If you've got a list of people and you'd like to publish them on the web as Linked Open Data, MicroAuthority is a simple, bare-bones system that provides an API, search, and other tools for developers and digital researchers.
MicroAuthority creates both HTML & Linked Open Data endpoints for institutional Agent authority files—lists of people, organizations, and groups. Data is provided by the institution as a CSV file. MicroAuthority automatically creates a unique URI for every row in that CSV file that works as both a webpage and as machine-readable Linked Open Data.
This has only been tested on datasets in the ~20,000 row size as of yet—it should scale to ~100,000 (or perhaps more), but nobody's tried yet. Beta software, etc, etc.
(These instructions are for people famillar with the process of installing Ruby software from Github. If this doesn't make sense, move on to the next section.)
git clone git@github.com:arttracks/microauthority.git
cd microauthority
gem install bundler foreman
bundle install
# Modify the `data/data.csv file before running the next two commands
rake sitemap:create
foreman start
Or, with slightly more context:
- Fork and clone the repository.
- Modify the
data/data.csv
file to reflect your data. Don't worry about missing data for columns—you don't have to fill them all in. The only required fields areid
andname
. Order doesn't matter. - Edit the
config/settings.yaml
file to add institution specific information. - Within the repo directory, install local dependencies:
gem install foreman bundler && bundle install
- Initialize the sitemap:
rake sitemap:create
To run the application locally: foreman start
will start the server running. Typing CONTROL-C
into the terminal will quit the server.
The site will now be running at http://localhost:5000. T
In order to setup MicroAuthority for your data, it's often helpful to run a copy of it on your own computer. This way, you can try it out and see what it's doing before you run it out on the internet somewhere. These instructions are designed to work on a Mac, but they would be similar or identical for any UNIX-based system. If you're interested in running this on a Windows system, get in touch—I would be helpful to work with you to figure out how to install it and to write documentation for Windows.
Before you can run the software, you need to download a copy of the code onto your computer. This code is available on Github, which is a website where people often put code that they would like other people to have access to. (You've obviously found it, since you're reading this here!)
There are two ways to download this—the easy way and the complicated way. The benefit of the easy way is that it's easy. The benefit of the complicated way is that it makes it easy to contribute changes you make to the software back to the project. If you think it's unlikely that you're going to be fixing bugs in MicroAuthority, and you just want to get it up and running, choose the easy way.
The easy way to get a copy is to download the software as a ZIP file directly from Github. This will download a directory containing most of the software that you need to run MicroAuthority. To get the rest, you'll need to install MicroAuthority's dependencies, which are the parts of MicroAuthority that were not written for this project, but are used within the software. An poor analogy would be to think of them as the software that MicroAuthority references. The Dependencies section of these instructions will describe how to do that. You can now skip the rest of this section and go to Navigating to the Directory.
The hard way to clone the Github repository. This uses a tool called git
, which is a tool for version control of software. Version control is similar to version tracking for Word documents—it keeps the history of how the software changes, who changed it, and when. It also allows you to "fork" or "branch" the code, so you can try out changes without losing the ability to go back to how the software was before you started changing things. For a full explanation, see the Getting Started: Git Basics chapter of the Pro Git book, which discussed many of the concepts at length.
If you'd rather just get started using Git, Github provides (a helpful set of instructions for installing git)[https://help.github.com/articles/set-up-git/]. If your computer doesn't already have a copy of it, you'll need to install a version of git
on your computer. It doesn't matter if you use the Desktop version or the Command Line version—both do the same thing.
One you have git installed and configured, you will to clone this repository. This will download a copy of the software that also contains special files (in a .git
directory) that contain a full copy of the history of the code, as well as metadata that describes where the software came from, how to download it again, and how to submit any changes you make to the software back to the project.
If you're using the command line you can do this by typing:
git clone git@github.com:arttracks/microauthority.git
which will create a directory called microauthority
that contains the software.
Now that you have a copy of the code, you should put that directory somewhere you'll remember it. If you downloaded it directly, you should also rename the directory to microauthority
. This isn't required for it to work, but the rest of the instructions assume that it's called that, and you'll be confused if you don't.
We're going to be using the Terminal, also known as "the command line". This is a special piece of software that comes with you computer that lets you run programs and look at files using a text-based interface. You type commands, and it does things. This is the way that all computers used to work, back before things like Windows or MacOS came along—and it's often much more efficient than clicking on icons. Programmers love efficiency, so most of the tools that we use run from the command line. For a basic tutorial on the command line and Terminal, see David Baumgold's tutorial (which is friendlier) or Apple's tutorial (which is more complete).
You'll need to open up Terminal on your computer, and then you're going to change Terminal's "working directory" to microauthority
. This is the same as opening the folder in Finder—it tells the computer that any files it needs and any commands you run should be done in the context of that folder. To do this, we need to use the command cd
, or change directory.
The command:
cd microauthority
will change the working directory to microauthority
. However, this will only work if you're in the directory one level up from it—you'll need to type the whole file path to that directory. If you don't know it, an easy trick is to type cd
(note the space after the d) and then drag the folder from the Finder into the Terminal. Dragging a folder into terminal will type the full path of that directory for you.
You should do this, because the rest of the commands that we'll be typing will assume that you're inside the microauthority
directory.
MicroAuthority is written using the Ruby programming language. MicroAuthority is also somewhat picky about the exact version of ruby that it works on. You can install there are instructions on how best to install Ruby [on the Ruby website], and many people use a program called rvm, or Ruby Version Manager to, you know, manage versions of Ruby. You don't need to use RVM, but if you're having issues getting Ruby installed, it might be worth looking into.
One you have a copy of Ruby installed and configured, you need to install Bundler. Bundler is a tool designed to download and install all the other libraries and related bits of software that you need to get Ruby software working. (It's probably worth mentioning that Ruby software libraries are called Gems. There's a fair amount of whimsy in the Ruby community, which is one of the reasons I write software in it.)
To install bundler, you need to use a tool called gem
that comes with Ruby. (Yes. You have installed a tool that will help you install a tool that helps you install tools. This is noticeably better than it used to be—in fact, Ruby's dependency management system is famously good. I'll allow you a moment to shudder when you consider what a bad dependency mamagement system might look like.)
on the command line, type:
gem install bundler
It should metaphorically whir and click a bit, and once it's done, you have a copy of Bunder available on your system.
you should also type
gem install foreman
which will install Foreman, a tool that makes it easier to launch software from the command line. While foreman
might be overkill for what you're doing right now, when we go to launch our copy of MicroAuthority on the internet we're going to need to use it, so we might as well use it from the start.
Lastly, we're going to type
bundle install
Which will download and install all the rest of the libraries that you'll need to use.
At this point, you have all the software installed on your computer, and you can run MicroAuthority. However, you haven't given it any data...which means it won't be very interesting.
MicroAuthority is designed to be easy to use for people who might be data professionals, but are not database professionals. Because of this, it doesn't have a database of any kind. Instead, when it starts, it reads all the data from a spreadsheet and stores it in the computer's memory. Thankfully, computers these days have loads of memory, so as long as you're dealing with merely tens of thousands of names, it doesn't cause a problem at all. If you're dealing with millions of names, this won't help you, but if you're dealing with millions of names, you're hopefully working on a larger project and know someone you can talk to about how to extend this to work in your situation.
By default, MicroAuthority will look for a spreadsheet in the data/
directory called data.csv
. It expects it to be a csv file, which is a standard form of spreadsheet data where there are commas between each value. Excel can export these, as can almost any other tool.
This spreadsheet is should have a header row, which means that the first row or line of the spreadsheet has the names of the columns. All the following rows should have your data in them.
The spreadsheet is only required to have one column: name
. The smallest possible spreadsheet that would work would look like this:
name |
---|
Jane Doe |
But that's not a very useful list, is it?
The other required column is id
, but if you don't have one, MicroAuthority will make one up for you using the order of the rows. I don't recommend letting it do this, though—it works, but it means that if you ever sort your data, or add a row anywhere except the bottom, it will change the ids assigned everything, and that's a bad thing. So don't do that.
Instead, you should have a column named id
, and the ID should be a unique identifier, which means that every one should be different for each row of the spreadsheet. Preferably, these should be numbers, but words work as well—, though to make the urls work well, we turn them into all-lowercase and replace spaces with underscores, as well as some other munging for more esoteric reasons. If you make a mistake with this MicroAuthority will not start, but it will give you a helpful message about which id
is duplicated.
To help you out, we've provided two example spreadsheets, both in the data/
directory. One is data/example-data.csv
, which is a set of names from the Carnegie Museum of Art's data export, and the other one is a very tiny spreadsheet called data/data.csv
. This second one is there for you to fill with your list of names.
If you don't have a list of names, and just want to see how this works, you can change the csv_file_name
setting in config/settings.yaml
from data.csv
to example-data.csv
, which will let you play with a moderately-sized list immediately. We'll talk more about that settings file later.
(David: Contiune writing from here.)
One you have a copy of it locally, you should modify the data/data.csv
file to reflect your own data. Don't worry about missing data for columns—you don't have to fill them all in.
To test it locally, you need to do the following:
bundle install
rake sitemap:create
foreman start
There is no database used for this project and no additional software needed. I recommend deploying it to something like Heroku—there's not a lot of dependencies involved.
If you update the data, you'll need to kill the website and restart it.
Explanation of how to use this.
The header matters.
This is best done in excel.
Authority File: Database Controlled Vocabulary/Standardized Vocabulary
Subject Heading/Keyword Search
A way to get information speaking the same language.
- They're talking about linking as the core value
- Cornell/Columbia Linked Data
API is a way to get data from a set of pages.
Not sure of APIs and Linked Data
Duplicatse?
Feel free to take this, fork it, and use it as you like. We'd really appreciate a link or a thanks as part of your deployment, but there's no requirement to do so.
Go ye, and make the web more link-tastic!
MicroAuthority is a project of the Art Tracks program at Carnegie Museum of Art.Funding has been provided by the National Endowment for the Humanities.