Skip to content

clclab/clclab-website

Repository files navigation

Netlify Status

Maintaining the CLC Lab website

InstallationNewsPublicationsBlogpostsDevelopmentNotes

The clc site is generated using Jekyll. The site is hosted at Netlify at clclab.netlify.com. Every time changes are made to the repository, the site is automatically regenerated. The address projects.illc.uva.nl/clclab redirects to Netlify.

Installation

Maintaing the website is easiest if you fist install Jekyll locally; see for example this guide on installing Jekyll with Bundler. After cloning the repository:

# Install all dependencies
bundle install

# To serve the website
bundle exec jekyll serve --livereload

Then go to http://127.0.0.1:4000, the website will be served here. Whenever you edit a file, the site is regenerated and your browser will automatically refresh the page. After making the required changes, push your commits to the remote repository. Netfify will automatically start regenerating the website.

See _data/news.yml. Every news item has a date and a text field. This is the format:

- date: yyyy-mm-dd
  text: >-
    Your news message as markdown. 
    So you can use **bold** or _italic_ text,
    [links](https://...), etc etc.

The publications of the clc lab are stored in assets/bibliography/clc_publications.bib. The file is used to generate the publications page.To avoid corrupting the BibTeX file, please use BibDesk to edit the file.

  • CLC-A-Conference={1}: custom field that indicates that this is a publication at an A conference.

Note: the other bibtex files in assets/bibliography contain references that are cited elsewhere on the website, e.g. in blogposts.

Details about (former) lab members are stored in _data/people.yml. Supported fields are:

  • name
  • type: one of current, former, collaborator
  • position: one of PI, postdoc, phd, master
  • url, github, twitter: links
  • email: email adress (please remove for former members)
  • img: picture. Please upload images to assets/people/
  • interests: a brief description of interests. Markdown allowed.
  • startyear and endyear: used for former members only

Blogposts are stored in _posts as markdownfiles named yyyy-mm-dd-title.md. Jekyll only processes files that start with a header: a section deliminated by three dashes. The header contains metadata about the file in YAML format, such as the title. Here's a starter for a blogpost:

---
layout: blog
title: "I know what you see"
subtitle: "Reading minds, part 1"
author: Iris Proff
published: 2018-11-23
image: /assets/posts/2018/Linda-Jolink-reading-minds-1.jpg
---

## Header 2

And some more **boldface** text or *italic* text.
You can cite papers {% cite Zuidema2003 %}. 
Like latex, you only need the cite key in your bibtex file (specified in the references section at the bottom of the file).

## Another section

And some more text

- list item
- list item 2

You can include images and youtube videos as follows:

{%
 include figure
 path="/assets/posts/2000/my-figure.jpg"
 alt="Description of the figure"
 caption="A caption"
 credits="Illustration by Jelle Zuidema"
%}

{%
  include youtube
  id="nsjDnYxJ0bo"
  caption="A caption for the video
%}

(but you can also just include html)

<div class="references">
  <h4>References</h4>
  {% bibliography --cited --file name_of_my_bibtex_file %}
</div>

Please keep the following in mind.

  • Style: Should be accessible or entertaining.
  • Images: If possible, include one or two pictures. Think about whether your results/methods/framework could be nicely visualized as an infographic or interactive plot.
  • Citations: Please use jekyll-scholar for citations: {% cite Zuidema2009 %}
  • You can use formulas.

Development

Directory structure:

  • /_data: YAML files with structured data.
  • /_posts and /_pages: blogposts and pages
  • /_includes and /_layouts: these directories contain the template files used by Jekyll when generating the website.
  • /assets: all publicly available files, such as css, js, images, pdfs, bibliography etc.
  • /src: javascript source code for filtering the publications. The code is saved to assets/js/bundle.js.

Notes