Migrating my resume from latex to HTML
New year, new resume? Join me as I rewrite my resume in HTML instead of \(\rm\LaTeX\)
Since my sophomore year or so, my resume has been in latex. The history is a bit murky because I apparently didn’t start tracking my resume source in git until well after college, but I have a vague recollection of how it all started. Up until that point, I’d been making my resumes in Microsoft Word, but I always found it to be a bit clunky, and I didn’t like the way mine looked. Around that time a bunch of my friends were all hyping up latex as the way they made their resumes, and that there were a plethora of good templates available online. I knew a little bit of latex at the time, so it seemed like a good idea to migrate everything over to latex. I started off with a pretty generic looking template, but soon got deep into customization. I played around with one and two column layouts, but as the years went by, I eventually settled on a single column, and also created a CV that contained every piece of experience I have (as opposed to my resume which is only my most “relevant” experiences).
A lot has changed since I migrated my resume to latex. For starters, I’m a lot better at latex now! But I’m also a lot better at HTML. You see, I’d always wanted to have my resume be in HTML, for a number of reasons, chief among them being that HTML seemed to offer more options for structuring data, and seemed way easier to tweak than latex - especially since I could trial CSS changes interactively in the browser. Editing my resume was also becoming a pain. I need the machine I was editing on to have a \(\rm\TeX\) installation, and quick tweaks couldn’t be done by just changing the source on github. I can remember at least one instance where for a few months the PDF in the repo was older than the source files, and I didn’t catch it until I’d submitted a few job apps. Additionally, as I started maintaining my CV, my CV would frequently lack updates to entries that were also present in my resume, or vice versa. I wanted to move the common info to a common location, but combining the files back into one document would require scripting, and I didn’t want to deal with latex errors that stemmed from combinations of documents - the error messages aren’t great to begin with. What held me back from migrating was that I didn’t feel comfortable writing the necessary CSS rules to make the HTML match the aesthetic of my latex resume. But, that has changed.
My resume is now written in HTML! The only tool I need to compile it into a PDF is any modern browser. You can view the HTML version here. There’s a few key things that makes this approach work for me:
CSS can specify page dimensions
Until recently, I didn’t know that CSS could specify width/height for documents in terms of inches or meters. This makes typesetting content that is intended for print possible. My resume contains this snippet of CSS:
body {
width: 8.5in;
height: 11in;
}
Which matches the dimensions of the standard US letter size (my old resume technically was formatted for A4, and I like the dimensions of A4 better, but US letter does make more sense since I’m in the US).
JavaScript can fill gaps where HTML is cumbersome to write by hand
As I was migrating my resume to HTML, one thing I found myself missing was
\url
. Which took a URL and generated a link, but also set the text of the link
to be the URL. This makes a lot of sense for links to be printed - I actually
don’t want any other alt text. My solution was to just write an a
tag with a
special class:
<a class="autolink">https://github.com/aneeshdurg/resume</a><
And then include a bit of JavaScript to fix the link later:
const fix_links = () => {
for (let el of document.querySelectorAll(".autolink")) {
el.href = el.innerText;
}
};
I opted to keep the text in the element instead of in the href
attribute as
the initial state so that the links would still render without JavaScript. Of
course, that was before I split the document up, which I talk about below
JavaScript can enable combining documents
Since I wanted to have a common location for snippets shared by my resume and
CV, I settled on a system of having div
s with a special class and a data
attribute and then having some JS replace the div
with the referenced file:
<div class="include" data-src="education.html"></div>
await Promise.all([...document.querySelectorAll(".include")].map(el => {
const src = "snippets/" + el.dataset.src;
return fetch(src).then(resp => {
resp.text().then(content => {
el.innerHTML = content;
});
})
}));
This has one key downside which is that without JS, the document won’t render. I
tried workarounds such as using iframe
s. But the issue was that without JS I
couldn’t apply CSS or get the right dimensions on the iframe
so it didn’t seem
worth it.
The biggest pro of this approach (aside from the shared data) is that editing individual entries in my resume is super easy, and way more ergonomic since I just need to open the appropriate file instead of finding it in a giant document. I choose to just store plain old HTML in each snippet instead of JSON or something so that I could have maximum flexibility.
Conclusion
The new HTML/JS/CSS source for my resume is undoubtedly more verbose than the latex counter part. \(\rm\LaTeX\) is purpose built for typesetting and it shows. However, I now don’t have to duplicate data for my resume and CV. That combined with the ease of editing, and my greater sense of confidence in my HTML/CSS/JS skills vs my latex skills make this change a worthwhile one for me. The source code is available here if you’re interested in browsing it. Thanks for reading!