This curated Wiki relies upon the Fred Hutch research community itself to improve, expand and evolve over time. Because the Wiki's content spans many research areas, we need and welcome contributions from a similarly wide range of researchers and Fred Hutch staff. Whether this contribution is as novice reviewers for a topic outside of their expertise or as expert contributors for those topics of most interest to them, no contribution is too little (or too large).
To contribute to the Wiki you only need to have your GitHub username added to the Fred Hutch Github Institution (and the AllFredHutchTeam). Email scicomp
and provide your GitHub username and request to be added. If you are interested in being part of the Wiki Reviewers Team (see below for their role), please email apaguiri
and Amy will add you to the GitHub team for that.
Content Contribution and Review Process
Contributing via an external editor
Building a copy of this Wiki locally
Building a copy of this Wiki on a rhino
We manage the content of this site via a set of markdown files that contain long, article-style text (our main pages in the site) and a handful of focused cookbook/demo-style documents (Resource Libraries).
Articles use an outline structure to allow graduated content, starting from basic information and progressing to more detailed content for expert users/readers. Using headers (H2 through H4) in the text allows the automatically rendered Table of Contents to facilitate readers' ability to jump around the documents to get to needed content as well as the creation of anchors to allow for linking directly to portions of content in a longer page. By keeping related text together in a small number of pages, it allows us to provide more context for people who are learning about a topic, allows finding of other related information a reader might not have known to look for, AND allows content providers to manage content in ONE PLACE rather than spread throughout the site.
Note: This site is created by researchers and staff who are not web designers nor technical writers by training! Thus, we have opted for a relatively flat organizational structure to keep it simpler for content curators and to reduce the risk of information becoming stale and irrelevant as much as possible.
The more focused, how-to style, Resource Library entries in both the Data Generation and Scientific Computing domains can use headers as well to populate a Table of Contents for each of these pages. These documents are intended to be fairly detailed examples or content that is linked to by Articles, but address a specific use case or example scenario that may be only intended for advanced users/readers. Once a number of related Resource library entries are created, Editors may consider consolidating the information and moving it into the main site as a full Article to highlight the content to new readers.
Both Articles and Resource Library entries are full-text searchable using the search feature (the magnifying glass in the header). This search ability is the primary strength behind this Wiki and will be the primary way people will find content, as, again, no web designers or technical writers are involved in this grassroots project.
NEW (as of March 2022) - Introducing Pathways! Pathways are a new approach we are incorporating which can be found by following the Pathways link in the sidebar from any domain. This Pathways page will host individual pages that provide users who want to do a commonly requested set of tasks, a list of pages/links in the order they'll want to read them, that will guide them along the pathway to doing what they want. We hope this might be an alternate mode for finding content in the Wiki that the community finds useful. If you have ideas for new Pathways, please file an issue and tell us about it.
There are multiple ways to edit content on our site, including:
For spot checks, small edits/refinements or those of us not familiar with git or GitHub:
- Clicking on the "Edit this Page" icon on any page, providing your edits, committing them to a branch and doing a pull request (GitHub will guide you in this process).
For larger edits, multi-page edits, structural changes or expert users of git/GitHub:
- Cloning the Wiki repository to your local machine, committing edits to a new branch, pushing those edits to GitHub and doing a pull request (requires knowledge of git and GitHub workflows).
To edit one of the content-containing markdowns (see below regarding Repo structure for more info about where these markdowns are) from GitHub, follow these steps:
-
Create a branch off the main branch for your edits. Do not fork the repo or others cannot submit additional edits to your content. Consider naming the branch in such a way that indicates what domain the edits will primarily be in (such as "generation-typos" or "intro-to-rhino"). Avoid making branches with names that don't attempt to describe the types of changes made whenever possible. For your content to be merged into the main branch, it will likely need to be edited by others, and it is possible that others may have substantial content to add to your edits. If the branches are named according to content being added (generally) then others can contribute to that content too.
-
Commit your edits to existing markdowns as you go, and update from the main branch before continuing to work on your branch. You will reduce future conflicts if you get in the habit of updating from the main branch and committing frequently.
-
Publish/push your branch to GitHub to share your edits with the group.
-
When you are done editing, create a pull request from your branch. This pull request step highlights your branch for consideration by potential contributors and editors! Suggest reviewers based on the content of the edits if you'd like by tagging their GitHub usernames (using @...). Request admin assistance if your content may be new and need to be hooked up to the sidebar or other web-specific needs (this is currently done by tagging
vortexing
orbmcgough
for a review).Note: If you are editing existing content and the page has a listing for the Primary Reviewers like this:
primary_reviewers: somegithubusername
then when you submit the pull request please request a review from those usernames. -
Reviewers will sign off on edits by approving or providing comments on a pull request, ideally one "expert" and one "novice" based on field of expertise. If there is a
primary_reviewers
listed for content then one of the reviews must be from one of those members. Others may move your content to combine it with other work, or make edits that you may want to review as well. Keep an eye on your pull requests and comments on it in order to check back in if someone's edits need your review as well. -
Once approving reviews have been obtained, the pull request can be merged into the main branch and then any edits go live to the site here.
This Wiki is intended to be curated by content owners, local experts, and service providers at Fred Hutch in order to ensure accuracy and relevancy to our community. Anyone can file Issues with suggestions and requests at any time. Direct contributions and reviews can only be made by users who have GitHub usernames affiliated with the Fred Hutch GitHub institution. Individuals from other institutions who are working with the Hutch community are encouraged to file an issue to discuss collaboration with the current project leaders.
Changes to the Wiki are assessed via pull requests to this GitHub repository. We use the primary_reviewers
tag in our markdowns to indicate when there is a resident expert who should be contacted via pull request review requests when content in that markdown is edited. This process of contribution and review from multiple different users allows us to make sure that the content evolves in such a way that it it both more interpretable to the intended audience (Fred Hutch affiliated staff), but also accurate, appropriate and continuously reviewed.
When a pull request is made, it automatically requests a pull request review from any member of the Wiki Reviewer Team. While reviews from other experts is welcomed, a review from a Wiki Reviewer Team member approving the changes is required prior to merging the pull request (this is enforced by protection of the main branch). In addition to regular communication among the Wiki Reviewer Team, we work with the the Hutch's Information Security Office to ensure published content maintains agreed upon standards for information security.
Please remember to make a markdown for yourself in our _contributors
directory so that we can give you credit for your contributions publicly on the site if you would like to.
You can also contribute to the wiki from external editors that can interoperate with GitHub. We have had good experience with Atom but other text editors have GitHub integration as well. Also there is a tutorial on how to use VSCode which is what you will want to use if you plan to contribute many screenshots or other images.
The content of this site is generated using GitHub "flavored" markdown. A cheat sheet for the code required to create things like headings and table is here. Our page TOC's are generated from these headings, so use ## H2 as your first level, and headings H2, H3 and H4 show up automatically in our TOC's.
If you would like to insert a link to another page in our site, please use:
[text you want to have highlighted](/domain/markdownfile_name/)
If it is a link to an external site use:
[text you want to have highlighted](https://my.url.com)
If you'd like to add images to your entry, some text editors (eg. Atom or VSCode via their respective plugins) allow for copy-and-pasting of images. You can read some instructions on how to get set up with VSCode in one of the Computing Demo's.
One edit is that in order for Jekyll to correctly render the images in a page, they should be placed in the assets
subdirectory of the directory containing the page being edited. The following text is the example format that the call to the image needs to be in for a markdown in the _compdemos folder:
![](/compdemos/assets/2018-06-13-16-47-59.png)
If the markdown you are editing is in one of the other folders you'll need to change the compdemos
string to whatever the text of your folder is; please leave out the underscore at the beginning of the folder name.
Both Atom and VSCode will make a directory called assets
in the directory where the markdown is, and then will copy your in-text image file there so you can commit it all to the repo.
If you want to adjust the size of your image, you can add your image using HTML syntax. Be sure to include the alt
attribute.
<img src="/compdemos/assets/2018-06-01-12-34-51.png" alt="VS Code extensions tab" width="500" height="600">
Before linking to external images, consider saving a copy and creating an internal link as outlined here. This helps ensure content doesn't break over time.
When linking to videos like screencasts you typically want to show an image screenshot and clicking on that screenshot starts the video. Images of videos are stored at https://img.youtube.com/vi and they use the same video id you find in Youtube URLs, so The Gift of Time is https://youtu.be/rN7cmb1K2yA
. To embed, insert this into markdown:
[![The Gift of Time](https://img.youtube.com/vi/rN7cmb1K2yA/0.jpg)](https://youtu.be/rN7cmb1K2yA "Click to see The Gift of Time")
It is also important to consider the following parameters for videos from outside sources:
rel=0
- this restricts the related videos shown at the end of payback to videos from the same channel rather than account-based recommendationsiv_load-policy
- set to 1 to display video annotations by default and 3 to disable annotations
So the above link modified would be:
[![The Gift of Time](https://img.youtube.com/vi/rN7cmb1K2yA/0.jpg)](https://youtu.be/rN7cmb1K2yA?rel=0&iv_load_policy=3 "Click to see the amazing kitten")
If you need to make screencasts, free software exist for Windows, Linux, and OSX.
Please if you need to reference a Fred Hutch username, do not write the entire email address out, just put the username in backticks like this:
`username`
The general contributor should likely have no reason/need to edit any of the files in the main directory of the repository, nor files in any other subfolders besides the ones described below. The folders below contain the content portions of the site, while the other folders and files contain all the necessary information to actually BUILD the website itself.
Data Generation Content, organized with filenames that start with xxx_ based on what section they are intended to show up in the sidebar: https://github.com/FredHutch/wiki/tree/main/_generation
Data Generation Resource Library (note all markdowns in this folder will be rendered): https://github.com/FredHutch/wiki/tree/main/_gendemos
Scientific Computing Content, organized with filenames that start with xxx_ based on what section they are intended to show up in the sidebar: https://github.com/FredHutch/wiki/tree/main/_scicomputing
Computing Resource Library (note all markdowns in this folder will be rendered): https://github.com/FredHutch/wiki/tree/main/_compdemos
Contributors List (note all markdowns in this folder will be rendered): https://github.com/FredHutch/wiki/tree/main/_contributors
For new contributor entries: https://github.com/FredHutch/wiki/blob/main/contributorTemplate.md
Pathways page (note all markdowns in this folder will be rendered): https://github.com/FredHutch/wiki/tree/main/_pathways
Everything merged into the main
branch will be automatically deployed to https://sciwiki.fredhutch.org/
Everything in any other branch pushed to GitHub will be deployed to https://sciwiki-preview.fredhutch.org/ ,
which is only accessible inside the Fred Hutch network.
This will always reflect the last (non-main
) commit/push to the repository.
You can check what branch and what commit is reflected by going to
https://sciwiki-preview.fredhutch.org/info.txt.
You may want to build a copy of this wiki locally (on your own computer) to make sure that it looks the way you want before pushing your changes.
- clone the repo somewhere
cd
to the directory where the repo is cloned
At this point there are two different methods you can use to build the site.
If you have Docker installed, run:
./docker-build.sh
- Install Ruby (version 1.9.2 or later). Note: most modern Mac computers already have Ruby installed. If you still need Ruby, it can be found here.
- On Mac, install xcode commandline tools
xcode-select --install
- You may need to install
bundler
. Typewhich bundler
to see if it is already installed. If nothing is returned, then installbundler
withgem install bundler
. If that fails, trysudo gem install bundler
. - You may need to install gems used by the site.
Type
gem install -g Gemfile
to install all of the gems the site uses.- If you get an error
mkmf.rb can't find the headers for ruby at /some/path/to/ruby.h
or something very similar, you need to install theruby-dev
package. - If you get an error that
zlib is missing
you need to install thezlib1g-dev
package.
- If you get an error
Now run
./build.sh
With both methods, you should be able to see the built site (once the build is done) by going to http://localhost:7979 in your browser.
To check for broken links, you can type
rake test
. This will exit with an error if there
are any broken links, and list the broken
links and the files they are found in.
If you are inside the Fred Hutch network, you can type
rake testlocal
and that will include internal URLs
in the check.
First, clone the repo on a rhino and check out the branch or commit you want to look at,.
ml Ruby/3.0.1-GCCcore-11.2.0
bundle install
bundle exec jekyll build
bundle exec jekyll serve -H 0.0.0.0 -P <port>
Select an unused port over 1024.
Point your browser at your current hostname on the port you specified in the serve
command above (example: rhino01:5678
). Once this has started, the site will be rebuilt when updates are made to any of the source files in the site.
The site contains functionality allowing content authors to define words such that a tooltip with the definition (and optional URL) appears when the mouse hovers over the word(s).
You can set this up using the following formatting:
{% glossary Foo %} is one of my favorite words.
Then in the file _data/glossary.yml
you can add a definition like this:
- term: Foo
definition: A nerd term meaning anything really.
url: https://en.wikipedia.org/wiki/Foobar
Note that the url
is optional. Is there is no relevant URL you can leave it out.
This functionality is possible by using the jekyll-glossary_tooltip plugin (see demo).
Note that this plugin is not one of the plugins approved by GitHub to be used with GitHub Pages, so this site is no longer hosted by Pages. Instead we build it ourselves using our CI/CD pipeline.
Generation resource library collection page:
https://github.com/FredHutch/wiki/blob/main/generationdemos.md
Computing resource library collection page:
https://github.com/FredHutch/wiki/blob/main/computingdemos.md
SciComp Announcement resource library collection page:
https://github.com/FredHutch/wiki/blob/main/scicompannounce.md
Contributors list collection page: https://github.com/FredHutch/wiki/blob/main/contributorslist.md
Pathways collection page:
https://github.com/FredHutch/wiki/blob/main/pathways.md
Main website configuration file: https://github.com/FredHutch/wiki/blob/main/_config.yml
Navigation yml: https://github.com/FredHutch/wiki/blob/main/_data.yml
Custom styling that overrides the remote theme: https://github.com/FredHutch/wiki/blob/main/assets/css
Header and footer configs: https://github.com/FredHutch/wiki/blob/main/_includes
Glossary yml: https://github.com/FredHutch/wiki/blob/main/_data/glossary.yml
Our index page: https://github.com/FredHutch/wiki/blob/main/index.md Our outreach page: https://github.com/FredHutch/wiki/blob/main/outreach.md