;We evaluate on four benchmarks. The model trains for ten epochs. Results are averaged over ten seeds.
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;Then tell Git the conflict is resolved:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight shell"&gt;&lt;code&gt;git add intro.tex
git commit &lt;span class="nt"&gt;-m&lt;/span&gt; &lt;span class="s2"&gt;"Merge priya: keep both edits"&lt;/span&gt;
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;Before you commit, search the file for &lt;code&gt;&amp;lt;&amp;lt;&amp;lt;&amp;lt;&amp;lt;&amp;lt;&amp;lt;&lt;/code&gt;, &lt;code&gt;=======&lt;/code&gt; and &lt;code&gt;&amp;gt;&amp;gt;&amp;gt;&amp;gt;&amp;gt;&amp;gt;&amp;gt;&lt;/code&gt;. A marker left in a &lt;code&gt;.tex&lt;/code&gt; file will break the build, or worse, end up in the PDF.&lt;/p&gt;

&lt;p&gt;If you want to give up and start over, &lt;code&gt;git merge --abort&lt;/code&gt; returns you to the state before the merge.&lt;/p&gt;

&lt;h2&gt;
  
  
  The same edits, one sentence per line
&lt;/h2&gt;

&lt;p&gt;Now repeat the experiment with the paragraph written one sentence per line:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;We evaluate on three benchmarks.
The model trains for ten epochs.
Results are averaged over five seeds.
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;Same two edits, same merge commands. This time Git reports a clean merge:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt; intro.tex | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;And the file contains both changes:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;We evaluate on four benchmarks.
The model trains for ten epochs.
Results are averaged over ten seeds.
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;LaTeX treats a single line break as a space, so the PDF looks the same. This one habit removes most prose conflicts. It does not remove all of them: if two people edit the &lt;strong&gt;same&lt;/strong&gt; sentence, you will still get a conflict, and you still decide which wording wins.&lt;/p&gt;

&lt;h2&gt;
  
  
  Three tips for the next conflict
&lt;/h2&gt;

&lt;ol&gt;
&lt;li&gt;
&lt;strong&gt;Read the whole conflict before editing.&lt;/strong&gt; Sometimes the two versions are not both right, and you need to talk to the co-author.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Resolve in the source, then compile.&lt;/strong&gt; A conflict that "looks fine" can still break the build.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Merge small and often.&lt;/strong&gt; Long-lived branches produce large conflicts. Short ones produce small ones.&lt;/li&gt;
&lt;/ol&gt;

&lt;h2&gt;
  
  
  If you would rather not merge at all
&lt;/h2&gt;

&lt;p&gt;Live collaborative editors avoid this step by merging edits as people type. I work on &lt;a href="https://letx.app/?utm_source=devto&amp;amp;utm_medium=post&amp;amp;utm_campaign=latex-collab-series&amp;amp;utm_content=day13" rel="noopener noreferrer"&gt;LetX&lt;/a&gt;, a collaborative LaTeX editor that uses a CRDT (Yjs) for this, with unlimited collaborators on its free plan. A live editor is not always the right choice, since Git gives you history, branches and automation that a live editor may not. I compare the two in &lt;a href="https://dev.to/letx_app_76f3cdf1c781dfd7/git-vs-a-collaborative-latex-editor-when-each-one-wins-3pjl"&gt;Git vs a collaborative LaTeX editor&lt;/a&gt;.&lt;/p&gt;

&lt;p&gt;This post is part of &lt;a href="https://dev.to/letx_app_76f3cdf1c781dfd7/how-to-choose-a-collaborative-research-writing-tool-a-9-point-checklist-4o2m"&gt;the series on collaborative research writing&lt;/a&gt;. Tomorrow: a question for you, about how your lab handles co-authored papers.&lt;/p&gt;

&lt;p&gt;What is the worst merge conflict you have had in a paper? Tell me in the comments.&lt;/p&gt;

</description>
      <category>latex</category>
      <category>git</category>
      <category>tutorial</category>
      <category>research</category>
    </item>
    <item>
      <title>One Shared .bib File for Co-Authors Without Merge Pain</title>
      <dc:creator>letx app</dc:creator>
      <pubDate>Fri, 06 Nov 2026 14:00:00 +0000</pubDate>
      <link>https://dev.to/letx_app_76f3cdf1c781dfd7/one-shared-bib-file-for-co-authors-without-merge-pain-204i</link>
      <guid>https://dev.to/letx_app_76f3cdf1c781dfd7/one-shared-bib-file-for-co-authors-without-merge-pain-204i</guid>
      <description>&lt;p&gt;The bibliography is the file everyone touches and nobody owns. Three co-authors add references in their own style, the same paper enters twice under different keys, and two people append an entry to the end of the file at the same moment, which is exactly where merges collide.&lt;/p&gt;

&lt;p&gt;Five rules prevent most of it. I ran the BibTeX examples below to confirm the messages.&lt;/p&gt;

&lt;h2&gt;
  
  
  Rule 1: one &lt;code&gt;.bib&lt;/code&gt; file, in the shared project
&lt;/h2&gt;

&lt;p&gt;Keep a single &lt;code&gt;refs.bib&lt;/code&gt; next to &lt;code&gt;main.tex&lt;/code&gt;. Do not let each author keep a private copy and paste entries in later, because that is how the same paper ends up with three different keys.&lt;/p&gt;

&lt;h2&gt;
  
  
  Rule 2: one key format, for everyone
&lt;/h2&gt;

&lt;p&gt;Agree a pattern and write it down. A common one is first author's surname, year, and a keyword from the title:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;vaswani2017attention
devlin2019bert
he2016deep
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;A fixed pattern means a co-author can guess a key without opening the file, and two people are less likely to add the same paper under different names.&lt;/p&gt;

&lt;h2&gt;
  
  
  Rule 3: format every entry the same way
&lt;/h2&gt;

&lt;p&gt;One field per line, the same indentation, and a trailing comma on every field line:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight bibtex"&gt;&lt;code&gt;&lt;span class="nc"&gt;@article&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;&lt;span class="nl"&gt;vaswani2017attention&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt;
  &lt;span class="na"&gt;author&lt;/span&gt;  &lt;span class="p"&gt;=&lt;/span&gt; &lt;span class="s"&gt;{Vaswani, Ashish and Shazeer, Noam}&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt;
  &lt;span class="na"&gt;title&lt;/span&gt;   &lt;span class="p"&gt;=&lt;/span&gt; &lt;span class="s"&gt;{Attention Is All You Need}&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt;
  &lt;span class="na"&gt;journal&lt;/span&gt; &lt;span class="p"&gt;=&lt;/span&gt; &lt;span class="s"&gt;{Advances in Neural Information Processing Systems}&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt;
  &lt;span class="na"&gt;year&lt;/span&gt;    &lt;span class="p"&gt;=&lt;/span&gt; &lt;span class="s"&gt;{2017}&lt;/span&gt;
&lt;span class="p"&gt;}&lt;/span&gt;
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;With one field per line, a version-control diff shows exactly which field changed. In a live editor it makes entries easy to scan. Consistent formatting also makes duplicate entries easier to spot by eye.&lt;/p&gt;

&lt;h2&gt;
  
  
  Rule 4: keep entries sorted by key
&lt;/h2&gt;

&lt;p&gt;If everyone appends to the end of the file, two people adding entries at the same time touch the same spot. Sorted entries spread the additions across the file, so different people usually edit different places. Reference managers and tools such as BibTool can normalize and sort a &lt;code&gt;.bib&lt;/code&gt; file, so you do not have to do it by hand.&lt;/p&gt;

&lt;h2&gt;
  
  
  Rule 5: check before you submit
&lt;/h2&gt;

&lt;p&gt;Two checks catch the most common problems, and BibTeX tells you about both.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;Duplicate keys.&lt;/strong&gt; If the same key appears twice, BibTeX stops at the second one:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;Repeated entry---line 7 of file refs.bib
 : @article{vaswani2017attention
 :                              ,
I'm skipping whatever remains of this entry
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;&lt;strong&gt;Missing fields.&lt;/strong&gt; If an entry lacks a field the style needs, you get a warning:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;Warning--empty journal in a2020x
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;Run BibTeX and read its output as part of the pre-submission check. Do not rely on the PDF alone, because a missing field can leave a reference half-formatted without any visible error.&lt;/p&gt;

&lt;h2&gt;
  
  
  What about a reference manager?
&lt;/h2&gt;

&lt;p&gt;If your team uses Zotero or a similar tool, you can export to &lt;code&gt;.bib&lt;/code&gt; and keep the library as the source of truth. Zotero's Better BibTeX add-on can keep an exported file updated automatically; check its documentation for the current setup. The trade-off is that the file is now generated, so edit entries in the manager, not in the &lt;code&gt;.bib&lt;/code&gt;, or the next export will overwrite your change.&lt;/p&gt;

&lt;p&gt;Pick one source of truth, either the manager or the &lt;code&gt;.bib&lt;/code&gt;, and tell everyone which it is.&lt;/p&gt;

&lt;h2&gt;
  
  
  A short policy to paste in your lab notes
&lt;/h2&gt;

&lt;ol&gt;
&lt;li&gt;All references live in one &lt;code&gt;refs.bib&lt;/code&gt; in the project.&lt;/li&gt;
&lt;li&gt;Keys follow &lt;code&gt;surname + year + keyword&lt;/code&gt;.&lt;/li&gt;
&lt;li&gt;One field per line, entries sorted by key.&lt;/li&gt;
&lt;li&gt;Before submission, one person runs BibTeX and clears every warning.&lt;/li&gt;
&lt;li&gt;If you use a reference manager, edits are made there, not in the exported file.&lt;/li&gt;
&lt;/ol&gt;

&lt;h2&gt;
  
  
  If your team writes in a shared editor
&lt;/h2&gt;

&lt;p&gt;A shared &lt;code&gt;.bib&lt;/code&gt; is easier when everyone edits the same project, since there is one copy to fix. I work on &lt;a href="https://letx.app/?utm_source=devto&amp;amp;utm_medium=post&amp;amp;utm_campaign=latex-collab-series&amp;amp;utm_content=day12" rel="noopener noreferrer"&gt;LetX&lt;/a&gt;, a collaborative LaTeX editor with unlimited collaborators on its free plan, and these rules apply there as in any other setup. The free plan has limits such as 40 compiles a day.&lt;/p&gt;

&lt;p&gt;Next in this series: a merge conflict in a &lt;code&gt;.tex&lt;/code&gt; file, replayed step by step. This post is part of &lt;a href="https://dev.to/letx_app_76f3cdf1c781dfd7/how-to-choose-a-collaborative-research-writing-tool-a-9-point-checklist-4o2m"&gt;the series on collaborative research writing&lt;/a&gt;.&lt;/p&gt;

&lt;p&gt;How does your team handle references? Tell me in the comments.&lt;/p&gt;

</description>
      <category>latex</category>
      <category>research</category>
      <category>tutorial</category>
      <category>productivity</category>
    </item>
    <item>
      <title>A Response-to-Reviewers Letter in LaTeX That Survives Revision</title>
      <dc:creator>letx app</dc:creator>
      <pubDate>Thu, 05 Nov 2026 14:00:00 +0000</pubDate>
      <link>https://dev.to/letx_app_76f3cdf1c781dfd7/a-response-to-reviewers-letter-in-latex-that-survives-revision-46jn</link>
      <guid>https://dev.to/letx_app_76f3cdf1c781dfd7/a-response-to-reviewers-letter-in-latex-that-survives-revision-46jn</guid>
      <description>&lt;p&gt;A revision is a second deadline, and the response letter is where many of them go wrong. Comments get skipped, responses drift from what the manuscript now says, and the editor cannot find where a change was made.&lt;/p&gt;

&lt;p&gt;A good response letter is mostly structure. This post gives you a LaTeX template that enforces the structure, a workflow for the manuscript changes, and a short checklist. I compiled the template to make sure it works.&lt;/p&gt;

&lt;p&gt;Journals differ in what they require, so check your venue's instructions for format and length. Treat this as a base to adapt.&lt;/p&gt;

&lt;h2&gt;
  
  
  The template
&lt;/h2&gt;

&lt;p&gt;It numbers each reviewer's comments automatically, quotes the comment in a shaded box, and gives you a &lt;code&gt;response&lt;/code&gt; block under each one. Changes you made in the manuscript are marked in blue so the editor can scan for them.&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight tex"&gt;&lt;code&gt;&lt;span class="k"&gt;\documentclass&lt;/span&gt;&lt;span class="na"&gt;[11pt]&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;article&lt;span class="p"&gt;}&lt;/span&gt;
&lt;span class="k"&gt;\usepackage&lt;/span&gt;&lt;span class="na"&gt;[margin=1in]&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;geometry&lt;span class="p"&gt;}&lt;/span&gt;
&lt;span class="k"&gt;\usepackage&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;xcolor&lt;span class="p"&gt;}&lt;/span&gt;
&lt;span class="k"&gt;\usepackage&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;tcolorbox&lt;span class="p"&gt;}&lt;/span&gt;

&lt;span class="k"&gt;\newcounter&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;rc&lt;span class="p"&gt;}&lt;/span&gt;
&lt;span class="k"&gt;\newcommand&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;&lt;span class="k"&gt;\reviewer&lt;/span&gt;&lt;span class="p"&gt;}&lt;/span&gt;[1]&lt;span class="p"&gt;{&lt;/span&gt;&lt;span class="k"&gt;\section*&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;Reviewer #1&lt;span class="p"&gt;}&lt;/span&gt;&lt;span class="k"&gt;\setcounter&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;rc&lt;span class="p"&gt;}{&lt;/span&gt;0&lt;span class="p"&gt;}}&lt;/span&gt;
&lt;span class="k"&gt;\newcommand&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;&lt;span class="k"&gt;\rcomment&lt;/span&gt;&lt;span class="p"&gt;}&lt;/span&gt;[1]&lt;span class="p"&gt;{&lt;/span&gt;&lt;span class="c"&gt;%&lt;/span&gt;
  &lt;span class="k"&gt;\stepcounter&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;rc&lt;span class="p"&gt;}&lt;/span&gt;&lt;span class="c"&gt;%&lt;/span&gt;
  &lt;span class="nt"&gt;\begin{tcolorbox}&lt;/span&gt;[colback=gray!8, colframe=gray!50, title=Comment &lt;span class="k"&gt;\arabic&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;rc&lt;span class="p"&gt;}&lt;/span&gt;]#1&lt;span class="nt"&gt;\end{tcolorbox}&lt;/span&gt;&lt;span class="p"&gt;}&lt;/span&gt;
&lt;span class="k"&gt;\newenvironment&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;response&lt;span class="p"&gt;}{&lt;/span&gt;&lt;span class="k"&gt;\par\noindent\textbf&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;Response.&lt;span class="p"&gt;}&lt;/span&gt;&lt;span class="k"&gt;\ &lt;/span&gt;&lt;span class="p"&gt;}{&lt;/span&gt;&lt;span class="k"&gt;\par\bigskip&lt;/span&gt;&lt;span class="p"&gt;}&lt;/span&gt;
&lt;span class="k"&gt;\newcommand&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;&lt;span class="k"&gt;\changed&lt;/span&gt;&lt;span class="p"&gt;}&lt;/span&gt;[1]&lt;span class="p"&gt;{{&lt;/span&gt;&lt;span class="k"&gt;\color&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;blue&lt;span class="p"&gt;}&lt;/span&gt;#1&lt;span class="p"&gt;}}&lt;/span&gt;

&lt;span class="nt"&gt;\begin{document}&lt;/span&gt;
&lt;span class="k"&gt;\section*&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;Response to Reviewers&lt;span class="p"&gt;}&lt;/span&gt;
We thank the reviewers. Changes in the manuscript are marked in &lt;span class="k"&gt;\changed&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;blue&lt;span class="p"&gt;}&lt;/span&gt;.

&lt;span class="k"&gt;\reviewer&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;1&lt;span class="p"&gt;}&lt;/span&gt;
&lt;span class="k"&gt;\rcomment&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;The evaluation uses only three benchmarks.&lt;span class="p"&gt;}&lt;/span&gt;
&lt;span class="nt"&gt;\begin{response}&lt;/span&gt;
We added a fourth benchmark (Section 5.2, Table 3).
&lt;span class="k"&gt;\changed&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;The result is consistent with the other three.&lt;span class="p"&gt;}&lt;/span&gt;
&lt;span class="nt"&gt;\end{response}&lt;/span&gt;

&lt;span class="k"&gt;\rcomment&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;Please state the learning rate.&lt;span class="p"&gt;}&lt;/span&gt;
&lt;span class="nt"&gt;\begin{response}&lt;/span&gt;
The learning rate is now given in Section 4.1.
&lt;span class="nt"&gt;\end{response}&lt;/span&gt;

&lt;span class="k"&gt;\reviewer&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;2&lt;span class="p"&gt;}&lt;/span&gt;
&lt;span class="k"&gt;\rcomment&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;The related work omits recent baselines.&lt;span class="p"&gt;}&lt;/span&gt;
&lt;span class="nt"&gt;\begin{response}&lt;/span&gt;
We agree and added three baselines to Section 2.
&lt;span class="nt"&gt;\end{response}&lt;/span&gt;
&lt;span class="nt"&gt;\end{document}&lt;/span&gt;
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;It compiles on a standard TeX Live install with &lt;code&gt;pdflatex&lt;/code&gt;. Each reviewer gets a heading, comments restart at 1 under each reviewer, and each response sits directly under the comment it answers.&lt;/p&gt;

&lt;h2&gt;
  
  
  The workflow
&lt;/h2&gt;

&lt;p&gt;&lt;strong&gt;1. Copy every reviewer comment verbatim.&lt;/strong&gt; Paste them in as &lt;code&gt;\rcomment{...}&lt;/code&gt;, one per point. If a reviewer's paragraph holds three points, split it into three comments so each gets its own answer.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;2. Make the manuscript changes first, then write the responses.&lt;/strong&gt; A response should describe what the manuscript says now. Writing responses before the edits is how letters end up describing changes that never happened.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;3. Point to a location every time.&lt;/strong&gt; "We added this in Section 5.2, Table 3" is easy to check. "We addressed this" is not.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;4. Mark changes in the manuscript.&lt;/strong&gt; Define a macro in the manuscript preamble, and wrap new or changed text in it:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight tex"&gt;&lt;code&gt;&lt;span class="k"&gt;\newif\ifshowrev&lt;/span&gt; &lt;span class="k"&gt;\showrevtrue&lt;/span&gt;            &lt;span class="c"&gt;% \showrevfalse for the final version&lt;/span&gt;
&lt;span class="k"&gt;\newcommand&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;&lt;span class="k"&gt;\rev&lt;/span&gt;&lt;span class="p"&gt;}&lt;/span&gt;[1]&lt;span class="p"&gt;{&lt;/span&gt;&lt;span class="k"&gt;\ifshowrev&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;&lt;span class="k"&gt;\color&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;blue&lt;span class="p"&gt;}&lt;/span&gt;#1&lt;span class="p"&gt;}&lt;/span&gt;&lt;span class="k"&gt;\else&lt;/span&gt; #1&lt;span class="k"&gt;\fi&lt;/span&gt;&lt;span class="p"&gt;}&lt;/span&gt;
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;Unlike a comment macro, this keeps the text in both modes and only changes its color. For a full marked-up file, &lt;code&gt;latexdiff&lt;/code&gt; between the submitted and revised versions is the usual route. I covered both in &lt;a href="https://dev.to/letx_app_76f3cdf1c781dfd7/track-changes-in-latex-latexdiff-the-changes-package-and-what-to-use-3ap2"&gt;yesterday's track-changes post&lt;/a&gt;.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;5. Cross-check at the end.&lt;/strong&gt; Count the comments in the letter against the reviewers' reports. Every one should have a response.&lt;/p&gt;

&lt;h2&gt;
  
  
  Tone rules that save rounds
&lt;/h2&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;Thank once, briefly.&lt;/strong&gt; Spend your words on the answers.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Answer the point, not the reviewer.&lt;/strong&gt; Stay on the content.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Disagree with evidence.&lt;/strong&gt; If you did not make a change, say why, and point to a result, a reference or a limitation you now state in the paper.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Never ignore a comment.&lt;/strong&gt; If you cannot address one, say so and explain.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Keep it short where you can.&lt;/strong&gt; A one-line fix needs a one-line response.&lt;/li&gt;
&lt;/ul&gt;

&lt;h2&gt;
  
  
  A checklist for the day you submit
&lt;/h2&gt;

&lt;ul&gt;
&lt;li&gt;[ ] Every reviewer comment appears in the letter, verbatim and numbered&lt;/li&gt;
&lt;li&gt;[ ] Every response gives a section, page or table reference&lt;/li&gt;
&lt;li&gt;[ ] The references match the revised manuscript (page numbers shift)&lt;/li&gt;
&lt;li&gt;[ ] Changed text is marked in the manuscript or in a diff file&lt;/li&gt;
&lt;li&gt;[ ] The letter compiles cleanly, with no &lt;code&gt;??&lt;/code&gt;
&lt;/li&gt;
&lt;li&gt;[ ] Co-authors have read the final PDF of both the letter and the paper&lt;/li&gt;
&lt;/ul&gt;

&lt;h2&gt;
  
  
  Working on it together
&lt;/h2&gt;

&lt;p&gt;A response letter is a shared document with a tight deadline, which is where live collaboration helps most: one co-author handles each reviewer, and everyone sees the letter and the manuscript change together. I work on &lt;a href="https://letx.app/?utm_source=devto&amp;amp;utm_medium=post&amp;amp;utm_campaign=latex-collab-series&amp;amp;utm_content=day11" rel="noopener noreferrer"&gt;LetX&lt;/a&gt;, a collaborative LaTeX editor with unlimited collaborators on its free plan, though the free plan limits compiles to 40 a day, which is worth knowing during a revision crunch. The template above works in any editor.&lt;/p&gt;

&lt;p&gt;Next in this series: one shared bibliography file for co-authors, without merge pain. This post is part of &lt;a href="https://dev.to/letx_app_76f3cdf1c781dfd7/how-to-choose-a-collaborative-research-writing-tool-a-9-point-checklist-4o2m"&gt;the series on collaborative research writing&lt;/a&gt;.&lt;/p&gt;

&lt;p&gt;What is the best response letter structure you have seen? Tell me in the comments.&lt;/p&gt;

</description>
      <category>latex</category>
      <category>research</category>
      <category>academia</category>
      <category>tutorial</category>
    </item>
    <item>
      <title>Track Changes in LaTeX: latexdiff, the changes Package, and What to Use</title>
      <dc:creator>letx app</dc:creator>
      <pubDate>Wed, 04 Nov 2026 14:00:00 +0000</pubDate>
      <link>https://dev.to/letx_app_76f3cdf1c781dfd7/track-changes-in-latex-latexdiff-the-changes-package-and-what-to-use-3ap2</link>
      <guid>https://dev.to/letx_app_76f3cdf1c781dfd7/track-changes-in-latex-latexdiff-the-changes-package-and-what-to-use-3ap2</guid>
      <description>&lt;p&gt;Word has Track Changes. LaTeX has nothing built in, which surprises many people who move over for a paper with co-authors. The good news is that there are three solid options, and they suit different moments in a paper's life.&lt;/p&gt;

&lt;p&gt;I ran every command and snippet below to check it works.&lt;/p&gt;

&lt;h2&gt;
  
  
  Option 1: latexdiff, for comparing two versions
&lt;/h2&gt;

&lt;p&gt;&lt;code&gt;latexdiff&lt;/code&gt; takes an old and a new version of a file and writes a third file with the differences marked: additions in blue wavy underline, deletions in red strikethrough.&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight shell"&gt;&lt;code&gt;latexdiff old.tex new.tex &lt;span class="o"&gt;&amp;gt;&lt;/span&gt; diff.tex
pdflatex diff.tex
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;With an old version containing "three benchmarks" and a new one containing "four benchmarks ... and uses early stopping", the changed words come out wrapped in markup like &lt;code&gt;\DIFdel{three}&lt;/code&gt; and &lt;code&gt;\DIFadd{four}&lt;/code&gt;, and the compiled PDF shows them colored.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;With Git&lt;/strong&gt;, you do not even need to save the old file. &lt;code&gt;latexdiff-vc&lt;/code&gt; fetches it from a commit:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight shell"&gt;&lt;code&gt;latexdiff-vc &lt;span class="nt"&gt;--git&lt;/span&gt; &lt;span class="nt"&gt;-r&lt;/span&gt; HEAD~1 main.tex
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;It extracts the old version, runs &lt;code&gt;latexdiff&lt;/code&gt;, and writes a file such as &lt;code&gt;main-diffHEAD~1.tex&lt;/code&gt;.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;For a paper split into many files&lt;/strong&gt;, add &lt;code&gt;--flatten&lt;/code&gt; so the included files are merged first:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight shell"&gt;&lt;code&gt;latexdiff &lt;span class="nt"&gt;--flatten&lt;/span&gt; old/main.tex new/main.tex &lt;span class="o"&gt;&amp;gt;&lt;/span&gt; diff.tex
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;&lt;strong&gt;Best for:&lt;/strong&gt; the marked-up manuscript that some journals ask for with a revision, and for checking what a co-author changed overnight.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;Watch out:&lt;/strong&gt; the output is a normal LaTeX file, so always compile it. Complex math and custom environments can confuse it, and sometimes the diff needs a manual fix before it builds.&lt;/p&gt;

&lt;h2&gt;
  
  
  Option 2: the &lt;code&gt;changes&lt;/code&gt; package, for marking edits as you go
&lt;/h2&gt;

&lt;p&gt;The &lt;code&gt;changes&lt;/code&gt; package lets each author mark their own edits inline, with a color per author:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight tex"&gt;&lt;code&gt;&lt;span class="k"&gt;\usepackage&lt;/span&gt;&lt;span class="na"&gt;[draft]&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;changes&lt;span class="p"&gt;}&lt;/span&gt;
&lt;span class="k"&gt;\definechangesauthor&lt;/span&gt;&lt;span class="na"&gt;[color=blue]&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;AK&lt;span class="p"&gt;}&lt;/span&gt;
&lt;span class="k"&gt;\definechangesauthor&lt;/span&gt;&lt;span class="na"&gt;[color=red]&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;SM&lt;span class="p"&gt;}&lt;/span&gt;

We evaluate on &lt;span class="k"&gt;\replaced&lt;/span&gt;&lt;span class="na"&gt;[id=AK]&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;four&lt;span class="p"&gt;}{&lt;/span&gt;three&lt;span class="p"&gt;}&lt;/span&gt; benchmarks&lt;span class="c"&gt;%&lt;/span&gt;
&lt;span class="k"&gt;\added&lt;/span&gt;&lt;span class="na"&gt;[id=SM, comment={Mention early stopping}]&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt; and use early stopping&lt;span class="p"&gt;}&lt;/span&gt;.
&lt;span class="k"&gt;\deleted&lt;/span&gt;&lt;span class="na"&gt;[id=AK]&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;This sentence is removed.&lt;span class="p"&gt;}&lt;/span&gt;
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;In &lt;code&gt;draft&lt;/code&gt; mode, the PDF shows additions, deletions and replacements in each author's color. When the edits are accepted, change one word:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight tex"&gt;&lt;code&gt;&lt;span class="k"&gt;\usepackage&lt;/span&gt;&lt;span class="na"&gt;[final]&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;changes&lt;span class="p"&gt;}&lt;/span&gt;
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;With &lt;code&gt;final&lt;/code&gt;, the compiled text reads: "We evaluate on four benchmarks and use early stopping." The deleted sentence and all the markup are gone. I compiled both modes to confirm.&lt;/p&gt;

&lt;p&gt;The commands you will use most:&lt;/p&gt;

&lt;div class="table-wrapper-paragraph"&gt;&lt;table&gt;
&lt;thead&gt;
&lt;tr&gt;
&lt;th&gt;Command&lt;/th&gt;
&lt;th&gt;Meaning&lt;/th&gt;
&lt;/tr&gt;
&lt;/thead&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td&gt;&lt;code&gt;\added[id=AK]{new}&lt;/code&gt;&lt;/td&gt;
&lt;td&gt;New text&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;&lt;code&gt;\deleted[id=AK]{old}&lt;/code&gt;&lt;/td&gt;
&lt;td&gt;Text to remove&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;&lt;code&gt;\replaced[id=AK]{new}{old}&lt;/code&gt;&lt;/td&gt;
&lt;td&gt;Replace old with new&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;&lt;code&gt;\highlight[id=AK]{text}&lt;/code&gt;&lt;/td&gt;
&lt;td&gt;Draw attention to text&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;&lt;code&gt;\comment[id=AK]{note}&lt;/code&gt;&lt;/td&gt;
&lt;td&gt;A comment, no text change&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;&lt;/div&gt;

&lt;p&gt;&lt;strong&gt;Best for:&lt;/strong&gt; a draft that several people edit over weeks, where you want to see who suggested what.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;Watch out:&lt;/strong&gt; markup is inside the source, so the text becomes harder to read. Accept or reject edits regularly, and remove the markup before submission by switching to &lt;code&gt;final&lt;/code&gt; and cleaning up.&lt;/p&gt;

&lt;h2&gt;
  
  
  Option 3: comment macros, for discussion without changing text
&lt;/h2&gt;

&lt;p&gt;When you want to ask a question rather than change a sentence, a comment command is lighter than tracked edits. I wrote up a &lt;a href="https://dev.to/letx_app_76f3cdf1c781dfd7/per-author-comment-macros-for-co-authors-in-latex-2l7a"&gt;per-author comment macro setup&lt;/a&gt; earlier in this series, including a strict mode that fails the build if a note is left behind.&lt;/p&gt;

&lt;h2&gt;
  
  
  Which one should you use?
&lt;/h2&gt;

&lt;div class="table-wrapper-paragraph"&gt;&lt;table&gt;
&lt;thead&gt;
&lt;tr&gt;
&lt;th&gt;You want to...&lt;/th&gt;
&lt;th&gt;Use&lt;/th&gt;
&lt;/tr&gt;
&lt;/thead&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td&gt;See what changed between two saved versions&lt;/td&gt;
&lt;td&gt;&lt;code&gt;latexdiff&lt;/code&gt;&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Show reviewers a marked-up revision&lt;/td&gt;
&lt;td&gt;
&lt;code&gt;latexdiff&lt;/code&gt; (check the journal's instructions)&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Mark suggestions inline, by author&lt;/td&gt;
&lt;td&gt;&lt;code&gt;changes&lt;/code&gt;&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Ask a question, no text change&lt;/td&gt;
&lt;td&gt;Comment macros&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Keep the source readable&lt;/td&gt;
&lt;td&gt;
&lt;code&gt;latexdiff&lt;/code&gt; or comments, not &lt;code&gt;changes&lt;/code&gt;
&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;&lt;/div&gt;

&lt;p&gt;A practical combination for a paper with co-authors: comment macros during drafting, &lt;code&gt;changes&lt;/code&gt; for suggested edits in the middle stage, and &lt;code&gt;latexdiff&lt;/code&gt; when you produce the revision for reviewers.&lt;/p&gt;

&lt;h2&gt;
  
  
  A note on live editors
&lt;/h2&gt;

&lt;p&gt;Some collaborative editors have their own review features. Overleaf lists track changes among its paid-plan features. I work on &lt;a href="https://letx.app/?utm_source=devto&amp;amp;utm_medium=post&amp;amp;utm_campaign=latex-collab-series&amp;amp;utm_content=day10" rel="noopener noreferrer"&gt;LetX&lt;/a&gt;, and the pages I have read do not describe a built-in track-changes feature, so for review I would use the three options above and check the current feature list yourself. Because LetX keeps a version history (the free plan keeps the last 15 versions, and full history is on Pro), you can also export two versions and run &lt;code&gt;latexdiff&lt;/code&gt; between them.&lt;/p&gt;

&lt;p&gt;Next in this series: writing a response-to-reviewers letter that survives revision. This post is part of &lt;a href="https://dev.to/letx_app_76f3cdf1c781dfd7/how-to-choose-a-collaborative-research-writing-tool-a-9-point-checklist-4o2m"&gt;the series on collaborative research writing&lt;/a&gt;.&lt;/p&gt;

&lt;p&gt;Which of the three do you use today? Tell me in the comments.&lt;/p&gt;

</description>
      <category>latex</category>
      <category>research</category>
      <category>tutorial</category>
      <category>academia</category>
    </item>
    <item>
      <title>Git vs a Collaborative LaTeX Editor: When Each One Wins</title>
      <dc:creator>letx app</dc:creator>
      <pubDate>Tue, 03 Nov 2026 14:00:00 +0000</pubDate>
      <link>https://dev.to/letx_app_76f3cdf1c781dfd7/git-vs-a-collaborative-latex-editor-when-each-one-wins-3pjl</link>
      <guid>https://dev.to/letx_app_76f3cdf1c781dfd7/git-vs-a-collaborative-latex-editor-when-each-one-wins-3pjl</guid>
      <description>&lt;p&gt;"Should we just use Git?" is the most common question in a lab that has outgrown email. The honest answer is that Git and a collaborative LaTeX editor solve different problems, and many teams do best using both.&lt;/p&gt;

&lt;p&gt;This post compares them on the things that matter for a co-authored paper, and shows a setup that gets the benefits of each.&lt;/p&gt;

&lt;h2&gt;
  
  
  What Git is good at
&lt;/h2&gt;

&lt;p&gt;Git was built for source code, and it shines when your paper behaves like code:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;Reproducibility.&lt;/strong&gt; If your figures are generated by scripts in the same repository, one commit holds the code, the data and the paper that came from them.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;A permanent history.&lt;/strong&gt; Every change is recorded and can be compared, tagged and restored, with no limit on how far back you go.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Branches.&lt;/strong&gt; Each author, or each big rewrite, can live on its own branch until it is ready.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Automation.&lt;/strong&gt; A CI job can compile the paper on every push and fail if it breaks.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Ownership.&lt;/strong&gt; The files live on your machines and your server, and no vendor is involved.&lt;/li&gt;
&lt;/ul&gt;

&lt;h2&gt;
  
  
  Where Git hurts for LaTeX
&lt;/h2&gt;

&lt;p&gt;Git compares files &lt;strong&gt;line by line&lt;/strong&gt;. LaTeX paragraphs are usually one very long line, so two edits to different sentences in the same paragraph count as a conflict. I reproduced this with a real repository, and I replay it step by step later this week; here is the short version.&lt;/p&gt;

&lt;p&gt;Two authors edit different sentences in one paragraph:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;We evaluate on three benchmarks. The model trains for ten epochs. Results are averaged over five seeds.
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;One changes &lt;code&gt;three&lt;/code&gt; to &lt;code&gt;four&lt;/code&gt;. The other changes &lt;code&gt;five&lt;/code&gt; to &lt;code&gt;ten&lt;/code&gt;. Merging gives:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;CONFLICT (content): Merge conflict in intro.tex
Automatic merge failed; fix conflicts and then commit the result.
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;Other costs:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;Everyone has to know Git.&lt;/strong&gt; One co-author who has never used it becomes the bottleneck, or the person who emails a file back.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Everyone needs a local TeX installation&lt;/strong&gt;, or access to a build server, to see the PDF.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;No live view.&lt;/strong&gt; You see a co-author's work after they commit and you pull, not while they type.&lt;/li&gt;
&lt;/ul&gt;

&lt;h2&gt;
  
  
  Where a collaborative editor is good
&lt;/h2&gt;

&lt;p&gt;A live editor, such as Overleaf or LetX, removes the setup and the merge step:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;Simultaneous editing.&lt;/strong&gt; Several people edit the same project at once and the edits merge as they type.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;No installation.&lt;/strong&gt; A browser and a link are enough, so a new co-author joins in a minute.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Instant PDF.&lt;/strong&gt; Everyone compiles against the same TeX setup, so "it works on my machine" goes away.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Low skill threshold.&lt;/strong&gt; A senior co-author who only writes prose can contribute without learning a tool.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;It has costs too. History may be shorter than Git's, depending on the plan. Limits on collaborators, compiles and build time depend on the vendor and the tier. And a live editor does not, by itself, give you reproducible figures or CI.&lt;/p&gt;

&lt;h2&gt;
  
  
  The decision table
&lt;/h2&gt;

&lt;div class="table-wrapper-paragraph"&gt;&lt;table&gt;
&lt;thead&gt;
&lt;tr&gt;
&lt;th&gt;Your situation&lt;/th&gt;
&lt;th&gt;Better fit&lt;/th&gt;
&lt;/tr&gt;
&lt;/thead&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td&gt;Figures and tables are generated by code in the same repo&lt;/td&gt;
&lt;td&gt;Git&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Co-authors have mixed technical skill&lt;/td&gt;
&lt;td&gt;Collaborative editor&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Several people write the same sections near the deadline&lt;/td&gt;
&lt;td&gt;Collaborative editor&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;You need a permanent, auditable history&lt;/td&gt;
&lt;td&gt;Git&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;You want CI that compiles on every change&lt;/td&gt;
&lt;td&gt;Git&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Fast onboarding of a new student or collaborator&lt;/td&gt;
&lt;td&gt;Collaborative editor&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Work must stay on your own infrastructure&lt;/td&gt;
&lt;td&gt;Git&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;&lt;/div&gt;

&lt;h2&gt;
  
  
  A setup that uses both
&lt;/h2&gt;

&lt;p&gt;You can keep the editor for writing and Git for the record:&lt;/p&gt;

&lt;ol&gt;
&lt;li&gt;
&lt;strong&gt;Write in the editor.&lt;/strong&gt; Co-authors work on the text together.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Sync to Git.&lt;/strong&gt; Most editors offer some Git or GitHub link. LetX lists GitHub integration on its Pro and Academic Pro plans, and Overleaf lists integrations on its paid plans. Check each vendor's current plan page, since this changes.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Keep code and data in the repository.&lt;/strong&gt; The figures come from scripts, and the paper pulls them in.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Tag the submission.&lt;/strong&gt; Tag the commit you submit, so the exact source is recoverable years later.&lt;/li&gt;
&lt;/ol&gt;

&lt;h2&gt;
  
  
  If you stay with Git alone
&lt;/h2&gt;

&lt;p&gt;Three practices make LaTeX much less painful in Git:&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;1. One sentence per line.&lt;/strong&gt; Git then merges edits to different sentences without a conflict. I tested it on the same example, and the merge succeeded and kept both changes.&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;We evaluate on four benchmarks.
The model trains for ten epochs.
Results are averaged over ten seeds.
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;LaTeX treats a single newline as a space, so the PDF looks the same.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;2. A &lt;code&gt;.gitignore&lt;/code&gt; for build files.&lt;/strong&gt;&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;*.aux
*.log
*.out
*.toc
*.bbl
*.blg
*.fls
*.fdb_latexmk
*.synctex.gz
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;&lt;strong&gt;3. A diff you can read.&lt;/strong&gt; &lt;code&gt;latexdiff&lt;/code&gt; compares two versions and produces a marked-up source, which is easier to review than raw &lt;code&gt;git diff&lt;/code&gt;. I cover it in tomorrow's post on track changes.&lt;/p&gt;

&lt;h2&gt;
  
  
  My view, and my bias
&lt;/h2&gt;

&lt;p&gt;I work on &lt;a href="https://letx.app/?utm_source=devto&amp;amp;utm_medium=post&amp;amp;utm_campaign=latex-collab-series&amp;amp;utm_content=day09" rel="noopener noreferrer"&gt;LetX&lt;/a&gt;, a collaborative LaTeX editor, so weigh that. My honest read is that teams with mixed experience and tight deadlines usually write faster in a live editor, and teams whose papers are tied to code often want Git as the source of truth. Many want both. The free LetX plan allows unlimited collaborators, and its limits (40 compiles a day, the last 15 versions) are on the &lt;a href="https://letx.app/pricing/?utm_source=devto&amp;amp;utm_medium=post&amp;amp;utm_campaign=latex-collab-series&amp;amp;utm_content=day09" rel="noopener noreferrer"&gt;pricing page&lt;/a&gt;.&lt;/p&gt;

&lt;p&gt;If you want the full checklist for choosing a tool, see &lt;a href="https://dev.to/letx_app_76f3cdf1c781dfd7/how-to-choose-a-collaborative-research-writing-tool-a-9-point-checklist-4o2m"&gt;the 9-point checklist&lt;/a&gt;.&lt;/p&gt;

&lt;p&gt;Which does your lab use, Git, an editor, or both? Tell me in the comments.&lt;/p&gt;

</description>
      <category>latex</category>
      <category>git</category>
      <category>research</category>
      <category>productivity</category>
    </item>
    <item>
      <title>Per-Author Comment Macros for Co-Authors in LaTeX</title>
      <dc:creator>letx app</dc:creator>
      <pubDate>Mon, 02 Nov 2026 14:00:00 +0000</pubDate>
      <link>https://dev.to/letx_app_76f3cdf1c781dfd7/per-author-comment-macros-for-co-authors-in-latex-2l7a</link>
      <guid>https://dev.to/letx_app_76f3cdf1c781dfd7/per-author-comment-macros-for-co-authors-in-latex-2l7a</guid>
      <description>&lt;p&gt;The worst place to find a leftover &lt;code&gt;% TODO ask Sam about the licence&lt;/code&gt; is in the PDF you just uploaded to the submission system.&lt;/p&gt;

&lt;p&gt;Comments in a co-authored LaTeX paper need three properties. Each author's notes should be easy to tell apart. All notes should disappear with one change. And a note that is still there on submission day should stop you, not slip through.&lt;/p&gt;

&lt;p&gt;This takes about ten lines of LaTeX. I tested every snippet below.&lt;/p&gt;

&lt;h2&gt;
  
  
  The basic version: one command per author
&lt;/h2&gt;

&lt;p&gt;Put this in a shared &lt;code&gt;macros.tex&lt;/code&gt; that every author loads (see &lt;a href="https://dev.to/letx_app_76f3cdf1c781dfd7/stop-notation-drift-in-co-authored-latex-papers-with-one-macrostex-1nab"&gt;why one macros file matters&lt;/a&gt;):&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight tex"&gt;&lt;code&gt;&lt;span class="c"&gt;% macros.tex&lt;/span&gt;
&lt;span class="k"&gt;\usepackage&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;xcolor&lt;span class="p"&gt;}&lt;/span&gt;
&lt;span class="k"&gt;\newif\ifcomments&lt;/span&gt;
&lt;span class="k"&gt;\commentstrue&lt;/span&gt;            &lt;span class="c"&gt;% use \commentsfalse to hide every note&lt;/span&gt;

&lt;span class="k"&gt;\newcommand&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;&lt;span class="k"&gt;\authnote&lt;/span&gt;&lt;span class="p"&gt;}&lt;/span&gt;[3]&lt;span class="p"&gt;{&lt;/span&gt;&lt;span class="k"&gt;\ifcomments&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;&lt;span class="k"&gt;\color&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;#2&lt;span class="p"&gt;}&lt;/span&gt;[#1: #3]&lt;span class="p"&gt;}&lt;/span&gt;&lt;span class="k"&gt;\fi&lt;/span&gt;&lt;span class="p"&gt;}&lt;/span&gt;
&lt;span class="k"&gt;\newcommand&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;&lt;span class="k"&gt;\ak&lt;/span&gt;&lt;span class="p"&gt;}&lt;/span&gt;[1]&lt;span class="p"&gt;{&lt;/span&gt;&lt;span class="k"&gt;\authnote&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;AK&lt;span class="p"&gt;}{&lt;/span&gt;blue&lt;span class="p"&gt;}{&lt;/span&gt;#1&lt;span class="p"&gt;}}&lt;/span&gt;
&lt;span class="k"&gt;\newcommand&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;&lt;span class="k"&gt;\sm&lt;/span&gt;&lt;span class="p"&gt;}&lt;/span&gt;[1]&lt;span class="p"&gt;{&lt;/span&gt;&lt;span class="k"&gt;\authnote&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;SM&lt;span class="p"&gt;}{&lt;/span&gt;red!70!black&lt;span class="p"&gt;}{&lt;/span&gt;#1&lt;span class="p"&gt;}}&lt;/span&gt;
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;In the text:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight tex"&gt;&lt;code&gt;We evaluate on three benchmarks. &lt;span class="k"&gt;\sm&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;Is the third licence cleared?&lt;span class="p"&gt;}&lt;/span&gt;
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;With &lt;code&gt;\commentstrue&lt;/code&gt;, the PDF shows &lt;code&gt;[SM: Is the third licence cleared?]&lt;/code&gt; in dark red. With &lt;code&gt;\commentsfalse&lt;/code&gt;, the note vanishes and the sentence reads cleanly. Each new author is one more &lt;code&gt;\newcommand&lt;/code&gt; line with their initials and a color.&lt;/p&gt;

&lt;h2&gt;
  
  
  The safer version: fail the build if a note remains
&lt;/h2&gt;

&lt;p&gt;Hiding a note is not the same as resolving it. A hidden note still sits in the source, and the next person to flip the switch brings it back. A strict mode turns any remaining note into a compile error:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight tex"&gt;&lt;code&gt;&lt;span class="k"&gt;\newif\ifcomments&lt;/span&gt; &lt;span class="k"&gt;\commentstrue&lt;/span&gt;
&lt;span class="k"&gt;\newif\ifstrict&lt;/span&gt;   &lt;span class="k"&gt;\strictfalse&lt;/span&gt;   &lt;span class="c"&gt;% use \stricttrue for the submission build&lt;/span&gt;

&lt;span class="k"&gt;\newcommand&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;&lt;span class="k"&gt;\authnote&lt;/span&gt;&lt;span class="p"&gt;}&lt;/span&gt;[3]&lt;span class="p"&gt;{&lt;/span&gt;&lt;span class="c"&gt;%&lt;/span&gt;
  &lt;span class="k"&gt;\ifstrict&lt;/span&gt;
    &lt;span class="k"&gt;\PackageError&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;authnote&lt;span class="p"&gt;}{&lt;/span&gt;Unresolved note by #1: #3&lt;span class="p"&gt;}{&lt;/span&gt;Resolve or delete this note before submitting.&lt;span class="p"&gt;}&lt;/span&gt;&lt;span class="c"&gt;%&lt;/span&gt;
  &lt;span class="k"&gt;\else&lt;/span&gt;
    &lt;span class="k"&gt;\ifcomments&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;&lt;span class="k"&gt;\color&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;#2&lt;span class="p"&gt;}&lt;/span&gt;[#1: #3]&lt;span class="p"&gt;}&lt;/span&gt;&lt;span class="k"&gt;\fi&lt;/span&gt;
  &lt;span class="k"&gt;\fi&lt;/span&gt;&lt;span class="p"&gt;}&lt;/span&gt;
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;With &lt;code&gt;\stricttrue&lt;/code&gt;, compiling the same sentence gives:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;! Package authnote Error: Unresolved note by SM: Is the third licence cleared?.
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;Now the submission checklist has a concrete step: switch to strict, compile, and fix every error until the build is clean. No search needed.&lt;/p&gt;

&lt;h2&gt;
  
  
  Margin notes with todonotes
&lt;/h2&gt;

&lt;p&gt;If you prefer notes in the margin and a list at the end, the &lt;code&gt;todonotes&lt;/code&gt; package does that, and it has a built-in switch:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight tex"&gt;&lt;code&gt;&lt;span class="k"&gt;\usepackage&lt;/span&gt;&lt;span class="na"&gt;[colorinlistoftodos]&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;todonotes&lt;span class="p"&gt;}&lt;/span&gt;   &lt;span class="c"&gt;% notes on&lt;/span&gt;
&lt;span class="c"&gt;% \usepackage[disable]{todonotes}            % notes off&lt;/span&gt;

We evaluate on three benchmarks.&lt;span class="k"&gt;\todo&lt;/span&gt;&lt;span class="na"&gt;[color=blue!20]&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;AK: check the third licence&lt;span class="p"&gt;}&lt;/span&gt;
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;Switching to &lt;code&gt;[disable]&lt;/code&gt; removes the margin notes from the PDF. The package is a good fit for long drafts, because &lt;code&gt;\listoftodos&lt;/code&gt; gives you a single list of everything open. The macros above are lighter and make the strict check easy, so I use them for papers and &lt;code&gt;todonotes&lt;/code&gt; for theses.&lt;/p&gt;

&lt;h2&gt;
  
  
  Three habits that make this work
&lt;/h2&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;Write the note where the problem is.&lt;/strong&gt; A note at the end of the paragraph is easier to act on than one in a separate file.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Resolve by deleting.&lt;/strong&gt; When a note is handled, delete it. Do not leave resolved notes in the source "for history"; version history exists for that.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Agree the colors up front.&lt;/strong&gt; Two authors with similar colors defeats the purpose.&lt;/li&gt;
&lt;/ul&gt;

&lt;h2&gt;
  
  
  If you write together in a live editor
&lt;/h2&gt;

&lt;p&gt;These macros do not depend on any tool, so they work in any setup. If your team edits in the same project at the same time, they also remove the need to reconcile comments after the fact, because each note carries its author's initials. I work on &lt;a href="https://letx.app/?utm_source=devto&amp;amp;utm_medium=post&amp;amp;utm_campaign=latex-collab-series&amp;amp;utm_content=day08" rel="noopener noreferrer"&gt;LetX&lt;/a&gt;, a collaborative LaTeX editor whose free plan allows unlimited collaborators, so a team of five can use five comment commands in one project. The free plan has other limits, such as 40 compiles a day, so check the &lt;a href="https://letx.app/pricing/?utm_source=devto&amp;amp;utm_medium=post&amp;amp;utm_campaign=latex-collab-series&amp;amp;utm_content=day08" rel="noopener noreferrer"&gt;pricing page&lt;/a&gt;.&lt;/p&gt;

&lt;p&gt;Next in this series: Git versus a collaborative editor for papers, and when each one wins. This post is part of &lt;a href="https://dev.to/letx_app_76f3cdf1c781dfd7/how-to-choose-a-collaborative-research-writing-tool-a-9-point-checklist-4o2m"&gt;the series on collaborative research writing&lt;/a&gt;.&lt;/p&gt;

&lt;p&gt;What is the most useful comment you have ever left in a co-authored draft? Tell me in the comments.&lt;/p&gt;

</description>
      <category>latex</category>
      <category>research</category>
      <category>productivity</category>
      <category>tutorial</category>
    </item>
    <item>
      <title>Writing Your Thesis in LaTeX With Your Advisor in the Loop</title>
      <dc:creator>letx app</dc:creator>
      <pubDate>Sun, 01 Nov 2026 14:00:00 +0000</pubDate>
      <link>https://dev.to/letx_app_76f3cdf1c781dfd7/writing-your-thesis-in-latex-with-your-advisor-in-the-loop-1fi4</link>
      <guid>https://dev.to/letx_app_76f3cdf1c781dfd7/writing-your-thesis-in-latex-with-your-advisor-in-the-loop-1fi4</guid>
      <description>&lt;p&gt;A thesis is the longest document most people will ever write in LaTeX, and it is also the one with the most feedback loops. Your advisor reads drafts, sends comments, and expects the next version to address them.&lt;/p&gt;

&lt;p&gt;Most of the pain comes from how the project is organised and how feedback travels. This post is a setup that keeps a long document manageable and makes your advisor's feedback easy to act on.&lt;/p&gt;

&lt;h2&gt;
  
  
  Structure the project as one file per chapter
&lt;/h2&gt;

&lt;p&gt;Use a main file that only assembles chapters. Keep each chapter in its own file:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;thesis/
├── main.tex
├── macros.tex
├── refs.bib
├── chapters/
│   ├── 01-introduction.tex
│   ├── 02-background.tex
│   ├── 03-method.tex
│   ├── 04-results.tex
│   └── 05-conclusion.tex
└── figures/
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;





&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight tex"&gt;&lt;code&gt;&lt;span class="c"&gt;% main.tex&lt;/span&gt;
&lt;span class="k"&gt;\documentclass&lt;/span&gt;&lt;span class="na"&gt;[12pt]&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;report&lt;span class="p"&gt;}&lt;/span&gt;
&lt;span class="k"&gt;\input&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;macros&lt;span class="p"&gt;}&lt;/span&gt;

&lt;span class="nt"&gt;\begin{document}&lt;/span&gt;
&lt;span class="k"&gt;\tableofcontents&lt;/span&gt;

&lt;span class="k"&gt;\include&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;chapters/01-introduction&lt;span class="p"&gt;}&lt;/span&gt;
&lt;span class="k"&gt;\include&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;chapters/02-background&lt;span class="p"&gt;}&lt;/span&gt;
&lt;span class="k"&gt;\include&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;chapters/03-method&lt;span class="p"&gt;}&lt;/span&gt;
&lt;span class="k"&gt;\include&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;chapters/04-results&lt;span class="p"&gt;}&lt;/span&gt;
&lt;span class="k"&gt;\include&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;chapters/05-conclusion&lt;span class="p"&gt;}&lt;/span&gt;

&lt;span class="k"&gt;\bibliographystyle&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;plain&lt;span class="p"&gt;}&lt;/span&gt;
&lt;span class="k"&gt;\bibliography&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;refs&lt;span class="p"&gt;}&lt;/span&gt;
&lt;span class="nt"&gt;\end{document}&lt;/span&gt;
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;Note that &lt;code&gt;\include&lt;/code&gt; starts each chapter on a new page, which is what you want for chapters. Use &lt;code&gt;\input&lt;/code&gt; for pieces that should flow inline. Do not include the &lt;code&gt;.tex&lt;/code&gt; extension.&lt;/p&gt;

&lt;h2&gt;
  
  
  Compile one chapter at a time
&lt;/h2&gt;

&lt;p&gt;A 150-page thesis can take a while to build. &lt;code&gt;\includeonly&lt;/code&gt; lets you compile only the chapter you are working on, while keeping page numbers and cross-references consistent from the previous full build:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight tex"&gt;&lt;code&gt;&lt;span class="c"&gt;% in the preamble, before \begin{document}&lt;/span&gt;
&lt;span class="k"&gt;\includeonly&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;chapters/03-method&lt;span class="p"&gt;}&lt;/span&gt;
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;Run a full build first so the auxiliary files exist, then use &lt;code&gt;\includeonly&lt;/code&gt; for daily work. Remove it, or list every chapter, for the final build. A common slip is leaving it on and wondering why chapters are missing from the PDF.&lt;/p&gt;

&lt;h2&gt;
  
  
  Make feedback easy to give
&lt;/h2&gt;

&lt;p&gt;Your advisor's time is the scarcest resource in the project. Reduce the effort of reviewing.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;Add line numbers to review drafts.&lt;/strong&gt; They let your advisor say "line 214" instead of "the third paragraph on page 41":&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight tex"&gt;&lt;code&gt;&lt;span class="k"&gt;\usepackage&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;lineno&lt;span class="p"&gt;}&lt;/span&gt;
&lt;span class="k"&gt;\linenumbers&lt;/span&gt;
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;Turn them off for the final version.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;Reuse the per-author comment macros.&lt;/strong&gt; The same colour-coded commands from my earlier posts work for advisor notes, with one switch to hide them all:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight tex"&gt;&lt;code&gt;&lt;span class="k"&gt;\usepackage&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;xcolor&lt;span class="p"&gt;}&lt;/span&gt;
&lt;span class="k"&gt;\newif\ifcomments&lt;/span&gt;
&lt;span class="k"&gt;\commentstrue&lt;/span&gt;

&lt;span class="k"&gt;\newcommand&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;&lt;span class="k"&gt;\authnote&lt;/span&gt;&lt;span class="p"&gt;}&lt;/span&gt;[3]&lt;span class="p"&gt;{&lt;/span&gt;&lt;span class="k"&gt;\ifcomments&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;&lt;span class="k"&gt;\color&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;#2&lt;span class="p"&gt;}&lt;/span&gt;[#1: #3]&lt;span class="p"&gt;}&lt;/span&gt;&lt;span class="k"&gt;\fi&lt;/span&gt;&lt;span class="p"&gt;}&lt;/span&gt;
&lt;span class="k"&gt;\newcommand&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;&lt;span class="k"&gt;\adv&lt;/span&gt;&lt;span class="p"&gt;}&lt;/span&gt;[1]&lt;span class="p"&gt;{&lt;/span&gt;&lt;span class="k"&gt;\authnote&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;Advisor&lt;span class="p"&gt;}{&lt;/span&gt;red&lt;span class="p"&gt;}{&lt;/span&gt;#1&lt;span class="p"&gt;}}&lt;/span&gt;
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;&lt;strong&gt;Send one chapter at a time.&lt;/strong&gt; A full thesis is too large to review in one sitting. Agree a rhythm, for example one chapter every two weeks, and send a compiled PDF with a short note on what you want checked: the argument, the structure, or the maths.&lt;/p&gt;

&lt;h2&gt;
  
  
  Snapshot every draft you send
&lt;/h2&gt;

&lt;p&gt;Each time you send a draft to your advisor, save a named snapshot, such as &lt;code&gt;ch3-draft-2026-11-01&lt;/code&gt;. When they reply two weeks later, you will know exactly which version they read, and you can compare it with your current text. This matters when you have kept editing while waiting for feedback.&lt;/p&gt;

&lt;p&gt;A named version in your editor, a Git tag, or a dated copy of the folder all work. What matters is naming it when you send, not afterwards.&lt;/p&gt;

&lt;h2&gt;
  
  
  A review cycle that works
&lt;/h2&gt;

&lt;ol&gt;
&lt;li&gt;
&lt;strong&gt;Draft&lt;/strong&gt; one chapter, with line numbers on.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Snapshot&lt;/strong&gt; and compile the PDF.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Send&lt;/strong&gt; the PDF with a short note on what to look at.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Receive&lt;/strong&gt; comments, either on the PDF or in the source.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Respond&lt;/strong&gt; by resolving each comment, and keep a short list of what you changed and what you decided not to change, with a reason.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Snapshot&lt;/strong&gt; again and send the next chapter.&lt;/li&gt;
&lt;/ol&gt;

&lt;p&gt;Keeping that list of changes is the habit that saves the most time later, because the same comments tend to come back when the full thesis is read together.&lt;/p&gt;

&lt;h2&gt;
  
  
  Work with your advisor inside the project
&lt;/h2&gt;

&lt;p&gt;You can send PDFs, or you can bring your advisor into the project. If your tool allows it, adding your advisor as a collaborator means they can read the source, leave comments in the text, and see changes as they happen. Some advisors prefer a PDF and a pen, so ask what they prefer before changing how they work.&lt;/p&gt;

&lt;p&gt;In &lt;a href="https://letx.app" rel="noopener noreferrer"&gt;LetX&lt;/a&gt;, collaborators are unlimited on the free plan, so inviting an advisor and committee members costs nothing extra. The research-team page mentions a supervisor reviewing drafts as part of a typical lab workflow, but it does not describe dedicated advisor tools, so I would rely on the comment macros above and check the current features yourself. I work on LetX, so test it with your own chapters before you move a whole thesis.&lt;/p&gt;

&lt;h2&gt;
  
  
  Templates, privacy and the final stretch
&lt;/h2&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;University templates.&lt;/strong&gt; Many universities require a specific class or layout. LetX lists &lt;a href="https://letx.app/templates/universities/" rel="noopener noreferrer"&gt;university thesis templates&lt;/a&gt; (87 on the templates page, with 40 linked individually, including MIT, Stanford, Oxford, ETH Zürich, NUS, IIT Bombay, BUET and Dhaka University). Always check your graduate school's current guidelines.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Privacy.&lt;/strong&gt; Projects in LetX are private by default and the site states documents are not used to train AI models, which matters for an unpublished thesis or one under embargo. Read the &lt;a href="https://letx.app/security/" rel="noopener noreferrer"&gt;security page&lt;/a&gt; and decide for yourself.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Final build.&lt;/strong&gt; Remove &lt;code&gt;\includeonly&lt;/code&gt;, turn comments off with &lt;code&gt;\commentsfalse&lt;/code&gt;, remove line numbers, compile from scratch, and read the full PDF for &lt;code&gt;??&lt;/code&gt; references and overfull boxes.&lt;/li&gt;
&lt;/ul&gt;

&lt;h2&gt;
  
  
  Checklist
&lt;/h2&gt;

&lt;ul&gt;
&lt;li&gt;[ ] One file per chapter, assembled with &lt;code&gt;\include&lt;/code&gt;
&lt;/li&gt;
&lt;li&gt;[ ] &lt;code&gt;\includeonly&lt;/code&gt; for daily work, removed for the final build&lt;/li&gt;
&lt;li&gt;[ ] Line numbers on review drafts&lt;/li&gt;
&lt;li&gt;[ ] A comment macro for your advisor, with a global switch&lt;/li&gt;
&lt;li&gt;[ ] A named snapshot every time you send a draft&lt;/li&gt;
&lt;li&gt;[ ] A running list of comments addressed&lt;/li&gt;
&lt;li&gt;[ ] University template and graduate school rules checked&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;What is the best advisor feedback habit you have seen? Tell me in the comments. This wraps up the first week of this series, and I will keep posting daily.&lt;/p&gt;

</description>
      <category>latex</category>
      <category>phd</category>
      <category>academia</category>
      <category>productivity</category>
    </item>
    <item>
      <title>The Best LaTeX Templates for IEEE, ACM, Springer and Nature (and the Traps in Each)</title>
      <dc:creator>letx app</dc:creator>
      <pubDate>Sat, 31 Oct 2026 14:00:00 +0000</pubDate>
      <link>https://dev.to/letx_app_76f3cdf1c781dfd7/the-best-latex-templates-for-ieee-acm-springer-and-nature-and-the-traps-in-each-3m8b</link>
      <guid>https://dev.to/letx_app_76f3cdf1c781dfd7/the-best-latex-templates-for-ieee-acm-springer-and-nature-and-the-traps-in-each-3m8b</guid>
      <description>&lt;p&gt;Your paper's formatting is decided before you write a word: the venue picks the template. The real work is knowing which template to use and which traps each one hides.&lt;/p&gt;

&lt;p&gt;This post covers the four publishers people ask about most: IEEE, ACM, Springer, and Nature. The rule for all four: &lt;strong&gt;start from the publisher's current official template and confirm it on the venue's call for papers.&lt;/strong&gt; Templates change, and some venues use modified versions.&lt;/p&gt;

&lt;h2&gt;
  
  
  IEEE: IEEEtran
&lt;/h2&gt;

&lt;p&gt;IEEE papers use the &lt;code&gt;IEEEtran&lt;/code&gt; class. The version I checked in the IEEEtran how-to is 1.8b. The main options:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight tex"&gt;&lt;code&gt;&lt;span class="c"&gt;% Conference paper&lt;/span&gt;
&lt;span class="k"&gt;\documentclass&lt;/span&gt;&lt;span class="na"&gt;[conference]&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;IEEEtran&lt;span class="p"&gt;}&lt;/span&gt;

&lt;span class="c"&gt;% Journal article&lt;/span&gt;
&lt;span class="k"&gt;\documentclass&lt;/span&gt;&lt;span class="na"&gt;[journal]&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;IEEEtran&lt;span class="p"&gt;}&lt;/span&gt;

&lt;span class="c"&gt;% Computer Society journal&lt;/span&gt;
&lt;span class="k"&gt;\documentclass&lt;/span&gt;&lt;span class="na"&gt;[10pt,journal,compsoc]&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;IEEEtran&lt;span class="p"&gt;}&lt;/span&gt;
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;&lt;strong&gt;Traps:&lt;/strong&gt;&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;Wrong variant.&lt;/strong&gt; Some IEEE publications use modified IEEEtran templates with their own instructions. IEEE's Template Selector is the authoritative source for the current one.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Blank first page.&lt;/strong&gt; Put the closing brace of the &lt;code&gt;\author&lt;/code&gt; group at the end of the last &lt;code&gt;\thanks&lt;/code&gt; group, or a blank first page may appear.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Label placement.&lt;/strong&gt; Do not put &lt;code&gt;\label&lt;/code&gt; before a figure or table &lt;code&gt;\caption&lt;/code&gt;, because the cross-reference will point at the wrong number.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Display math.&lt;/strong&gt; Avoid &lt;code&gt;eqnarray&lt;/code&gt; and &lt;code&gt;$$&lt;/code&gt;; use &lt;code&gt;align&lt;/code&gt; or &lt;code&gt;equation&lt;/code&gt;.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Compose in two columns.&lt;/strong&gt; IEEE recommends writing in the final two-column layout so figures and equations fit.&lt;/li&gt;
&lt;/ul&gt;

&lt;h2&gt;
  
  
  ACM: acmart
&lt;/h2&gt;

&lt;p&gt;ACM uses a single class, &lt;code&gt;acmart&lt;/code&gt;, for both journals and proceedings. The options you will use most:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight tex"&gt;&lt;code&gt;&lt;span class="c"&gt;% Camera-ready conference proceedings (most common)&lt;/span&gt;
&lt;span class="k"&gt;\documentclass&lt;/span&gt;&lt;span class="na"&gt;[sigconf]&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;acmart&lt;span class="p"&gt;}&lt;/span&gt;

&lt;span class="c"&gt;% Review submission, single column&lt;/span&gt;
&lt;span class="k"&gt;\documentclass&lt;/span&gt;&lt;span class="na"&gt;[manuscript,screen,review]&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;acmart&lt;span class="p"&gt;}&lt;/span&gt;

&lt;span class="c"&gt;% Double-anonymous review&lt;/span&gt;
&lt;span class="k"&gt;\documentclass&lt;/span&gt;&lt;span class="na"&gt;[sigconf,anonymous,review]&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;acmart&lt;span class="p"&gt;}&lt;/span&gt;
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;&lt;strong&gt;Traps:&lt;/strong&gt;&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;Do not change the layout.&lt;/strong&gt; acmart forbids modifying margins, font sizes or spacing, and papers that do are sent back for revision.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Fonts.&lt;/strong&gt; The template uses the Libertine font family. Do not load &lt;code&gt;lmodern&lt;/code&gt; or &lt;code&gt;ltimes&lt;/code&gt;.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Placeholder metadata.&lt;/strong&gt; The sample file has placeholder rights information (&lt;code&gt;\setcopyright&lt;/code&gt;, &lt;code&gt;\acmDOI&lt;/code&gt;, &lt;code&gt;\acmConference&lt;/code&gt; and others) and placeholder CCS concepts marked "Do Not Use This Code". Replace them with the values from your rights form and the ACM classification tool.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Accessibility.&lt;/strong&gt; The template supports figure descriptions, and venues increasingly expect them.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Confirm the style.&lt;/strong&gt; Journals use different options such as &lt;code&gt;acmsmall&lt;/code&gt;, &lt;code&gt;acmlarge&lt;/code&gt; or &lt;code&gt;acmtog&lt;/code&gt;, so check which one your venue wants.&lt;/li&gt;
&lt;/ul&gt;

&lt;h2&gt;
  
  
  Springer: two templates, not one
&lt;/h2&gt;

&lt;p&gt;Springer has two different paths, and mixing them up is a common mistake.&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;Journals (including Nature Portfolio and BMC):&lt;/strong&gt; the Springer Nature LaTeX template. The journal package I saw is dated December 2024.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Conference proceedings (including LNCS):&lt;/strong&gt; a separate proceedings template. The Springer Nature page says proceedings contributions should follow the proceedings guidelines, not the journal template.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;&lt;strong&gt;Traps:&lt;/strong&gt;&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;Wrong template for the venue.&lt;/strong&gt; An LNCS paper on the journal template, or the other way round, will be corrected by the production team and cost you time.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Custom fonts.&lt;/strong&gt; Springer's submission guidance says not to use them.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Special characters.&lt;/strong&gt; Convert diacritics to TeX code, for example &lt;code&gt;\"a&lt;/code&gt;.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Folders.&lt;/strong&gt; For their submission systems, put all files in a single directory and use local paths in &lt;code&gt;\includegraphics&lt;/code&gt;.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Bibliography.&lt;/strong&gt; Springer Nature's guidance says the most reliable method for its submission systems is to paste the &lt;code&gt;.bbl&lt;/code&gt; content into your main &lt;code&gt;.tex&lt;/code&gt; file in place of &lt;code&gt;\bibliography{...}&lt;/code&gt;.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Always compile locally&lt;/strong&gt; and fix errors before uploading. Their systems require files that compile with &lt;code&gt;pdflatex&lt;/code&gt;, submitted as a ZIP.&lt;/li&gt;
&lt;/ul&gt;

&lt;h2&gt;
  
  
  Nature: keep it simple
&lt;/h2&gt;

&lt;p&gt;Nature's journals are the opposite of IEEE and ACM. There is no strict journal template to match at first submission.&lt;/p&gt;

&lt;p&gt;From the Springer Nature community guidance for Nature Astronomy:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;Use a simple setup such as the standard &lt;code&gt;article&lt;/code&gt; class, and &lt;strong&gt;avoid templates that need their own class file.&lt;/strong&gt;
&lt;/li&gt;
&lt;li&gt;For the initial submission, a self-contained PDF is simplest.&lt;/li&gt;
&lt;li&gt;Keep the source as a single &lt;code&gt;.tex&lt;/code&gt; file and paste your &lt;code&gt;.bbl&lt;/code&gt; contents into it.&lt;/li&gt;
&lt;li&gt;Use only packages included in a standard installation, since non-standard packages can break compilation.&lt;/li&gt;
&lt;li&gt;Abbreviate journal names in references in ISO-4 style (the &lt;code&gt;jabbrv&lt;/code&gt; package helps).&lt;/li&gt;
&lt;li&gt;If extra figures move to Extended Data or Supplementary Information, redefine the labels with &lt;code&gt;\renewcommand&lt;/code&gt;.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;If your paper goes through Springer Nature's eJP system with the Springer Nature template, their page notes you will need the &lt;code&gt;pdflatex&lt;/code&gt; option in the preamble. Each Nature journal has its own author instructions, so read them before you format anything.&lt;/p&gt;

&lt;h2&gt;
  
  
  A quick comparison
&lt;/h2&gt;

&lt;div class="table-wrapper-paragraph"&gt;&lt;table&gt;
&lt;thead&gt;
&lt;tr&gt;
&lt;th&gt;&lt;/th&gt;
&lt;th&gt;IEEE&lt;/th&gt;
&lt;th&gt;ACM&lt;/th&gt;
&lt;th&gt;Springer&lt;/th&gt;
&lt;th&gt;Nature&lt;/th&gt;
&lt;/tr&gt;
&lt;/thead&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td&gt;Class&lt;/td&gt;
&lt;td&gt;&lt;code&gt;IEEEtran&lt;/code&gt;&lt;/td&gt;
&lt;td&gt;&lt;code&gt;acmart&lt;/code&gt;&lt;/td&gt;
&lt;td&gt;Journal template or proceedings template&lt;/td&gt;
&lt;td&gt;Plain &lt;code&gt;article&lt;/code&gt; is fine&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Strictness&lt;/td&gt;
&lt;td&gt;High, with variants&lt;/td&gt;
&lt;td&gt;Very high, layout locked&lt;/td&gt;
&lt;td&gt;High&lt;/td&gt;
&lt;td&gt;Low at first submission&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Layout edits allowed&lt;/td&gt;
&lt;td&gt;Little&lt;/td&gt;
&lt;td&gt;No&lt;/td&gt;
&lt;td&gt;Little&lt;/td&gt;
&lt;td&gt;Not applicable&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Biggest trap&lt;/td&gt;
&lt;td&gt;Using the wrong variant&lt;/td&gt;
&lt;td&gt;Editing the layout or placeholders&lt;/td&gt;
&lt;td&gt;Mixing journal and proceedings templates&lt;/td&gt;
&lt;td&gt;Fancy classes and packages&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;&lt;/div&gt;

&lt;h2&gt;
  
  
  Starting from a working template
&lt;/h2&gt;

&lt;p&gt;Setting up a template by hand is where hours disappear. &lt;a href="https://letx.app" rel="noopener noreferrer"&gt;LetX&lt;/a&gt; has a &lt;a href="https://letx.app/templates/" rel="noopener noreferrer"&gt;template library&lt;/a&gt; with 1000+ templates. Its conference category includes IEEE, ACM sigconf and Springer LNCS, and its journal category includes IEEE, ACM, Springer, Elsevier, APS RevTeX and Nature. The free plan includes all of them.&lt;/p&gt;

&lt;p&gt;I work on LetX, so treat that as a convenience and not a substitute for the checks above. Whichever tool you use, &lt;strong&gt;download the current template from the publisher, or confirm that the copy you use is current&lt;/strong&gt;, because templates are updated.&lt;/p&gt;

&lt;h2&gt;
  
  
  Pre-submission template check
&lt;/h2&gt;

&lt;ul&gt;
&lt;li&gt;[ ] Template confirmed against the venue's call for papers&lt;/li&gt;
&lt;li&gt;[ ] Right variant: conference or journal, proceedings or journal template&lt;/li&gt;
&lt;li&gt;[ ] No layout edits where they are forbidden&lt;/li&gt;
&lt;li&gt;[ ] Placeholder metadata replaced&lt;/li&gt;
&lt;li&gt;[ ] Compiles cleanly from scratch, with no &lt;code&gt;??&lt;/code&gt; in the PDF&lt;/li&gt;
&lt;li&gt;[ ] Bibliography handled the way the submission system wants&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;Which template has given you the most trouble? Tell me in the comments. Tomorrow: writing your thesis in LaTeX with your advisor in the loop.&lt;/p&gt;

</description>
      <category>latex</category>
      <category>research</category>
      <category>academia</category>
      <category>tutorial</category>
    </item>
    <item>
      <title>Fixing LaTeX Compile Errors With an AI Assistant (and When It Gets It Wrong)</title>
      <dc:creator>letx app</dc:creator>
      <pubDate>Fri, 30 Oct 2026 14:00:00 +0000</pubDate>
      <link>https://dev.to/letx_app_76f3cdf1c781dfd7/fixing-latex-compile-errors-with-an-ai-assistant-and-when-it-gets-it-wrong-2p85</link>
      <guid>https://dev.to/letx_app_76f3cdf1c781dfd7/fixing-latex-compile-errors-with-an-ai-assistant-and-when-it-gets-it-wrong-2p85</guid>
      <description>&lt;p&gt;"Undefined control sequence." "Missing $ inserted." "Emergency stop."&lt;/p&gt;

&lt;p&gt;LaTeX error messages are famous for being cryptic. The good news is that a small number of errors cause most of the pain, and an AI assistant can fix many of them, provided it can see your files and your compile log. This post covers how to read the log yourself, the six errors you will meet most, and what AI help can and cannot do.&lt;/p&gt;

&lt;h2&gt;
  
  
  Read the log in three parts
&lt;/h2&gt;

&lt;p&gt;Every LaTeX error has the same shape:&lt;/p&gt;

&lt;ol&gt;
&lt;li&gt;
&lt;strong&gt;The message&lt;/strong&gt;, which starts with &lt;code&gt;!&lt;/code&gt; or names the file and line, for example &lt;code&gt;Undefined control sequence.&lt;/code&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;The line number&lt;/strong&gt;, for example &lt;code&gt;./main.tex:3:&lt;/code&gt;.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;The context&lt;/strong&gt;, a snippet showing where LaTeX stopped.&lt;/li&gt;
&lt;/ol&gt;

&lt;p&gt;Compile with &lt;code&gt;-file-line-error&lt;/code&gt; to get the file and line in one place:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight shell"&gt;&lt;code&gt;pdflatex &lt;span class="nt"&gt;-interaction&lt;/span&gt;&lt;span class="o"&gt;=&lt;/span&gt;nonstopmode &lt;span class="nt"&gt;-file-line-error&lt;/span&gt; main.tex
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;Always fix the &lt;strong&gt;first&lt;/strong&gt; error. Later ones are often just fallout from it.&lt;/p&gt;

&lt;h2&gt;
  
  
  The six errors you will see most
&lt;/h2&gt;

&lt;p&gt;I reproduced each of these with a minimal file so the messages below are real.&lt;/p&gt;

&lt;h3&gt;
  
  
  1. Missing $ inserted
&lt;/h3&gt;



&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight tex"&gt;&lt;code&gt;The loss is x&lt;span class="p"&gt;_&lt;/span&gt;i&lt;span class="p"&gt;^&lt;/span&gt;2 here.
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;





&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;./main.tex:3: Missing $ inserted.
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;Subscripts and superscripts only work in math mode. Fix: write &lt;code&gt;$x_i^2$&lt;/code&gt;.&lt;/p&gt;

&lt;h3&gt;
  
  
  2. Undefined control sequence
&lt;/h3&gt;



&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight tex"&gt;&lt;code&gt;&lt;span class="k"&gt;\includegraphcs&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;a&lt;span class="p"&gt;}&lt;/span&gt;
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;





&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;./main.tex:3: Undefined control sequence.
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;Usually a typo in a command name, or a command from a package you did not load. Fix: correct the spelling, or add the missing &lt;code&gt;\usepackage{...}&lt;/code&gt;.&lt;/p&gt;

&lt;h3&gt;
  
  
  3. File not found
&lt;/h3&gt;



&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight tex"&gt;&lt;code&gt;&lt;span class="k"&gt;\usepackage&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;boooktabs&lt;span class="p"&gt;}&lt;/span&gt;
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;





&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;! LaTeX Error: File `boooktabs.sty' not found.
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;Misspelled package name or a missing file. This one is fatal, so no PDF is produced. Fix: check the spelling, or add the file to the project.&lt;/p&gt;

&lt;h3&gt;
  
  
  4. Environment mismatch
&lt;/h3&gt;



&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight tex"&gt;&lt;code&gt;&lt;span class="nt"&gt;\begin{itemize}&lt;/span&gt;
&lt;span class="k"&gt;\item&lt;/span&gt; a
&lt;span class="nt"&gt;\end{enumerate}&lt;/span&gt;
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;





&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;LaTeX Error: \begin{itemize} on input line 3 ended by \end{enumerate}.
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;An &lt;code&gt;\end&lt;/code&gt; that does not match its &lt;code&gt;\begin&lt;/code&gt;. Fix: make them match, and check for a missing &lt;code&gt;\end&lt;/code&gt; earlier in the file.&lt;/p&gt;

&lt;h3&gt;
  
  
  5. File ended while scanning use of \textbf
&lt;/h3&gt;



&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight tex"&gt;&lt;code&gt;&lt;span class="k"&gt;\textbf&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;bold text
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;





&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;! File ended while scanning use of \textbf .
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;An unclosed brace. The line number points at the end of the file, not at the real mistake, so search backwards for the opening &lt;code&gt;{&lt;/code&gt; that never closes.&lt;/p&gt;

&lt;h3&gt;
  
  
  6. Unclosed brace inside math
&lt;/h3&gt;



&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight tex"&gt;&lt;code&gt;a = &lt;span class="k"&gt;\frac&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;1&lt;span class="p"&gt;}{&lt;/span&gt;2
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;





&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;! File ended while scanning use of \frac .
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;Same cause as above. Editors with brace matching make these much faster to find.&lt;/p&gt;

&lt;h2&gt;
  
  
  Why a plain chat assistant often falls short
&lt;/h2&gt;

&lt;p&gt;You can paste an error into a general chat assistant and get a sensible answer. The limits are practical:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;It does not see your whole project, so it cannot tell that a macro is defined in another file.&lt;/li&gt;
&lt;li&gt;It cannot compile. It proposes a fix and you find out whether it worked by trying it.&lt;/li&gt;
&lt;li&gt;Each round trip costs copy, paste, and compile.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;For simple typos that is fine. For errors caused by interactions between packages, files and bibliography, the missing context matters.&lt;/p&gt;

&lt;h2&gt;
  
  
  What an AI assistant needs to fix LaTeX well
&lt;/h2&gt;

&lt;p&gt;A useful assistant needs three things:&lt;/p&gt;

&lt;ol&gt;
&lt;li&gt;
&lt;strong&gt;The project&lt;/strong&gt;, including every &lt;code&gt;.tex&lt;/code&gt; file and the &lt;code&gt;.bib&lt;/code&gt; file.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;The real compile log&lt;/strong&gt;, not a pasted fragment.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;The ability to recompile&lt;/strong&gt;, so it can check its own fix.&lt;/li&gt;
&lt;/ol&gt;

&lt;p&gt;That is the idea behind Lexi, the assistant built into &lt;a href="https://letx.app" rel="noopener noreferrer"&gt;LetX&lt;/a&gt;. According to the LetX site, it reads the whole project and the compile log, edits the source, and recompiles the PDF. The command &lt;code&gt;/fix&lt;/code&gt; runs that loop. Examples on the Lexi page include a mismatched list environment, a missing &lt;code&gt;xcolor&lt;/code&gt; package, and converting &lt;code&gt;\hline&lt;/code&gt; tables to &lt;code&gt;booktabs&lt;/code&gt;.&lt;/p&gt;

&lt;p&gt;Lexi has other commands beyond fixing errors, such as &lt;code&gt;/table&lt;/code&gt; for a compiling booktabs table, &lt;code&gt;/cite&lt;/code&gt;, &lt;code&gt;/shorten&lt;/code&gt;, &lt;code&gt;/convert&lt;/code&gt; for retargeting a paper to another venue, and &lt;code&gt;/slides&lt;/code&gt; for turning a paper into a Beamer deck. Type &lt;code&gt;/&lt;/code&gt; in the panel, or select text and press Cmd-K to edit only that passage.&lt;/p&gt;

&lt;h2&gt;
  
  
  Its limits, and how to use it safely
&lt;/h2&gt;

&lt;p&gt;The Lexi page itself reports that one of its five test tasks, repairing a broken bibliography, did not pass its check. That is a useful reminder for any AI tool:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;Read the diff.&lt;/strong&gt; Look at what changed, especially in the bibliography and in math.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Check the PDF&lt;/strong&gt;, not just that it compiled. A document can compile and still be wrong.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Keep a restore point.&lt;/strong&gt; LetX applies edits to the open document and keeps the previous version in history, one click away. Use that, or your own snapshot, before any large AI edit.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Fix the first error first&lt;/strong&gt;, whether you or an AI does it.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;Lexi is a Pro feature with a usage allowance. The free plan includes the editor, templates, collaboration and 40 compiles per day. Check the &lt;a href="https://letx.app/lexi/" rel="noopener noreferrer"&gt;Lexi page&lt;/a&gt; and the &lt;a href="https://letx.app/pricing/" rel="noopener noreferrer"&gt;pricing page&lt;/a&gt; for the current allowance. LetX also states that user documents are never used to train AI models, which matters if your paper is unpublished.&lt;/p&gt;

&lt;p&gt;If you prefer using Claude or ChatGPT, LetX also offers an MCP server that lets those assistants search templates, edit files and compile PDFs in a project.&lt;/p&gt;

&lt;h2&gt;
  
  
  A debugging routine that works with or without AI
&lt;/h2&gt;

&lt;ol&gt;
&lt;li&gt;Compile and read the &lt;strong&gt;first&lt;/strong&gt; error only.&lt;/li&gt;
&lt;li&gt;Open the file and line it names.&lt;/li&gt;
&lt;li&gt;Look backwards from that line for the real cause.&lt;/li&gt;
&lt;li&gt;Fix one thing, recompile, repeat.&lt;/li&gt;
&lt;li&gt;Run a final check for &lt;code&gt;??&lt;/code&gt; in the PDF, and for warnings in the log.&lt;/li&gt;
&lt;/ol&gt;

&lt;p&gt;Which LaTeX error cost you the most time? Tell me in the comments. Tomorrow: the best LaTeX templates for IEEE, ACM, Springer and Nature.&lt;/p&gt;

</description>
      <category>latex</category>
      <category>ai</category>
      <category>productivity</category>
      <category>tutorial</category>
    </item>
    <item>
      <title>How to Write a Research Paper With Co-Authors in 6 Steps</title>
      <dc:creator>letx app</dc:creator>
      <pubDate>Thu, 29 Oct 2026 14:00:00 +0000</pubDate>
      <link>https://dev.to/letx_app_76f3cdf1c781dfd7/how-to-write-a-research-paper-with-co-authors-in-6-steps-1a7a</link>
      <guid>https://dev.to/letx_app_76f3cdf1c781dfd7/how-to-write-a-research-paper-with-co-authors-in-6-steps-1a7a</guid>
      <description>&lt;p&gt;Writing a paper with co-authors fails in predictable ways. Someone overwrites someone else's paragraph. Nobody knows which version is current. Two people use different notation. Then the last week turns into a scramble.&lt;/p&gt;

&lt;p&gt;These six steps prevent most of that. They work in any editor, and I use LaTeX in the examples.&lt;/p&gt;

&lt;h2&gt;
  
  
  Step 1: Agree on the outline and the owners
&lt;/h2&gt;

&lt;p&gt;Before anyone writes a sentence, agree on two things.&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;The outline:&lt;/strong&gt; the section headings and a one-line goal for each section.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;The owners:&lt;/strong&gt; one person responsible for each section.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;An owner is not the only person who may edit a section. An owner is the person who decides what goes in it and who resolves disagreements. Everyone else suggests changes through comments.&lt;/p&gt;

&lt;p&gt;A table in a shared note is enough:&lt;/p&gt;

&lt;div class="table-wrapper-paragraph"&gt;&lt;table&gt;
&lt;thead&gt;
&lt;tr&gt;
&lt;th&gt;Section&lt;/th&gt;
&lt;th&gt;Owner&lt;/th&gt;
&lt;th&gt;First draft due&lt;/th&gt;
&lt;/tr&gt;
&lt;/thead&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td&gt;Abstract&lt;/td&gt;
&lt;td&gt;Sam&lt;/td&gt;
&lt;td&gt;Week 5&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Introduction&lt;/td&gt;
&lt;td&gt;Sam&lt;/td&gt;
&lt;td&gt;Week 3&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Method&lt;/td&gt;
&lt;td&gt;Priya&lt;/td&gt;
&lt;td&gt;Week 2&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Experiments&lt;/td&gt;
&lt;td&gt;Lee&lt;/td&gt;
&lt;td&gt;Week 3&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;&lt;/div&gt;

&lt;h2&gt;
  
  
  Step 2: Give every section its own file
&lt;/h2&gt;

&lt;p&gt;One giant &lt;code&gt;main.tex&lt;/code&gt; causes most edit collisions. Split the paper so &lt;code&gt;main.tex&lt;/code&gt; only assembles the pieces:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight tex"&gt;&lt;code&gt;&lt;span class="c"&gt;% main.tex&lt;/span&gt;
&lt;span class="k"&gt;\documentclass&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;article&lt;span class="p"&gt;}&lt;/span&gt;
&lt;span class="k"&gt;\input&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;macros&lt;span class="p"&gt;}&lt;/span&gt;

&lt;span class="nt"&gt;\begin{document}&lt;/span&gt;
&lt;span class="k"&gt;\input&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;sections/abstract&lt;span class="p"&gt;}&lt;/span&gt;
&lt;span class="k"&gt;\input&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;sections/intro&lt;span class="p"&gt;}&lt;/span&gt;
&lt;span class="k"&gt;\input&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;sections/method&lt;span class="p"&gt;}&lt;/span&gt;
&lt;span class="k"&gt;\input&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;sections/experiments&lt;span class="p"&gt;}&lt;/span&gt;
&lt;span class="k"&gt;\input&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;sections/conclusion&lt;span class="p"&gt;}&lt;/span&gt;

&lt;span class="k"&gt;\bibliographystyle&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;plain&lt;span class="p"&gt;}&lt;/span&gt;
&lt;span class="k"&gt;\bibliography&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;refs&lt;span class="p"&gt;}&lt;/span&gt;
&lt;span class="nt"&gt;\end{document}&lt;/span&gt;
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;Now two people can work on &lt;code&gt;method.tex&lt;/code&gt; and &lt;code&gt;experiments.tex&lt;/code&gt; without touching the same lines.&lt;/p&gt;

&lt;h2&gt;
  
  
  Step 3: Keep notation in one &lt;code&gt;macros.tex&lt;/code&gt;
&lt;/h2&gt;

&lt;p&gt;Put every shared symbol in one file, and make one rule: nobody defines a macro inside a section file. This stops the same vector appearing in three different styles. I wrote this up in &lt;a href="https://dev.to/letx_app_76f3cdf1c781dfd7/stop-notation-drift-in-co-authored-latex-papers-with-one-macrostex-1nab"&gt;Stop Notation Drift in Co-Authored LaTeX Papers With One macros.tex&lt;/a&gt;.&lt;/p&gt;

&lt;h2&gt;
  
  
  Step 4: Give every author a comment command
&lt;/h2&gt;

&lt;p&gt;Comments like &lt;code&gt;% TODO fix this&lt;/code&gt; get lost. Give each author a color and a command, plus one switch that hides everything for the final build:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight tex"&gt;&lt;code&gt;&lt;span class="c"&gt;% in macros.tex&lt;/span&gt;
&lt;span class="k"&gt;\usepackage&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;xcolor&lt;span class="p"&gt;}&lt;/span&gt;
&lt;span class="k"&gt;\newif\ifcomments&lt;/span&gt;
&lt;span class="k"&gt;\commentstrue&lt;/span&gt;   &lt;span class="c"&gt;% use \commentsfalse for the submission build&lt;/span&gt;

&lt;span class="k"&gt;\newcommand&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;&lt;span class="k"&gt;\authnote&lt;/span&gt;&lt;span class="p"&gt;}&lt;/span&gt;[3]&lt;span class="p"&gt;{&lt;/span&gt;&lt;span class="k"&gt;\ifcomments&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;&lt;span class="k"&gt;\color&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;#2&lt;span class="p"&gt;}&lt;/span&gt;[#1: #3]&lt;span class="p"&gt;}&lt;/span&gt;&lt;span class="k"&gt;\fi&lt;/span&gt;&lt;span class="p"&gt;}&lt;/span&gt;
&lt;span class="k"&gt;\newcommand&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;&lt;span class="k"&gt;\sm&lt;/span&gt;&lt;span class="p"&gt;}&lt;/span&gt;[1]&lt;span class="p"&gt;{&lt;/span&gt;&lt;span class="k"&gt;\authnote&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;SM&lt;span class="p"&gt;}{&lt;/span&gt;red&lt;span class="p"&gt;}{&lt;/span&gt;#1&lt;span class="p"&gt;}}&lt;/span&gt;
&lt;span class="k"&gt;\newcommand&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;&lt;span class="k"&gt;\pr&lt;/span&gt;&lt;span class="p"&gt;}&lt;/span&gt;[1]&lt;span class="p"&gt;{&lt;/span&gt;&lt;span class="k"&gt;\authnote&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;PR&lt;span class="p"&gt;}{&lt;/span&gt;blue&lt;span class="p"&gt;}{&lt;/span&gt;#1&lt;span class="p"&gt;}}&lt;/span&gt;
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;In the text:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight tex"&gt;&lt;code&gt;We evaluate on three benchmarks. &lt;span class="k"&gt;\pr&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;Is the third one's licence cleared?&lt;span class="p"&gt;}&lt;/span&gt;
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;Notes appear in the PDF while drafting, and one line removes all of them before submission. Search the project for &lt;code&gt;\sm{&lt;/code&gt; and &lt;code&gt;\pr{&lt;/code&gt; at the end to catch strays.&lt;/p&gt;

&lt;h2&gt;
  
  
  Step 5: Snapshot before every big rewrite
&lt;/h2&gt;

&lt;p&gt;Large rewrites are where work gets lost. Before anyone restructures a section, save a version you can return to. That can be a Git commit, a named version in your editor, or a copy of the folder. Name it by intent, like "before restructuring experiments", not "v12".&lt;/p&gt;

&lt;p&gt;Check how much history your tool keeps before deadline week. Some free plans keep a limited number of versions or a short time window.&lt;/p&gt;

&lt;h2&gt;
  
  
  Step 6: Freeze one week before the deadline
&lt;/h2&gt;

&lt;p&gt;The final week is where organized teams fall apart. A simple schedule helps:&lt;/p&gt;

&lt;ol&gt;
&lt;li&gt;
&lt;strong&gt;Day -7:&lt;/strong&gt; Content freeze. Fixes only, no new claims or experiments.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Day -5:&lt;/strong&gt; One person resolves all comments and switches &lt;code&gt;\commentsfalse&lt;/code&gt;.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Day -3:&lt;/strong&gt; Build the final PDF and check the page limit. Fix length with layout changes, not last-minute cuts to results.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Day -2:&lt;/strong&gt; Everyone reads the compiled PDF, not the source. Look for &lt;code&gt;??&lt;/code&gt; references, overfull boxes and incomplete bibliography entries.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Day -1:&lt;/strong&gt; Upload to the submission system and test the upload.&lt;/li&gt;
&lt;/ol&gt;

&lt;p&gt;Reading the PDF matters more than it sounds. The source hides problems that a reader sees in seconds.&lt;/p&gt;

&lt;h2&gt;
  
  
  Picking a tool for these steps
&lt;/h2&gt;

&lt;p&gt;These steps work with email and Git, but they are easier when everyone edits the same project live. With a live editor, there is no "which version is current" problem, and a one-file-per-section structure keeps edits from stepping on each other.&lt;/p&gt;

&lt;p&gt;If you want to try it, &lt;a href="https://letx.app" rel="noopener noreferrer"&gt;LetX&lt;/a&gt; is a collaborative LaTeX editor with unlimited collaborators on its free plan (it caps free compiles at 40 per day). You can import an existing project as a ZIP and invite co-authors with a link. I work on it, so see yesterday's post for &lt;a href="https://dev.to/letx_app_76f3cdf1c781dfd7/letx-vs-overleaf-a-free-overleaf-alternative-with-unlimited-collaborators-1495"&gt;how it compares with Overleaf&lt;/a&gt; and test it on a real paper before committing.&lt;/p&gt;

&lt;h2&gt;
  
  
  A checklist to copy
&lt;/h2&gt;

&lt;ul&gt;
&lt;li&gt;[ ] Outline and owners agreed before writing&lt;/li&gt;
&lt;li&gt;[ ] One file per section&lt;/li&gt;
&lt;li&gt;[ ] One shared &lt;code&gt;macros.tex&lt;/code&gt;
&lt;/li&gt;
&lt;li&gt;[ ] One comment command per author, with a switch to hide them&lt;/li&gt;
&lt;li&gt;[ ] A named snapshot before every big rewrite&lt;/li&gt;
&lt;li&gt;[ ] A content freeze one week out&lt;/li&gt;
&lt;li&gt;[ ] Everyone reads the compiled PDF before submission&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;Which step does your team skip? Tell me in the comments. Tomorrow: fixing LaTeX compile errors with an AI assistant.&lt;/p&gt;

</description>
      <category>latex</category>
      <category>research</category>
      <category>productivity</category>
      <category>collaboration</category>
    </item>
    <item>
      <title>LetX vs Overleaf: A Free Overleaf Alternative With Unlimited Collaborators</title>
      <dc:creator>letx app</dc:creator>
      <pubDate>Wed, 28 Oct 2026 14:00:00 +0000</pubDate>
      <link>https://dev.to/letx_app_76f3cdf1c781dfd7/letx-vs-overleaf-a-free-overleaf-alternative-with-unlimited-collaborators-1495</link>
      <guid>https://dev.to/letx_app_76f3cdf1c781dfd7/letx-vs-overleaf-a-free-overleaf-alternative-with-unlimited-collaborators-1495</guid>
      <description>&lt;p&gt;If you are searching for an Overleaf alternative, you are probably in one of three situations. You have a paper with three or more authors and hit the free plan's collaborator limit. You lost version history you needed. Or you are paying per seat for a lab and the total looks high.&lt;/p&gt;

&lt;p&gt;This post compares the two tools on the nine points from &lt;a href="https://dev.to/letx_app_76f3cdf1c781dfd7/how-to-choose-a-collaborative-research-writing-tool-a-9-point-checklist-4o2m"&gt;yesterday's checklist&lt;/a&gt;. I work on LetX, so I am biased, and I have tried to say where Overleaf is the better choice. All numbers below come from each vendor's own pages as I read them on 26 October 2026. Plans change, so check both pricing pages before you decide.&lt;/p&gt;

&lt;h2&gt;
  
  
  The short version
&lt;/h2&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;Use LetX if&lt;/strong&gt; you write in teams of three or more, want unlimited collaborators on a free plan, and care about compile speed.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Use Overleaf if&lt;/strong&gt; you work alone or with one co-author, need its paid features like track changes, or your institution already provides it.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Use neither&lt;/strong&gt; if your team is happy in a local setup with Git and has the skills to maintain it.&lt;/li&gt;
&lt;/ul&gt;

&lt;h2&gt;
  
  
  Free plans side by side
&lt;/h2&gt;

&lt;div class="table-wrapper-paragraph"&gt;&lt;table&gt;
&lt;thead&gt;
&lt;tr&gt;
&lt;th&gt;&lt;/th&gt;
&lt;th&gt;Overleaf Free&lt;/th&gt;
&lt;th&gt;LetX Free&lt;/th&gt;
&lt;/tr&gt;
&lt;/thead&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td&gt;Collaborators per project&lt;/td&gt;
&lt;td&gt;One collaborator&lt;/td&gt;
&lt;td&gt;Unlimited&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Projects&lt;/td&gt;
&lt;td&gt;Not stated on the plans page&lt;/td&gt;
&lt;td&gt;Unlimited&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Compiles&lt;/td&gt;
&lt;td&gt;Limited by compile timeout&lt;/td&gt;
&lt;td&gt;40 per day, builds up to 2 minutes&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Version history&lt;/td&gt;
&lt;td&gt;Previous 24 hours&lt;/td&gt;
&lt;td&gt;Last 15 versions&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Templates&lt;/td&gt;
&lt;td&gt;Included&lt;/td&gt;
&lt;td&gt;All 1000+ included&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;AI assistant&lt;/td&gt;
&lt;td&gt;5 uses per day on the plans page&lt;/td&gt;
&lt;td&gt;Lexi is a Pro feature&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Price&lt;/td&gt;
&lt;td&gt;$0&lt;/td&gt;
&lt;td&gt;$0&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;&lt;/div&gt;

&lt;p&gt;Two things stand out. LetX's free plan has no collaborator cap but does cap compiles at 40 per day. Overleaf's free plan caps collaborators but, as far as the plans page states, does not give a daily compile count. If your team compiles constantly, the 40-per-day limit is the number to watch.&lt;/p&gt;

&lt;h2&gt;
  
  
  Paid plans side by side
&lt;/h2&gt;

&lt;div class="table-wrapper-paragraph"&gt;&lt;table&gt;
&lt;thead&gt;
&lt;tr&gt;
&lt;th&gt;&lt;/th&gt;
&lt;th&gt;Overleaf&lt;/th&gt;
&lt;th&gt;LetX&lt;/th&gt;
&lt;/tr&gt;
&lt;/thead&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td&gt;Student / academic&lt;/td&gt;
&lt;td&gt;$8.25 per month billed yearly ($13 monthly), 10 collaborators per project&lt;/td&gt;
&lt;td&gt;Academic Pro $9 per month or $90 per year, unlimited collaborators&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Mid tier&lt;/td&gt;
&lt;td&gt;Standard, $16.75 per month billed yearly ($25 monthly), 10 collaborators&lt;/td&gt;
&lt;td&gt;n/a&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Top tier&lt;/td&gt;
&lt;td&gt;Pro, $33.25 per month billed yearly ($45 monthly), unlimited collaborators, SSO and admin features&lt;/td&gt;
&lt;td&gt;Pro $18 per month or $180 per year, unlimited collaborators&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Unlimited compiles&lt;/td&gt;
&lt;td&gt;Paid plans extend the compile timeout&lt;/td&gt;
&lt;td&gt;Included in Pro and Academic Pro&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Full version history&lt;/td&gt;
&lt;td&gt;Paid plans&lt;/td&gt;
&lt;td&gt;Pro and Academic Pro&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;GitHub integration&lt;/td&gt;
&lt;td&gt;Paid plans (integrations)&lt;/td&gt;
&lt;td&gt;Pro and Academic Pro&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;&lt;/div&gt;

&lt;p&gt;A few notes on reading this table fairly:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;Overleaf's student plan requires enrollment at a degree-granting institution. LetX's academic plan covers students, faculty and university staff, and verification asks for a transcript or ID document.&lt;/li&gt;
&lt;li&gt;LetX's Academic Pro includes everything in Pro except that its monthly Lexi AI allowance is smaller.&lt;/li&gt;
&lt;li&gt;Overleaf's higher tiers add things like SSO and audit logging that matter to institutions. Compare against what your department needs, not against the headline price.&lt;/li&gt;
&lt;/ul&gt;

&lt;h2&gt;
  
  
  Where each one wins
&lt;/h2&gt;

&lt;p&gt;&lt;strong&gt;Overleaf is strong when:&lt;/strong&gt;&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;You work alone or with one co-author, so the collaborator cap never bites.&lt;/li&gt;
&lt;li&gt;You need track changes or its other paid review features.&lt;/li&gt;
&lt;li&gt;Your university already pays for it, which makes the price irrelevant.&lt;/li&gt;
&lt;li&gt;You want the largest existing ecosystem of tutorials and community answers.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;&lt;strong&gt;LetX is strong when:&lt;/strong&gt;&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;Your paper has three or more authors and you do not want to pay just to add a collaborator.&lt;/li&gt;
&lt;li&gt;You want live editing on a CRDT (Yjs), which merges simultaneous edits instead of locking or conflicting.&lt;/li&gt;
&lt;li&gt;You want a free plan that includes the whole template library.&lt;/li&gt;
&lt;li&gt;You want an AI assistant that reads the project and the compile log (Lexi, on Pro). I cover it in a later post.&lt;/li&gt;
&lt;/ul&gt;

&lt;h2&gt;
  
  
  Switching takes about two minutes
&lt;/h2&gt;

&lt;p&gt;You do not need to commit before you try it:&lt;/p&gt;

&lt;ol&gt;
&lt;li&gt;In Overleaf, download your project as source (the ZIP).&lt;/li&gt;
&lt;li&gt;In LetX, create a project by uploading that ZIP.&lt;/li&gt;
&lt;li&gt;Compile and check that images, &lt;code&gt;.bib&lt;/code&gt; files and folder structure came across.&lt;/li&gt;
&lt;li&gt;Invite a co-author and edit the same section at the same time.&lt;/li&gt;
&lt;/ol&gt;

&lt;p&gt;LetX runs standard TeX Live, so most projects compile without changes. If something does not, check packages and engine settings first.&lt;/p&gt;

&lt;h2&gt;
  
  
  What I would test before switching
&lt;/h2&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;Your slowest document.&lt;/strong&gt; Compile the longest paper or thesis you have and compare times on both tools.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;A real co-author conflict.&lt;/strong&gt; Two people, one paragraph, at the same time.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;The deadline scenario.&lt;/strong&gt; Count how many compiles you actually do in a day. If it is near 40, price Pro in.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;For the full list of claims and caveats, I keep a &lt;a href="https://letx.app/vs/overleaf/" rel="noopener noreferrer"&gt;LetX vs Overleaf comparison page&lt;/a&gt;, and a longer post on &lt;a href="https://letx.app/blog/overleaf-free-plan-limits-2026/" rel="noopener noreferrer"&gt;Overleaf's free plan limits&lt;/a&gt;. Try it free at &lt;a href="https://letx.app" rel="noopener noreferrer"&gt;letx.app&lt;/a&gt;.&lt;/p&gt;

&lt;p&gt;If you have switched in either direction, I would like to hear what surprised you. Tomorrow: how to write a paper with co-authors in six steps.&lt;/p&gt;

</description>
      <category>latex</category>
      <category>research</category>
      <category>productivity</category>
      <category>webdev</category>
    </item>
  </channel>
</rss>
