So you want to write some content for the site? Awesome!
First, reach out to someone to make sure what you are wanting to write isn't already being worked on. You can reach out to Pat ([email protected]) via email. You can also check the wiki topic on discuss to see what's possibly being worked on or in the queue (feel free to reply to that topic if you want feedback). We also have our IRC channel, #pixls.us on irc.freenode.net (webclient) if you want to drop in and ask someone (patdavid, DrSlony, andabata, LotR, houz, Mimir can all point you in the right direction - beware of houz though, he bites).
There are two main ways to write content for the site. The easy way, and the hard way. The easy way is for those folks who simply want to write a blog post or article and don't want to mess around with git or cloning the entire website. The hard way is for those folks who want to help with the website in some way or make 100% sure that their content looks the way they want.
Unless you are going to hack at the website code, I suggest the easy way...
The easy way doesn't require you to do anything fancy.
Articles/posts for the site are written in Markdown. These are just plain text files with some very lightweight markup to denote a few things (see the link for a basic Markdown introduction).
In order to view your Markdown rendered as it would appear on the website, we can use a browser + an appropriate extension*.
* You can also use any stand-alone text editor/previewer that you want. There are many options available for all platforms, as well as some neat on-line versions like dillinger.io and stackedit.io.
To preview everything as it would appear on the website, the extension will need to allow custom css as well.
In Chrome the MarkView extension is known to work. It not only shows the Markdown with custom styling, but it also auto-refreshes the view in the browser every time the parsed Markdown file is saved, which is nice.
-
Install MarkView.
- Allow it to work on local files:
In Chrome, go to:
chrome://extensions/to see all of your extensions.
Find the MarkView extesion in the list and make sure the option to "Allow access to file URLs" is checked:
- Allow it to work on local files:
In Chrome, go to:
-
Download the combined CSS file for pixls blogposts and save it somewhere easy to find:
https://raw.githubusercontent.com/patdavid/markview-css/master/pixls-blogpost.css -
Add the CSS as an option in MarkView.
- From the
chrome://extensionsMarkView entry, click Options.
Alternatively, from the MarkView icon on the Chrome toolbar, Right-Click and choose Options. - Under the section "Custom Styling Instructions":
- Find "Add Theme CSS into Selection" and click the "Choose File" button.
- Choose the
pixls-blogpost.cssfile you downloaded in the previous step and click the "Add Theme" button.
- From the
-
Select a "Document Theme".
- In the section "Selection", under "Document Theme the drop-down should have an entry for
pixls-blogpostnow. Choose that theme:
- In the section "Selection", under "Document Theme the drop-down should have an entry for
After losing over an hour looking around, I don't see any really viable Firefox solution to live-previewing Markdown files while using custom CSS to match the pixls.us website. The best option I could find right now appears to be Markdown Viewer,but it doesn't allow using custom styles or any auto-refreshing of the page on save.
Actually clone this repository and install the necessary packages to build it. Check the README.md file. We will also be writing a CONTRIBUTING.md file soon for further details.
Besides using normal Markdown there are a couple of things to keep in mind for formatting while writing.
Every markdown file requires that there be frontmatter included.
If there will be any question on the characters being used, remember to escape them by enclosing the data in quotes. In particular having things like colons (like CSS in lede-style for instance).
Until I get my crap together more, it's probably safer to copy the frontmatter from the last post of the type you're using (and modify as needed).
Required frontmatter include:
datetitleauthorlayout
The date metadata should be some form of ISO 8601. It could technically be as sparse as a simple YYYY but more commonly:
date: 2016-07-12T08:36:55-05:00
or
date: 2016-07-12
The article title. Duh. :)
If there are any odd characters it would be safer to enclose the title in double quotes...
title: "What're the best ways to help people? :)"
The author name for the post. This will appear at the bottom of the post.
This specifies which type of template gets called for this post. There are two main layout types we're using:
layout: article.hbtfor articleslayout: blog-posts.hbtfor blog posts
Other common frontmatter that is normally used (but shouldn't break the site build if excluded) include:
sub-titlelede-imglede-img-attributionlede-styleauthor-imgauthor-bio
The sub-title for the post. Not needed, but have fun with it if you use it! :)
The lede-img is the header image on posts. Try to keep the image large in dimensions to accommodate bigger screen sizes (try to aim for a minimum of 1650px wide). Feel free to aggressively compress as neeeded. I've been trying to target around no more than 300kb.
The lede-img-thumb is the smaller thumbnail view of the lede-img. It should be sized to around 1024px wide and also aggressively compressed to around 75kb-100kb at most.
This should be html to attribute the source and license of the lede image. For instance, the "Old Oak" lede image on the Sharing Galore uses the following metadata for attribution:
lede-attribution: "<a href='https://discuss.pixls.us/t/old-oak-a-tutorial/1627'>Old Oak</a> by <a href='http://londonlight.org/zp/'>Morgan Hardwood</a> <a class='cc' href='https://creativecommons.org/licenses/by-sa/4.0/'>cba</a>"Just in case the lede image needs some slight modification to work well (centering and shifting) this will be injected into the page style (after the other styles). This should be normal CSS.
lede-style: 'background-position: 0 0;'
This is the headshot or avatar for the author.
Normally this is 150px square.
While this is not a required metadata, if it's not specified it will default to Pat David's ugly mug.
The image should be located at src/images/authors/ but can be in the local directory. The path should reflect its location.
A short blurb about the author. This can include links and other information and should be kept to a few sentences at most.
Aside: this should probably be moved to a yaml file or other template partial so that it can be called for authors by name (including the author-img probably).
For a blog post:
date:
title:
sub-title:
lede-img:
lede-img-thumb:
lede-attribution:
lede-style:
author:
author-img:
author-bio:
collection: blogposts
layout:
Or an article:
date:
title:
sub-title:
lede-img:
lede-attribution:
author:
author-img:
author-bio:
type: "article"
collection: tutorial
layout: article.hbt
When commands need to be referenced there is a special class available called Cmd to provide different styling:
<span class='Cmd'>Colors → Desaturate...</span>Sometimes some text needs to be set aside from the main writing. There is an 'aside' class for this:
<p class='aside'>This is some section of text that needs to be stylistically offset from the surrounding text.</p>Images on posts and articles are normally inserted wrapped with a <figure> tag. This is for both formatting and to make available the use of a <figcaption> tag to caption the image with semantic markup. Markdown is smart enough to allow you to write straight HTML when needed and will pass it through to the output.
To include an image, with no caption:
<figure>
<img src='IMAGE.png' alt='Alt-Text'>
</figure>To include an image with a caption is straightforward:
<figure>
<img src='IMAGE.png' alt='Alt-text'>
<figcaption>
This is a caption for my image!
</figcaption>
</figure>You can also include a plain <img> tag if you want but this is not recommended.
By default, the image will be constrained to the max-width of the <p> elements. This means that at the moment the width of images will be constrained to a max of 640px.
Unless you add a class to the <figure> tag. If you add class='big-vid' to the <figure> tag it will trigger a different style, allowing the image width to go up to 960px:
<figure class='big-vid'>
<img src='IMAGE.png' alt='Alt-Text'>
</figure>The reason for this is that videos share the same styling to make them large as well.
Embedded videos, like Youtube, can be inserted directly if they are wrapped in a <div class='fluid-vid'> tag. This is to ensure that the iframe video is responsive with the rest of the site across various browser widths.
<div class='fluid-vid'>
<iframe width="640" height="360" src="https://www.youtube-nocookie.com/embed/cvA9KmLg7sY?rel=0" frameborder="0" allowfullscreen></iframe>
</div>If you want to embed a big version of the video, you can wrap it in two sets of <div>'s: one for fluid-vid and another for big-vid:
<div class='big-vid'>
<div class='fluid-vid'>
<iframe width="640" height="360" src="https://www.youtube-nocookie.com/embed/cvA9KmLg7sY?rel=0" frameborder="0" allowfullscreen></iframe>
</div>
</div>This also works for other services that use <iframe> to embed things (like Google Maps or OpenStreetMap).
We are a community dedicated to Free Software photography after all - so it makes sense that we might want to release our images with permissive licensing as well! We include the Creative Commons icon fonts on the site (https://creativecommons.org/about/downloads/).
To use them, add the class cc to your element then use the appropriate text to show the desired icon. The below <span> will display the CC logo, By-Attribution, Share-Alike:
<span class='cc'>
cba
</span>Icons Font
Logo and icons in an embeddable TTF.
a: SA, b: BY, c: CC Circle, d: ND, n: NC,
m: Sampling, s: Share, r: Remix, C: CC Full Logo
The ttf font file.